Professional Documents
Culture Documents
Telecommunications Law and Regulation Fifth Edition by Ian Walden PDF
Telecommunications Law and Regulation Fifth Edition by Ian Walden PDF
Telecommunications Law and Regulation Fifth Edition by Ian Walden PDF
Telecommunications
Law and Regulation
Fifth Edition
E d it e d b y
Ian Walden
1
iv
1
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© The Contributors 2018
The moral rights of the authors have been asserted
Fourth Edition published in 2012
Fifth Edition published in 2018
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Crown copyright material is reproduced under Class Licence
Number C01P0000148 with the permission of OPSI
and the Queen’s Printer for Scotland
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2018942280
ISBN 978–0 –19–8 80741–4
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
v
Preface
The origins of this book lie in a LLM course in Telecommunications Law, for which
I became responsible when I joined the Centre for Commercial Law Studies, at
Queen Mary University of London in 1992, and still teach today. One problem
faced by our students in those early days was the lack of a suitable textbook, as
those targeted at the professional adviser market were priced appropriately. The
first edition, published by Blackstone in 2001, was the affordable and accessible
single work that we felt was missing in the market. The publication of the second
edition, in 2005, was by Oxford University Press. For the third edition, in 2009, my
co-editor, John Angel, stepped down from the role due to other commitments. I
would like to record my thanks to John for his invaluable input to this project. The
fourth edition was published in 2012.
This fifth edition substantially updates the fourth edition. In terms of new ma-
terial, all the chapters have changed significantly, reflecting the ongoing legal
and regulatory developments occurring in the sector. While no new chapters
were required, we welcome some new authors and remain immensely grateful to
all the contributors for their input. The organization of the book remains struc-
tured into six parts: Fundamentals; Regulatory Regimes; Key Regulatory Issues;
Telecommunications Transactions; Communications Content; and International
Regulatory Regimes.
The telecommunications sector continues to be of strategic importance to
states, both as an activity in its own right as well as an infrastructure over which
trade is carried out. Rapid technological and market developments confront the
legal frameworks that are designed to regulate the sector, challenging legislators,
regulators, and advisers. This book attempts, but inevitably fails, to keep up with
such developments and the challenges they generate. While the law should be cor-
rect up to January 2018, we have managed to sneak some amendments in up until
June 2018.
Professor Ian Walden
Centre for Commercial Law Studies, Queen Mary, University of London
Of Counsel, Baker McKenzie
June 2018
vi
vi
Contents
Table of Cases ix
Table of Statutes xix
Table of EU Legislation xxxi
Table of International Instruments xli
List of Contributors xlv
PART I: FUNDAMENTALS
viii Contents
Index 895
ix
Table of Cases
A&M Records, Inc v Napster, Inc, 239 F 3d 1004 (2001) (‘Napster’). . . . . . . . . . . . . . . . . . . . . . . . . 737–8
ABC/Général des Eaux/Canal+/WHSmith (Case IV/M.110) [1991] OJ C244 . . . . . . . . . . . . . . . . 568–9
AEG Telefunken v Commission (Case 107/82) [1983] ECR 3151. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
AGC v ISTAT (Case C-240/15), 28 July 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Agincourt Steamship Co Ltd v Eastern Extension, Australasia and China Telegraph
Co Ltd [1907] 2 KB 305, CA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 804
Aimster Copyright Litigation, In Re, 334 F 3d 643 (7th Cir 2003). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 737
Airtours/First Choice (IV/M. 1524) [2000] OJ L 93/01, 13.4.2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Airtours v Commission (T-342/99) [2002] 5 CMLR 7. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
AKZO Chemie BV v Commission of the European Communities (C-62/86) [1991]
ECR I-3359, [1993] 5 CMLR 215 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36, 555, 559
Alliance for Community Media v FCC, 529 F 3d 763 (6th Cir 2008) . . . . . . . . . . . . . . . . . . . . . . . . . . 202
ALS Scan v Remarq Communities Inc, 239 F 3d 619 (4th Cir 2001). . . . . . . . . . . . . . . . . . . . . . . . . . . 743
Ameren Corporation v FCC, 865 F 3d 1009 (8th Cir 2017) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
American Bird Conservancy, Inc and Forest Conservation Council v FCC,
516 F 3d 1027 (DC Cir 2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
American Civil Liberties Union v Reno 21 US 844 (1997). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735
AOL and Time Warner (Case No COMP/M.1845), OJ L 268/28, 9.10.2001. . . . . . . . . . . . . . . . . . . . . 162
Arcor AG & Co KG and Others v Bundesrepublik Deutschland, (C-152/07 and C-154/07)
[2008] ECR I-5959; [2008] 3 CMLR 37. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Arcor AG & Co KG v Bundesrepublik Deutschland (C-55/06) 24 April 2008 . . . . . . . . . . . . . . . . . . . 64
Areva/Siemens (Case COMP/39736), 18 June 2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
AT&T Corp v Iowa Utilities Board, 525 US 366 (1999) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
AT&T v City of Portland, 216 F 3d 871 (2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
AT&T Wireless PCS v Virginia Beach, 155 F 3d 423 (4th Cir 1998). . . . . . . . . . . . . . . . . . . . . . . . . . 254–5
Attorney-General v Edison Telephone Company of London (1880) 6 QBD 244 . . . . . . . 103, 293, 400
Austria v Commission (C-411/02) ECJ, ECRI-8155, 16 March 2004. . . . . . . . . . . . . . . . . . . . . . . 160, 674
Autronic AG v Switzerland (1990) 12 EHRR 485. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 684
A v France (1994) 17 EHRR 462. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 670
BAI and Commission v Bayer (C-2 and 3/01P) [2004] ECRI-23. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
Bais Yaakov of Spring Valley v FCC, No. 14-1234 (DC Cir 31 March 2017). . . . . . . . . . . . . . . . . . . . . 278
Bărbulescu v Romania (5 September 2017) (Application No. 61496/08) . . . . . . . . . . . . . . . . 649, 673–4
Base Co. NV v Ministeraad (Case C-1/14), (2015), Judgment of
Third Chamber, 11 June 2015. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Base NV and others v Ministeraad (Case C-389/08), 6 October 2010. . . . . . . . . . . . . . . . . . . . . . . . . 181
Base NV v Commission (T-295-06), 22 February 2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Bell Atlantic Telephone Companies v FCC, 206 F 3d 1 (2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
Bellsouth/SBC/J V (Case COMP/M.1946), OJ C 202/5, 15 July 2000. . . . . . . . . . . . . . . . . . . . . . . . . . . 144
x
x Table of Cases
Cable & Wireless et al v FCC, No 97–1612, DC Cir, 12 January 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . 825
Centrafarm v Sterling Drug, Case 15-74 [1974] ECR 1147. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
Centrafarm v Winthrop, Case 16-74 [1974] ECR 1183 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
Central Hudson Gas & Elec Corp v Public Service Commission, 447 US 557 (1980) . . . . . . . . . . . 274
Centre Belge d’Etudes de Marché-Télé-Marketing v Compagnie Luxembourgeoise de
Telediffusion SA and Information Publicite Benelux SA (311/8 4) [1986] 2 CMLR 558 . . . . . . . 161
Chambers v DPP [2012] EWHC 2157 (Admin). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 720
Cincinnati v Discovery Network, Inc, 507 US 410 (1993) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274–5
Cityhook Ltd v OFT and Others [2007] CAT 18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 806
Comcast v FCC, 600 F 3d 642 (2010). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212, 764, 770–1, 775
Commission v Belgium (C-11/95) [1996] ECR I-4115. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693
Commission v Belgium (C-221/1), OJ C 274/14, 9 November 2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Commission v Council of the European Union (Case C-687/15), 25 October 2017. . . . . . . . . . 417–18
Commission v France (C-146/0 0) [2001] ECR I-9767 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Commission v Germany (C-424/07) [2009] ECI I-11431. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13, 182, 460
Commission v Germany (International Dairy Agreement) [1996] ECR I-3989. . . . . . . . . . . . . . . . 837
Commission v Greece (C-396/99) [2001] ECR I-7577. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Commission v Luxembourg (C-97/01) [2003] ECR I-5797, [2003] Info TLR 420. . . . . . . . . . . . . . . . 160
Commission v Poland (C-277/07) 13 November 2008. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454
Commission v Portugal (C-429/99) [2001] ECR I-7605 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Commission v Spain (C-500/01) [2004] ECR I-583, [2004] Info TLR 99. . . . . . . . . . . . . . . . . . . . . . . . 160
Commission v Sweden (C-246/07) 20 April 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Commission v UK (C-222/94) [1996] ECR I-4025 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 690, 696
Commission v VW (C-74/04P) [2006] ECR I-6585. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
Computer Inquiry I (Regulatory Pricing Problems Presented by the
Interdependence of Computer and Communication Facilities, Final Decision
and Order, 28 FCC 2d 267 (1971)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Computer Inquiry II (Amendment of 64.702 of the Commission’s Rules and
Regulations, Second Computer Inquiry, Final Decision, 77 FCC 2d 384 (1980)). . . . . . . . . 210–11
xi
Table of Cases xi
JML Direct v Freesat UK [2009] EWHC 616 (Ch), affirmed in [2010] EWCA Civ 34. . . . . . . . . . . . . 707
Johnson v Youden [1950] 1 KB 544 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 680
Lap v Axelrod, 95 A2d 457 (NY App Div 3d Dept 1953), appeal denied, 460 NE 2d 1360. . . . . . . . 302
Liberty Global/Virgin Media (Case COMP/M.6880), 15 April 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Liberty Global/Ziggo (Case COMP/M.7000), OJ [2015] C 145/7. . . . . . . . . . . . . . . . . . . 534–5, 569, 576
L’Oréal SA & Others v eBay International AG & Others (C-324/09), 12 July 2011. . . . . . . 741, 747, 751
L’Oréal v eBay [2009] EWHC 1094 (Ch) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750–1
Louisiana Pub Serv Commission v FCC, 476 US 355 (1986). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
National Association of Telecom, Officers and Advisors v FCC (DC Cir 2017). . . . . . . . . . . . . . . . . 202
National Biscuit Co v City of Philadelphia, 98 A 2d 182, 187–188 (Pa, 1953). . . . . . . . . . . . . . . . . . . 292
National Cable & Telecommunications Association, Inc v Brand X Internet
Services 545 US 967 (2005) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
National Cable & Telecommunications Association, Inc v Gulf Power Co,
534 US 327 (2002). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
New Media Online v Bundeskommunikationssenat (Case C-374/14) [2015],
21 October 2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716
News Corp/BSkyB (Case M.5932), 21 December 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Newscorp/Telepiu (COMP/M.2876). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Nokia/Navteq (M.4942), 2 July 2008, [2009] OJEU C13/8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581
North Carolina Utilities Commission v FCC, 537 F 2d 787, (4th Cir 1976),
cert denied, 429 US 1027 (1976) (NCUC I). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Norwich Pharmacal Co v Customs and Excise Commissioners [1973]
UKHL 6, [1974] AC 133 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750
Nungesser and Eisele v Commission (Case 258/78) [1982] ECR 2015. . . . . . . . . . . . . . . . . . . . . . . . . 547
Peel Land and Property (Ports No 3) Ltd v TS Sheerness Steel Ltd [2013] EWHC 1658 (Ch). . . . . 370
People of the State of California v FCC, 905 F 2d 1217 (9th Cir 1990). . . . . . . . . . . . . . . . . . . . . . . . . 211
Perfect 10, Inc v Amazon, Inc, 508 F 3d 1146, 1172–73 (9th Cir 2007). . . . . . . . . . . . . . . . . . . . . . . . . 737
Perfect 10, Inc v Google Inc, 416 F Supp (2nd 828) CD Cal 2006. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 737
xv
Table of Cases xv
Western Union Telegraph Co v Call Publishing Co, 181 US 92, 98 (1901) . . . . . . . . . . . . . . . . . . . . . 766
Wilson v Newspaper and Mail Deliverer’s Union of NY, 197 A 720, 722 (NI Ch 1938). . . . . . . . . . 288
WorldCom and MCI (Case No. IV/M.1069), OJ L 116/1, 4.5.1999). . . . . . . . . . . . . . . . . . . . . . . . . 162, 440
WorldCom v FCC, 288 F 3d 429 (2002). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Wouters (Case C-309/99) [2002] ECR I-1577. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540
Table of Statutes
xx Table of Statutes
Finland Russia
Information Society Code (917/2014)������������766 Federal Law No. 374-FZ������������������������������������661
Germany Slovenia
Federal Telecommunications Law on Electronic
Law (TKG), § 6 ����������������������������������� 627, 633 Communications 2012 ����������������������������766
Teleservices Act 1997 (TDG)����������������������������739
Solomon Islands
India Telecommunications Act 2009
Regulations (No. 2 of 2016)������������������������������766 (No 20 of 2009)��������������������������������������������859
xxi
xl Table of EU Legislation
List of Contributors
law for sixteen years as an associate with the law firm of Wilson, Elser, Moskowitz,
Edelman & Dicker in New York and in the US financial services industry. Her ex-
perience includes insurance regulatory compliance, appellate litigation and state
government relations for providers of life, health and property/casualty insurance
and pension products. Among her varied functions as Senior Counsel at TIAA-
CREF, the world’s largest private pension system, where she worked for seven
years, Anne served as counsel to the IT divisions.
Dr Karen Lee is a lecturer at the School of Law of the University of New England
(Australia). Her PhD, for which she received the University of New South Wales’s
PhD Research Excellence Award, involved an in-depth study of self-regulatory
rule-making in the Australian telecommunications sector. Prior to becoming
an academic, she worked in the TMT department of the London office of Denton
Wilde Sapte (now Dentons), where she specialized in telecommunications regula-
tion. She was seconded to the Office of Telecommunications in 1998–99 and to the
Department of Trade and Industry in 2002. A graduate of the Indiana University
Maurer School of Law, she is a qualified lawyer in New South Wales, Illinois, and
England and Wales. She is the author of The Legitimacy and Responsiveness of
Industry Rule-making (published by Hart) and has published in the Federal Law
Review, the Media and Arts Law Review and the Australian Journal of Competition
and Consumer Law. In 2016, she was the runner up for the Giandomenico Majone
Prize for Best Conference Paper Written and Presented by Early-Stage Researchers,
awarded by the European Consortium of Political Research’s Standing Group on
Regulatory Governance.
Graeme Maguire is the global head of Bird & Bird’s Tech & Comms Group. He is
based in London but usually somewhere else. His client work focuses on inter-
national tech transactions and communications regulatory issues including com-
munications infrastructure projects, network sharing, spectrum/5G matters and
IoT, OTT and telco cloud business models. He holds an MA in Natural Sciences/
Law from Downing College, Cambridge and a diploma in IP Law and Practice from
Bristol University. Publications include the ‘Communications and Broadcasting
xvli
Regulation’ chapter of the 4th edition of Graham Smith’s Internet Law and
Regulations. Graeme has spent time on secondment in a competition policy role
at Oftel (now Ofcom) and British Telecommunications and prior to joining Bird
& Bird in 2005 was a partner at Linklaters TMT group. Email: graeme.maguire@
twobirds.com.
Dr Bostjan Makarovic has had over sixteen years of experience in European and
international telecommunications law and regulation. He was head of telecom-
munications division with the Slovenian telecoms regulator, acted as a regulatory
advisor to a number of telcos and other ICT businesses, and consultant to regula-
tory authorities in the EMEA region on topics such as regulatory reform, the shift to
NGA, and net neutrality. He acted as member of European Union Communications
Committee (CoCom), member of European Regulators’ Group (ERG) Contact
Network, and advised the Presidency of the Council of the European Union in
relation to the 2009 review of the electronic communications regulatory frame-
work. He holds Queen Mary, University of London PhD in the regulation of Next
Generation Networks. He is the founder of Aphaia, a consultancy dealing with ICT
regulation and policy, a fully qualified lawyer in Slovenia, and an IAPP-certified
international privacy professional (CIPP/E).
Chris Marsden is Professor of Internet Law at the University of Sussex and a re-
nowned international expert on Internet and new media law, having researched
and taught in the field since its foundation over twenty years ago (1996 was arguably
Year Zero). Chris researches regulation by code—whether legal, software or social
code. He is author of seven books and over 130 research publications on Internet
law and regulation, including ‘Net Neutrality: From Policy to Law to Regulation’
(2017), ‘Regulating Code’ (2013 with Ian Brown), ‘Internet Co-regulation’ (2011).
Chris was formerly Professor of Law at Essex (2007–13), having previously re-
searched at RAND (2005–7), Oxford (2004–5), Warwick (1997–2000). He held
Visiting Fellowships at Harvard, Melbourne, Cambridge, Oxford, USC-Annenberg,
Keio, GLOCOM Tokyo, and FGV Rio de Janeiro. He has founded and led teams to
successful completion of over twenty externally funded international collabora-
tive projects, worth in total over £6m, including Openlaws.eu [2014–16] and FP7
European Internet Science (EINS) [2011–15].
David Satola is Lead Counsel, Technology and Innovation, in the World Bank Legal
Department, focusing on legal aspects (transactional and regulatory) of telecom-
munications reforms, internet governance, human rights on the internet, cyber-
security/cyber-crime and competition regulation involving ICTs. His work has
spanned more than eighty-five countries. He was seconded to the UN’s Working
Group on Internet Governance and acts as the Bank’s Observer to UNCITRAL and
ICANN’s GAC. He is the Chair of the Internet Governance Task Force of the ABA.
Lorna Woods is Professor of Internet Law at the University of Essex. Lorna started
her career as a practising solicitor in a TMT practice in the City of London. She has
extensive experience in the field of media policy and communications regulation,
including social media and the internet, and she has published widely in this area,
as well as contributed to a range of studies and parliamentary inquiries.
l
1
Part I
FUNDAMENTAL S
2
3
TELECOMMUNIC ATIONS L AW
AND R EGUL ATION
AN INTRODUCTION
Ian Walden
This book examines national, regional, and international legal and regulatory
frameworks governing the telecommunications sector, particularly the provi-
sion of all forms of network infrastructure, communication services, and equip-
ment supplied for the transmission of data and information. The book is entitled
‘telecommunications’ rather than ‘communications’, despite the best attempts of
European Union law to recast the terminology. Telecommunications remains the
preferred term for a number of reasons. First, the change of regulatory terminology
4
4 Part I Fundamentals
is still not reflected in industry discourse, let alone among the wider general
public. Second, the book is intended as a text for a global audience, not just the
UK or Europe, so it does not seem appropriate for EU terminology to be imposed
on our readers. Third, the many historical and cross-jurisdictional aspects of the
book recommend consistency as an aid to comprehension. The book does, how-
ever, use the terms ‘telecommunications’ and ‘communications’ interchangeably.
1
See Financial Services Authority Press Release, ‘Telecoms lending—fi rms must remain vigilant’, FSA/PN/
153/2000, 7 December 2000.
2
See <https://w ww.ovum.com/ovum-forecasts-1-5 -t rillion-i n-revenues-f rom-g lobal-telecoms-media-
market-i n-2021/>.
3
Ofcom Communications Market Report: UK, 3 August 2017, at 1.3.
5
needs to become familiar. Such terminology relates, in large part, to the tech-
nology being used, the structure of the industry, and the products and services
being supplied. These issues are examined briefly in the next section.
Only a few years ago, the scope of telecommunications technology would have
been easy to define: telephony, fax, and mobile. However, now there is a rapidly
changing technological environment, which means even systems that we use
every day, like the telephone, are now regarded as being ‘legacy’ technology. The
current drivers for change are simple: the ever increasing use of the mobile and
the internet. In many countries, ‘fixed-mobile substitution’ is common; while in
developing countries, mobile is by far the most dominant technology. Voice over
the internet applications and services has meant very cheap telephone calls from
anywhere in the world by connecting over the internet to a service provider in the
destination country, who then routes the telephone call locally. Going even fur-
ther, instant messaging systems (eg Apple’s FaceTime and WhatsApp) provide the
capability of making voice and video calls directly between devices free of charge.
Such ‘over-t he-top’ or OTT services challenge traditional regulatory concepts and
practices, which governments and regulators are still struggling to address.5
However, while these services and capabilities are evolving rapidly, a lot of
the underlying technologies are common. In all systems there are fundamental
categories of equipment that the telecommunications network, of whatever type,
must use. These are:
• transmission systems;
• switching or routing equipment;
• terminal equipment;
• network management systems;
• billing systems.
4
Professor Laurie Cuthbert, Queen Mary, University of London, helped draft this section originally.
5
See in particular Chapters 4, 14, and 15.
6
6 Part I Fundamentals
although some legacy systems using copper cables still exist. Fixed point-to-point
radio links, using microwave, are also used where the terrain is difficult or where
the network node (particularly a base station in mobile networks) is isolated.
At the level of the ‘access network’ there is a wide variety of different transmis-
sion systems. One of the challenges in getting broadband services to residential
customers has been the cost of replacing the old ‘twisted metallic pair’ telephone
cables, and much effort has been spent in improving the technology to allow
higher bitrates over these copper cables, since the cost of replacing them is pro-
hibitive. From ISDN to ADSL (Asynchronous Digital Subscriber Line) and G.fast,
new techniques allow broadband to be offered to residential customers over their
existing telephone lines, at least to those who are close enough to the telephone
exchange. Other fixed access transmission systems use co-a xial cables (cable TV),
optical fibres, or point-to-point radio links.
In mobile networks, the access network is the link between the mobile handset
and the network base station (or BTS—Base Transceiver Station). The type of net-
work (GSM, 3, 4, or 5G) defines the type of transmission used over the radio link.
Another radio access method that is very common is that for WLANs (Wireless
Local Area Networks), often known as ‘WiFi’. The common standard for this is
IEEE 802.11, with numerous iterations since it was first published in 1997 enabling
ever greater capacity to be transmitted over the same radio link. Within organiza-
tions the predominant wireline access technique for computer communications
is Ethernet, often using Unshielded Twisted Pair (UTP) cables, although WLANs
are increasingly being implemented now that the security of such systems is being
improved.
While transmission systems get information from A to B, users want to be able to
connect to different people, or to different websites—t his means that connections
have to be ‘switched’ or ‘routed’ to the right destination. With telephone networks
this was done using switches (called ‘exchanges’ in the UK) but with internet-t ype
networks (IP networks) the devices performing that function are called routers.
The reason for this difference is that telephone networks are traditionally ‘circuit-
switched’, whereas IP networks are ‘packet-switched’. Circuit-switched means
that a connection is set up for the whole duration of a telephone call; in packet
switching, information is broken into units called ‘packets’ that are independently
routed across the network. The important differences between the two techniques
are that:
• the delay in a circuit-switched network is fixed, but in packet networks the nature
of the packet routing means that delays between packets can be very variable;
• it can be harder to guarantee ‘Quality of Service’ (QoS) over packet networks.
Overall this meant that packet-switched networks were generally better suited
for data connections and circuit-switched networks for voice (which is particu-
larly sensitive to delay and variations in delay). However, the predominance of
packet-based IP (Internet Protocol) for computer communications has led to a
major change in how networks are structured, with all communications networks
moving to using IP rather than circuit switching.6 This change has been enabled
as a result of intense research effort to get good quality voice communications
with IP. The difference between routers and switches is in fact much more complex
(and confusing) than the simple explanation above. IP traffic often passes through
equipment called ‘switches’ in the local area network—a nd these have a different
function. To complicate matters even further, new architectures for IP networks
introduce the concept of ‘switched routers’; such as MPLS (multi-protocol label
switching).
In the business world, the local telephone system (PBX— Private Branch
Exchange) has evolved from being a traditional telephony switch to a fully IP
system with IP phones, or even with ‘soft phones’ on the PC. Of course, no network
would work if the user did not have any equipment to use with it and it is often the
capabilities of the terminal equipment that attracts users rather than that of the
underlying network.
An important aspect of any communications network is its reliability and avail-
ability, particularly when congestion occurs. Ensuring this is a function of network
management systems, ie complex software programs that control the operation
and performance of the various network elements; this is true of all types of net-
work, whether telephony, mobile, or internet.
Also of crucial importance is the billing system—no network operator could
survive in business without one! Modern telephony billing systems are very com-
plicated, recording the details of every transmission and applying a wide range
of tariffs based on the type of network user (eg retail or wholesale customer or
interconnecting operator) and type of communication services (eg text messaging
and voice calls). The internet has utilized very different tariff structures from trad-
itional telephone networks, such as flat rate rather than minute-based tariffs,
which has enabled the implementation of much simpler (and cheaper) billing sys-
tems. However, there are now signs that volume-based charging (where the user
pays for the overall amount of data transferred) are starting to appear for internet
6
In June 2015, BT announced that it plans to switch off its PSTN and ISDN networks by 2025.
8
8 Part I Fundamentals
This book is primarily concerned with the rules and regulations governing
the provision of telecommunications equipment, network infrastructure, and
services (eg the transmission of data), rather than the law governing the content
of the traffic being sent across telecommunication networks. The latter is gener-
ally perceived as the domain of ‘media law’7 or ‘internet law’ rather than ‘telecom-
munications law’. However, one recurring issue in telecommunications law is the
problem of distinguishing clearly between issues of carriage and issues of content,
particularly with the emergence of apps offering communications functionalities
and calls for ‘net neutrality’. This edition contains a section addressing various
content-related aspects, in respect of personal data and privacy (Chapter 13), the
impact of broadcasting regulation (Chapter 14), and the position of ISPs regarding
liability for, and control over, the content and services they provide (Chapter 15).
Even the categorization of carriage as a service has evolved, with the development
of commodity markets for trading carriage in terms of telecommunication min-
utes.8 Such economic re-categorization can have profound implications for policy
makers and regulators.
The various aspects of telecommunications law addressed in this book can be
broadly distinguished into competition or economic issues and non-economic
public policy issues. Competition law is primarily concerned with establishing
and ensuring the sustainability of competitive markets, at a national, regional, and
global level. Telecommunications as a sector capable of establishing a compara-
tive advantage in international trade was recognized by the UK Government at the
outset of liberalization, in the early 1980s. In the Telecommunications Act 1984,
for example, four of the ten general duties imposed upon the regulator addressed
trade-related aspects of telecommunications, from encouraging the provision of
transit services, traffic being routed through the UK, to the supply of telecommu-
nications apparatus (s 3(2)). For developing countries, the prospect of becoming a
regional hub in the emerging information economy is promoted as an opportunity
arising from market liberalization.
Non-competition public policy issues have historically focused on the provision
of telecommunication services to the population as a whole: the issue of universal
7
eg Goldberg, Sutter, and Walden, Media Law and Practice (Oxford: OUP, 2nd edn, 2019).
8
eg RouteTrader <http://r tx.routetrader.com>.
9
service. Concerns about the growth of a ‘digital divide’ between the information
rich and poor is a manifestation of such political imperatives. However, other non-
competition issues include consumer protection, environmental concerns, health
and safety matters, as well as the protection of personal privacy and the debate
over ‘network neutrality’ (see further Section 1.7 below).
It is inevitable that the seismic shifts in the structure of the telecommunications
sector are reflected in a complex and rapidly changing legal framework. The lib-
eralization of the sector has usually required significant legal intervention, the
classic exemption to the rule being New Zealand, which initially simply opened
up the sector to competition without the imposition of a regulatory framework,
but has subsequently had to establish a regulatory authority.9 In parallel with the
pursuance of liberalization, the rapid and dramatic technological developments
have compounded the problems faced by policy makers, legislators, and regu-
lators when trying to establish legal clarity and certainty for an industry under-
going convergence with other industries.10 The internet is the classic example of
this technological phenomenon. The existence of a clear legal and regulatory dis-
tinction between issues of carriage, the primary focus of the book, and issues of
content is therefore dissolving in the face of such technological change.
This chapter introduces some of the key themes present within the field of tele-
communications law. These themes are then considered in greater detail in one or
more of the following substantive chapters.
9
All restrictions on the supply of services were removed in 1989. However, by 2001, a Telecommunications
Commissioner was appointed within the Commerce Commission, with substantial further enforcement
powers being granted to the Commissioner in 2006.
10
See Standage, T, The Victorian Internet (London: Phoenix, 1998), which describes the revolutionary im-
pact of the telegraph and Carr, N, The Big Switch (New York: Norton, 2013).
10
10 Part I Fundamentals
Similarly, in the UK, the Communications Act 2003 requires Ofcom to have re-
gard to the possibility of addressing regulatory matters through ‘effective self-
regulation’ (s 6(2)). The technical complexity of the telecommunications market
has always meant that much of the regulatory input on particular issues, such as
interconnection, simply consisted of the convening and oversight of particular in-
dustry groups, intervening only in the event of impasse. However, as regulators
reduce or withdraw from market intervention, then increasingly reliance is likely
to be made upon industry to regulate itself.
A third concept often linked in the past with liberalization and deregulation was
that of privatization: the conversion of the incumbent operator from being a state-
owned public body to a privately owned entity. As with deregulation, the nature of
the relationship with the process of liberalization has been far from straightfor-
ward. The policy drivers behind privatization of the incumbent have tended to be
based around state revenue concerns rather than the objective of liberalization.
The provision of a modern telecommunication infrastructure requires massive
capital investment, a funding-burden which governments have not been prepared
to shoulder. Attracting private sector finance is generally seen as the only feasible
mechanism for meeting the policy objective of modernizing this strategic eco-
nomic sector.
Concerns that a state-owned incumbent might inhibit market entry have come
a clear second to such revenue-raising concerns. Indeed, governments have re-
mained remarkably attached to the ‘national champion’, with the majority of the
OECD countries continuing to have some stake in the incumbent.11 However,
the process of privatization has, itself, sometimes acted as a barrier to the pro-
cess of liberalization. In the UK, for example, the divestiture of BT occurred in
three stages, 1984, 1991, and 1993. However, at the time of the second sale, the
government was also undergoing a comprehensive review of the market, the
‘Duopoly Review’, in order to promote further liberalization (Chapter 3). During
11
OECD, Communications Outlook 2013 (11 July 2013), at Table 2.6.
12
12 Part I Fundamentals
this process, it was generally perceived that BT used the need to maintain share
price for the forthcoming sale as an effective tool in its negotiations with the gov-
ernment. Government stake holdings in incumbent operators have also been an
international trade issue. In the US, for example, concerns were raised in the US
legislature about Deutsche Telekom’s proposed merger with Voicestream, on the
basis that the German government continued to have a stake in its incumbent.
After privatization, a government may continue to be concerned about the per-
formance of the incumbent, particularly where, as in the UK, a significant pro-
portion of the shares are held by the general public, ie the electorate to which the
government is always accountable. In many countries, the need to attract inter-
national investment into the telecommunications sector, either through the sale
of a strategic stake in the incumbent, through Build–Operate–Transfer schemes or
financing new entrants, has actually driven the adoption of a comprehensive legal
framework for the provision of telecommunications networks and services. A lack
of legal certainty is seen as a significant discouragement to financial investment
and therefore to market entry (see Chapter 17).
eg Ofcom Report, ‘Results of research into consumer views on the importance of communications
13
15
eg UK General Conditions of Entitlement, Condition 3 ‘Proper and effective functioning of the network’
(May 2015). See further Chapter 6.
16
See further Chapter 13, at Section 13.3.
17
See OECD, Communications Outlook 2013 (11 July 2013), at Table 2.6.
18
See Case C-424/07, Commission v Federal Republic of Germany [2009] ECR I-11431 and Commission de-
cision (2006/621/EC) on the state-a id implemented by France for France Télécom, OJ L 257/11, 20 September
2006, respectively.
14
14 Part I Fundamentals
19
See Commission Communication, ‘Towards a new framework for Electronic Communications infra-
structure and associated services: The 1999 Communications Review’, COM 1999, 539, 10 November 1999.
20
A sixth, Postcomm, was subsumed into Ofcom on 1 October 2011, reuniting post and telecommunica-
tions at a regulatory level, even though the industries remain distinct, while the incorporation of the BBC
Trust’s regulatory functions into Ofcom represents a seventh.
15
21
See generally Coates, K, ‘Regulating the telecommunications sector: Substituting practical cooperation
for the risks of competition’, in McCrudden (ed), Regulation and Deregulation (Oxford: Clarendon Press, 1998),
at 249–274.
22
<http://c a.globalstar.com/en/>.
16
16 Part I Fundamentals
such financial assistance, which may require the recipient jurisdiction to adopt a
pro-competitive legislative and regulatory framework for the telecommunications
market (Chapter 17). Indeed, such conditional financial assistance to developing
countries has been an extremely influential tool in the international harmoniza-
tion of telecommunications law.
governments $30 billion and $50 billion respectively. However, as with much eco-
nomic theory, rational actors often act irrationally, paying sums through fear of
market reaction as much as the business rationale. As a result, serious questions
have been raised about whether the benefits in terms of public revenue will be
achieved at the expense of the development of the market itself: through delayed
roll-out and higher charges for services.
Another important scarce resource is telephone numbers. Access to a number
and the right to control access to numbers needs to be subject to regulatory control
in order to facilitate market liberalization. However, strategic national planning
for the use and distribution of telephone numbers into the future can be an ex-
tremely difficult task and one which, if mistakes are made, can generate substan-
tial adverse public feeling towards the national regulatory authority. The domain
name and IP addressing scheme utilized for internet-based communications has
also generated regulatory issues, relating to its governance, scarcity, and impact
on other legal regimes, such as trade marks.
The right to access or utilize the private property of another for the provisioning
of networks and services is an issue that has sometimes been viewed as similar in
nature to the use of a scarce resource. Whilst the granting of rights of way need
not be limited, the exercise of a statutory right to interfere with another’s property
has such potentially significant consequences for the owner and/or occupier of the
property that regulatory controls are inevitably necessary. Not least, the exercise
of such rights interferes with an individual’s right to enjoy their possessions and
their right of privacy, as enshrined in national and international law.23
As telecommunications networks proliferate in a competitive market, it is pos-
sible that people challenging the exercise of statutory rights may increasingly raise
such human rights concerns against operators building networks across private
land. Recognized limitations to an individual’s right of privacy on grounds such
as the ‘economic well-being of the country’ or the ‘rights and freedoms of others’24
may be sustainable as a basis upon which to interfere during the process of liberal-
ization, but may seem less ‘necessary’ once a market is fully competitive.
The construction of international telecommunications networks raises issues of
access to public resources, both state-based, such as the electromagnetic spec-
trum, as well as resources recognized under public international law as the ‘prop-
erty of all mankind’, specifically outer space and the high seas (Chapter 16). Public
policy makers and regulators are also giving greater consideration to environ-
mental concerns, such as the siting of transmitters for wireless communications
23
eg the European Convention for the Protection of Human Rights and Fundamental Freedoms, Art 8(1),
and Protocol 1, Art 1.
24
Ibid, at Art 8(2).
18
18 Part I Fundamentals
Disputes and complaints may arise between market participants, between the
regulator and the regulatees, and between providers and their customers. In
the latter case, especially where consumers are involved, sectoral dispute settle-
ment schemes are designed both to redress an inevitable imbalance between the
parties, as well as facilitating access to justice for the consumer. Intervening in
disputes between market participants has been a critical component of the lib-
eralization process, primarily because of the position of the incumbent. Where
markets are fully competitive, however, such regulatory intervention may be seen
as an unnecessary use of public money when the parties have equal recourse to
alternative legal processes.26
The manner in which a regulator exercises its powers is an issue of concern to
telecommunication lawyers. As with any public authority, the regulator will be
continuously required to exercise its discretion in respect of when, where, and
how it intervenes in the operation of the market. The complex nature of regula-
tory invention in the sector, particularly in respect of cost-related matters such
as price controls, may require that regulatees have the right to appeal against
regulatory decisions through a de novo appeal procedure (Chapter 4). Regulatory
decisions will also be subject to judicial review on procedural grounds, chal-
lenging a decision on the basis of irregularity, irrationality, illegality, and
proportionality.
The frequency and manner in which decisions are challenged will also impact
on the operation of the whole regulatory framework. Legal activism by operators,
frequently challenging the decisions made by the regulator, may effectively slow
down the decision-making process, as regulators become cautious and exces-
sively procedural in order to stem legal challenges and the associated commit-
ment of public resources. Legal interventions in regulatory decision-making are
more often of benefit to the incumbent, than new entrants.
T-Mobile, BT, H3G, C&W, Vodafone & Orange v Ofcom [2008] CAT 12, at paras 89 and 94.
25
20 Part I Fundamentals
incumbent operator and other operators with similar market influence, such as in
the mobile sector. Within Europe, such asymmetric regulatory controls are placed
on organizations designated as having ‘significant market power’ (Chapter 4);
while at an international level, the equivalent term in the WTO’s Reference Paper
is ‘major supplier’ (Chapter 16).
Much of the literature in the field adopts a fundamental distinction between so-
called ex ante and ex post regulatory controls. For the purpose of this book, the
phrase ex ante is used in respect of regulatory measures that proactively control
the structure and/or behaviour of market players going forward; while ex post re-
fers to measures that arise in reaction to the decisions and activities of entities.
Establishing the costs associated with the provision of telecommunications net-
works and services is key to their effective regulation. Interconnection charges can
represent from a third to a half of a new entrant’s costs; therefore regulatory control
over such charges through ‘cost-orientation’ requirements is critical to enabling
competition (Chapter 8). Likewise, universal service policy requires the identi-
fication of those service elements that are ‘provided at a loss or provided under
cost conditions falling outside normal commercial standards’,27 before regulators
provide appropriate financial support mechanisms. However, determining and
verifying such cost-based obligations is often an extremely controversial regu-
latory process, in terms of attribution, calculation methodology, eg whether his-
torical or forward-looking, and the establishment of appropriate cost accounting
systems by regulated operators (Chapter 2).
Tariff controls are present under most regimes, whether at a retail or wholesale
level. Such controls are generally perceived as being the most appropriate mech-
anism for ensuring that a dominant operator is controlled whilst providing suf-
ficient incentives to encourage economic efficiency. Such controls are, however,
notoriously difficult to get right in terms of balancing the interests of customers,
competitors, and the dominant operator.
Related to tariff controls are requirements upon operators to disclose infor-
mation about various aspects of their business activities, either to the regulator,
competitors, or consumers: eg tariff filings and technical standards for intercon-
nection. Information asymmetry is an inevitable regulatory problem in a complex
sector such as telecommunications. Transparency obligations are designed to re-
move the likelihood of anti-competitive practices and to provide a certain degree
of legal certainty, for example, through obligations to publish standard contrac-
tual terms and conditions (eg a Reference Interconnection Offer). The publication
27
Directive 2002/22/EC on universal service and users’ rights relating to electronic communications net-
works and services, OJ L 108/7, 24 April 2002, at Art 12 and Annex IV, Part A.
21
FCC Report, ‘A New Federal Communications Commission for the 21st Century’, 1999.
28
See Emtel Ltd v The Information Technology and Communication Technologies Authority & ors (2017) SCJ
29
294, at para 253.
30
See case Telecom Corporation of NZ Ltd v Clear Communications Ltd (1992) 4 NZBLC.
31
The words of Professor Areeda, quoted with approval in Verizon Communications Inc v Law Offices of
Curtis V. Trinko (02-6 82) 540 US 398 (2004), at 15.
2
22 Part I Fundamentals
In our information society, more and more technical standards are used in formu-
lating laws, regulations, decisions etc . . . standards are becoming more important
in drafting contractual obligations and interpreting the meaning thereof, whether
or not in the courtroom.34
32
Communication from the Commission, ‘The results of the public consultation on the 1999
Communications Review and Orientations for the new Regulatory Framework’, COM(2000)239, Brussels, 26
April 2000.
33
See generally Shapiro, C, and Varian, H, Information Rules: A Strategic Guide to the Network Economy
(Harvard: Harvard Business School Press, 1999).
34
Stuurman, C, ‘Legal aspects of standardization and certification of information technology and telecom-
munications: an overview’, in Amongst Friends in Computers and Law (Netherlands: Computer/L aw Series,
No 8, 1990), at 75–92.
23
The nature of the communications process requires that the various parties ad-
here to a certain agreed standard, whether in terms of language, protocol, num-
bers, or physical connection. The need for standardization to communicate
across national boundaries gave rise to the establishment of the International
Telecommunication Union, one of the oldest inter-governmental organizations
(Chapter 16). As the quote highlights, there is proliferation of standards within
the laws, regulations, and agreements governing the telecommunications
market.
Standards are critical to the process of liberalizing a market. New entrants
will be as dependent on the technical certainty that arises from the exist-
ence of published standards, as they require legal certainty upon which to
base their investments. The absence of appropriate standards has been used
by incumbents to delay the introduction of competing services. Within the
European Union, standards have been critical in the establishment of an
Internal Market for telecommunications equipment, networks, and services
(Chapter 4).
Numerous standards-making bodies operate in every aspect of the telecom-
munications market, as well as at a national, regional, and international level.
Historically, such bodies have tended to operate in accordance with complex
bureaucratic procedural mechanisms, which led to inevitable delays in decision
making. With the appearance of new technologies and environments, such as the
internet, such institutions have increasingly faced competition from new entities,
such as the Internet Engineering Task Force (IETF), operating under more flex-
ible and rapid processes. Participation in the work of such bodies can require op-
erators to devote significant financial and management resource, while failure
to participate may effectively hand control over the development of a particular
market to your competitors.
One important aspect of standards in the technology field is the possibility
that a particular standard may constitute the intellectual property of a com-
pany, such as a patented process. In 1999, a dispute arose between Ericsson
and Qualcomm over the ownership of certain patents related to Code Division
Multiple Access (CDMA) technology, which underpins third generation mobile
telephony. In such circumstances, competition law principles may be applic-
able, particularly the ‘essential facilities’ doctrine. 35 However, regulators may
be concerned to ensure that ex ante measures are in place to prohibit such prac-
tices (Chapter 10).
24 Part I Fundamentals
36
eg Commission Recommendation ‘On Leased lines interconnection pricing in a liberalized telecommu-
nications market’, C(1999)3863, 24 November 1999.
37
See Agreement between the European Communities and the Government of the United States of America
on the application of positive comity principles in the enforcement of their competition laws, OJ L 173/2 8, 18
June 1998.
25
during BT’s abortive attempt to merge with MCI in 1997, the Director General of
Telecommunications in the UK expressed concerns that one of the potential con-
sequences were the merger to be successful was that BT may end up with a sub-
stantial proportion of its assets residing overseas, as well as its investments, at
the expense of the domestic market.38 To address this concern, BT’s licence was
modified to include an annual reporting requirement whereby BT would effect-
ively guarantee that sufficient resources were maintained to meet its domestic
obligations.
26 Part I Fundamentals
details. However, the process of liberalization in Europe and the US, as well as in
many other countries, is sufficiently well advanced to provide us with a clear out-
line of some of the key aspects of international best practice in law and regula-
tion for the telecommunications or communications sector over the next five to
ten years.
27
THE ECONOMIC S OF
TELECOMMUNIC ATIONS R EGUL ATION
Lisa Correa1
2.1 Introduction 27
2.2 Rationale for Regulation 28
2.3 The Principles of Economic Regulation 29
2.4 The Economics of Telecoms 30
2.5 Scope of the Regulatory Control 34
2.6 Forms of Economic Regulatory Control 40
2.7 Important Considerations when Setting Regulatory Controls 41
2.8 Structure of the Regulated Industry 46
2.9 Privatization and Liberalization of Communications in the UK 48
2.10 Retail Price Regulation in the UK 54
2.11 Tariff Rebalancing Issues 58
2.12 Social Obligations and Further Constraints on Retail Prices 59
2.13 The Need for an Effective Interconnection Regime 63
2.14 Interconnection Cost Methodologies 64
2.15 Interconnection and Access Regulation in the UK 68
2.16 Concluding Remarks 97
2 .1 INTRODUC TION
1
The author would like to thank Federica Maiorano for comments. Any views expressed in this paper are
those of the author and do not necessarily reflect the views of the various institutions for which she works.
28
28 Part I Fundamentals
• 100 per cent share of the market at the time of privatization and thus 100 per cent
control of customers;
• The accumulated assets, and economies of scale2 and experience of the telecoms
market; and
• Ownership of vital networks or privileged use of public rights of way to which
potential competitors must perforce have access if they were to compete.
2
Economies of scale are the cost advantages that firms obtain due to size, output, or scale of operation, with
cost per unit of output generally decreasing with increasing scale as costs are spread out over more units of
output—see discussion in Section 2.4.4.
29
3
Super-normal profits relate to the concept of monopoly profits. In a competitive situation, it is assumed
that any excess profit will be competed away by competition. However, in a monopoly environment, this is not
the case and hence super-normal or excess profit is earned.
4
The rationale for imposing regulatory measures is generally based on the notion of market failure. This
situation exists when a market fails to function properly. Market failure can arise under various circumstances.
In such cases the introduction of appropriate regulatory measures can provide a way of eliminating, or at least
reducing, the market failures thereby providing protection to citizens and consumers, and businesses.
30
30 Part I Fundamentals
5
Marginal cost is the cost associated with the provision of an additional unit of output while incremental
cost is the cost associated with an additional specified increment of output.
31
would make a loss. In setting charges for services, communications networks need
to make some allowance for common cost recovery. In general, in markets with
significant common costs, an efficient pricing structure will be one where prices
lie between the incremental and standalone costs of those services. Standalone
costs refer to the costs of producing a product or service in isolation. They therefore
include the incremental cost for that service plus all the common costs that would
be incurred regardless of whether only one service is supplied. This means that the
standalone costs of a service would be significantly higher than the incremental
cost of that same service. This large gap between incremental and standalone
costs means however that there are various methods of recovering common costs
(see Section 2.14.3).
From an economic point of view, an efficient structure for the recovery of
common costs should reflect the demand conditions for different products. More
price inelastic products (where the demand for the product does not increase or
decrease correspondingly with a fall or rise in its price) should attract a higher
proportional mark-up to incremental costs—t his is often referred to as ‘Ramsey
pricing’.6
However, the overall sizes of different product markets will also influence how
a firm recovers its common costs across its product set. A large product market
may make a significant contribution to overall common cost recovery by virtue
of high volumes, even if the per unit mark-up for common costs for that product
is not particularly high. Given the preponderance of common costs, there is much
opportunity for various kinds of price discrimination.7 The degree to which this is
possible is however dependent on the demand conditions in the market.
6
Named after Frank P Ramsey, who set out this issue in 1927.
7
Price discrimination often features in economic theory as a manifestation of monopoly power. However,
in communications, the presence of significant economies of scale and scope means that price discrimination
between different customer groups (depending on its structure) may provide a means of not only allowing the
firm to recover its costs, but it could also lead to an increase in output and to customers who might have other-
wise been priced out of the market, being served.
8
An externality is the cost or benefit which is not internalized by the person that consumes the service.
32
32 Part I Fundamentals
critical to scale the network but it is likely to make these decisions based only on
the direct cost of and profit opportunity from production. The firm is unlikely to
consider the indirect costs to those harmed by an incorrect decision. Specifically,
if demand exceeds capacity and the network is congested, it would be the con-
sumer (not the firm) who suffers because his/her calls or data packets are not de-
livered.9 In this situation, this may mean that some high-valued demands may not
be served. This negative externality effect together with the systematic variation of
demand over time means peak-load pricing10 schemes are common.
9
In this example, the firm would not suffer and in fact could benefit from this incorrect decision because
if demand needs to be rationed, a mechanism to do this would be to raise price.
10
Firms that deal in markets with fluctuating demands such as peak and off-peak periods will incur some
costs that are common to both periods and other costs which are separable to whichever is served. In the case
of communications, due to the level of common and fixed costs in the network, costs are low in off-peak periods
hence resulting in low prices, whilst in peak periods prices are high because of lower available capacity.
11
If there is only one customer on a telecoms network, there is not much point as one would not want to call
oneself. If however other individuals join the network, the calling opportunities increase and so from a social
viewpoint, everyone benefits.
12
See the ITU World Telecommunications Development Report 1994 and Saunders RJ, Warford JJ, and
Wellenius B, Telecommunications and Economic Development (Baltimore MD: John Hopkins University
Press, 1983).
13
The calling party pays principle is a billing option whereby the person making the call is charged for its
full costs. The total cost of each call placed by a subscriber is split in two parts. The first part is the amount
that the caller’s provider is charging to provide the service to the calling party. However, another part of the
charge is the amount that the provider of the call-receiver will charge the caller’s network, to terminate the
call onto his network.
14
If there are two networks, one large with many customers and another which is just starting up, then no-
one would want to join the smaller network, unless the two networks were interconnected.
3
customers are attracted to an operators’ network based on the cost of call origin-
ation, not the cost of termination.15 There is no incentive for the called party to be
connected to more than one network operator and in the unlikely event that this
did occur, it could entail wasteful duplication. In essence, the called party network
has control over a bottleneck monopoly, in which it can gain monopoly profits by
inflating the call termination fees. This means there is a strong case for regulation
of this service and its charges not only today, but for some considerable time to
come on those operators with significant market power.16
15
In Europe, Australia, and New Zealand, the calling party pays (CPP) principle has generally applied. In
contrast, in the US and Canada, the receiving party pays (RPP) principle applies. The choice of RPP or CPP
generally reflects historical choice. Other things being equal, a charge for receiving calls under RPP may
discourage receipt of calls and in some circumstances, may discourage subscribers from connecting to the
network. On the other hand, CPP may discourage the making of calls but may via discounted subscriptions
encourage greater subscriber penetration. In the early days of communications in UK and Europe, the view
was that encouraging subscriber penetration was important, and so CPP was adopted. CPP means however
that call termination is a bottleneck, and so where CPP is in place, every network has market power over call
termination.
16
See further Chapter 4, at Section 4.6. It is possible to have SMP in both origination and termination but it
is less likely to be the case that everyone would have SMP in origination. It is more likely that there would be
a tight oligopoly. If there were sufficient countervailing buyer power, operators would likely switch to a ‘bill &
keep’ regime, depending on traffic flows and the billing costs. In the fixed market, all operators have SMP on
termination but because the regime is based on pure LRIC, which is very low, they have adopted a bill & keep
regime because that is less costly than actually billing other operators.
34
34 Part I Fundamentals
the calling party pays principle, while origination is mainly a bottleneck because
of sunk costs17 and economies of scale.
17
A sunk cost is a cost that has already been incurred and cannot be recovered if the firm exits the market,
eg an optical fibre once laid will not be removed and reused.
18
Often markets are defined based on competition law principles, specifically upon the principle of
the ‘hypothetical monopoly’. This concept is already well established in antitrust legislation, both in the
European Union and in the US, and provides the standard framework for market definition analysis in com-
petition policy cases.
19
eg the supply of paper for use in publishing. See Case IV/M166 OJ (1992) C58/20, Torras/Sarrio [1992] 4
CMLR 341. For customers, different grades of paper are not viewed as substitutes, but because they are pro-
duced using the same plant and raw materials, it is relatively easy for manufacturers to switch production
between different grades. If a ‘hypothetical monopolist’ in one grade of paper tried to set prices above com-
petitive levels, manufacturers producing other grades could easily start supplying that grade; market power is
thus constrained by substitution by suppliers.
35
20
eg if we were to consider the market for bottled waters and found that via the SSNIP test, the monopolist
could charge a small but significant non-t ransitory price premium above the competitive level then the rele-
vant market, in this case, would be bottled waters. However, if the SSNIP test showed that the monopolist was
prevented from charging a small but significant non-t ransitory price premium above the competitive level
then we would need to repeat the test with the inclusion of the closest substitute such as all non-a lcoholic
beverages etc.
21
In the situation where there is market power and current prices are assumed to be at the ‘competitive’
price level, then if prices are increased, sufficient switching would likely occur for it to become unprofitable.
This is what we would however expect in a sector with market power because if the price rise was profitable,
the firm would have implemented it already.
22
See United States v E.I du Pont de Nemours & Co. 351 US 377 (1956); 76 S. Ct. 994; L. Ed. 1264. In that case,
Du Pont argued that cellophane did not constitute a separate market since it competed directly and closely
with other flexible packaging materials such aluminium foil, wax paper, and polyethylene. The problem
with this argument was that Du Pont, as the sole supplier of cellophane, is likely to have set the prices for its
products at a level where alternative products only provided a constraint on the pricing of cellophane if the
prevailing price was used as the ‘competitive’ price.
36
36 Part I Fundamentals
Market power is not, however, an absolute term but a matter of degree; the de-
gree of market power will depend on the circumstances of each case. In assessing
whether there is dominance, a case-by-case review is needed of:
• the structure of the market and the nature of competition prevailing in the
market;
• barriers to entry into the market; and
• countervailing buyer power.
23
Case 27/76, United Brands v Commission [1978] ECR 207, [1978] 1 CMLR 429.
24
eg if a firm has 35 per cent of the market, it might still be dominant if it has sixty-five competitors each
with 1 per cent. Where two firms have roughly equal market shares, even if they are high then single firm dom-
inance is unlikely to be found but collective dominance, whereby a group of firms jointly occupies a dominant
position, may be found under EU law.
25
Case C62/86, AKZO Chemie BV v Commission [1993] 5 CMLR 215.
37
also be considered consistent with dominance if other factors (ie weak position of
competitors) are also indicative of it. In the case of market shares below 40 per cent,
the EC considers it unlikely that an undertaking will be individually dominant if
its market share is below this threshold, although dominance could be established
below that figure if other relevant factors (ie the degree of vertical integration or
the firm’s control of essential inputs that are required by its competitors—see dis-
cussion below) provided strong evidence of dominance.
Market shares can be assessed by volume or value of sales. The appropriate
measure can vary between markets. Often volume measures are used for bulk
products such as wholesale conveyance minutes, and value measures for differ-
entiated products such as retail products. Whichever measure is used, for the as-
sessment of dominance, it is important that the history of the market shares of all
undertakings within the relevant market is considered. This is more informative
than considering market shares at a single point in time, partly because such a
snapshot might hide the dynamic nature of a market. For example, where markets
are growing fast, high market shares are less indicative of market power than in a
more mature or slow-g rowth market. It is important in these types of markets to
thus have a proper picture of the structure of the market to ensure that any desig-
nation of dominance does not prevent innovative activity from occurring.26
26
In many fast-g rowing industries, it is often the case that a particular firm takes the innovating initiative
which involves considerable up-f ront investment and so becomes the market leader. Other firms then enter
the market and adopt the practices of the first-mover firm. Now because of customer inertia (ie customers take
time to switch their allegiances), initially, the first-mover firm is likely to have a higher market share. This may
not however reflect a dominant position because if it is relatively easy to enter the market then over a short
period, market shares could be eroded quickly. If the undertaking with a high market share was designated as
dominant, this could be detrimental for competition and the development of the industry in the long run as
there would be negative incentives on firms to innovate and become the first-mover in an industry.
27
An exit barrier is a similar concept: a cost borne by a firm leaving a market that a firm remaining in the
market does not have to bear. The existence of exit barriers can be important when considering sunk costs
since exit costs reduce the disposal value of an incumbent’s assets if it chooses to leave a market and may
therefore equally deter new entry.
38
38 Part I Fundamentals
are thus important in assessing whether a firm possesses market power. To get a
better picture of the state of the market, it is important that the regulator has hard
evidence of the recent history of the market. Such evidence might include a his-
torical record of entry into and exit from the market (or closely related markets) or,
if possible, fully documented evidence of plans to enter. Growth, or prospective
growth, in a market could also have a bearing on the likelihood of entry, as entry
will usually be more likely in a growing market than in a static or declining one.
The reason is because it will be easier in a growing market for an entrant to be ac-
commodated without any significant declines in prices and profits. The rate of in-
novation may also be important. In markets where high rates of innovation occur,
or are expected, innovation may overcome barriers to entry relatively quickly.
Indeed, any profits that result from an entry barrier created by successful innov-
ation may be an important incentive to innovate.
All of these issues need to be examined as part of a market review of competition
and dominance. The existence of fully informed customers should also be con-
sidered. If customers are fully aware of the options open to them then a dominant
firm is likely to find it difficult to leverage its market power.
In the UK, three types of entry barrier (see Table 2.2) are often distinguished,
although they may overlap at times:
40 Part I Fundamentals
It should be noted though that while the conditions mentioned above are im-
portant to analyse in a market assessment, buyer power does not always benefit
the final consumer. For this reason, a careful analysis of vertical relationships in
the market, on a case-by-case basis, is often also required.
The SSNIP test provides the standard framework for market definition analysis. Once
the market has been defined, the dominance assessment can then begin.29 As dis-
cussed, market shares are not conclusive on their own. The complex nature of competi-
tion in the market, potential barriers to entry, and buyer power must also be examined.
Once the scope of the regulatory control has been specified, the next step is to de-
termine the appropriate form of economic regulatory control to be imposed where
there has been a finding of SMP.
28
Office of Fair Trading, December 2004, ‘Assessment of market power’, at <https://w ww.gov.uk/govern-
ment/uploads/s ystem/uploads/attachment_data/fi le/2 84400/oft415.pdf>.
29
See Sections 2.9 and 2.15 for a discussion as to how these concepts have been applied in the UK and
see Chapter 9.
41
30
The WACC is the minimum return that a company must earn on an existing asset base to satisfy its cred-
itors, owners, and other providers of capital, or they will invest elsewhere.
31
In the UK, the measure of retail price inflation used was the Retail Price Index (RPI), which measures the
change in the cost of a representative sample of retail goods and services. Since 2013 it has been superseded
by the Consumer Price Index (CPI).
32
The extent to which the RPI –X approach provides incentives to improve productive efficiency is in part
a function of how the efficiency gains in one control period are treated in the next period. If the efficiency
gains in one charge control period are shared between the firm and consumers in the following charge control
period, then the firm has an incentive to find those gains.
42
42 Part I Fundamentals
• how the services to be regulated are to be grouped (ie the tariff baskets) and
whether further mechanisms are then necessary when controlling the different
groups (ie sub-caps etc);
• the duration of the price cap;
• how the firm’s movements of costs and productivity change over the life of the
price cap;
• how capital costs should be valued;
• the interaction of the price cap with quality; and
• demand-side factors that may hinder consumer choice.
circumstances, the NRA should consider whether there is a need for additional safe-
guard controls to provide a regulatory safety net for certain customer groups.
44 Part I Fundamentals
Historic Costs relate to what the assets cost in the first place, minus depreciation.
33
Replacement or Current Costs—w hat it costs to replace old assets with modern equivalent assets of equal
34
remaining life. This reflects general inflation, plus specific effects such as technical progress.
45
35
It is imperative that if informational remedies are implemented that they are tested to be accessible, rele-
vant, and targeted. Just providing more information may not be sufficient because behavioural factors (biases)
may mean customers do not engage with it. See OFT, ‘What does Behavioural Economics mean for Competition
Policy?’, March 2010, OFT1224, at <http://webarchive.nationalarchives.gov.uk/20140402142426/http:/w ww.
oft.gov.uk/shared_oft/economic_ research/oft1224.pdf>.
46
46 Part I Fundamentals
Economic regulation has taken various forms in different industries and coun-
tries. This is partly because each industry/country has an individual NRA, al-
lowing personal differences to manifest themselves. More fundamental, however,
are the underlying differences in the structure of the regulated industry. For ef-
fective regulation to occur, an effective model or industry structure needs to be
developed to allow effective competition to emerge.
The main model of competition in communications is one based on access to a
vertically integrated incumbent’s bottleneck facilities. This is because of the signifi-
cant scale and scope economies in telecoms networks (see Section 2.4). The incum-
bent also uses those same facilities to compete downstream against competitors to
whom it supplies access to the bottleneck facilities. A key issue is that control of the
bottleneck facilities (see Section 2.5) potentially puts the incumbent in a position of
advantage with respect to its competitors in the downstream market. This may allow
it to have both the incentive and ability to distort competition in the downstream
market through the way it exercises its control of the upstream bottleneck input.
There are several ways in which the regulator can constrain the vertically in-
tegrated firm’s incentives and ability to distort competition in the downstream
market. In particular, the regulator can pursue a range of different models of sep-
aration (see Table 2.3 below). Broadly, each model provides successively stronger
constraints on the ability of the vertically integrated incumbent to act on this in-
centive to distort competition in the downstream market. However, at the same
time, the measures imposed become more intrusive for the firm.
36
Table 2.3 is based on an article by Martin E Cave, ‘Six Degrees of Separation—Operational Separation as
a Remedy in European Telecommunications Regulation’ (2006) 64 Communications & Strategies 89–103, at
<https://mpra.ub.uni-muenchen.de/3572/1/M PRA_paper_3572.pdf>.
47
Models one to seven comprise behavioural remedies. These try to remove the
ability of the vertically integrated firm to engage in discriminatory conduct but
they do not remove the underlying incentives of the vertically integrated firm to
discriminate. They do however allow for flexibility of structures as technology and
market developments occur but because the underlying incentive to discriminate
still exists, they suffer from the asymmetry of information problem, and so require
considerable regulatory scrutiny.
Model eight, in contrast, comprises of a structural remedy, which removes both
the ability and the underlying incentive of the regulated firm to discriminate
against competitors. However, while this model removes the ability and incen-
tive to discriminate, structural separation may not in itself change the bottleneck
division’s incentives to operate efficiently, invest, or deliver a good quality of ser-
vice. Therefore, the NRA would still need to continue to regulate the structurally
separate bottleneck division to protect consumers in the absence of strong com-
petition. Structural separation could additionally see the loss of efficiencies made
possible by a vertically integrated structure, such as cost synergies and the removal
of double mark-ups.37 It is also a one-off intervention that is difficult to reverse. If
technological advances, regulatory changes, and differing trading patterns emerge,
model eight can institutionalize a structure that could become obsolete.
37
One of the oft-c ited benefits of vertical integration is that firms can benefit from a number of cost
synergies. In non-i ntegrated operations, every step in production may involve mark-ups so the reseller can
earn profit. By selling directly to end buyers, vertically integrated firms can ‘cut out the middle man’ removing
one or more steps of mark-ups along the way. This can therefore lead to lower prices for consumers.
48
48 Part I Fundamentals
Given the pros and cons of the different separation models above, whichever
model is adopted is highly dependent on the concerns that have been identified at
a point in time and the costs and benefits of intervention.
We now turn to a discussion as to how some of the concepts above were imple-
mented in the UK. Arguably, the UK experience can be divided into three phases
of liberalization and economic regulation (i) privatization and the early develop-
ment of competition; (ii) the end of the UK duopoly policy; and (iii) increasing con-
vergence between telecoms, broadcasting, and information technology. Below, we
take each of these in turn:
38
See further Chapter 8.
39
DTI, Competition and Choice: Telecommunications Policy for the 1990s (London: HMSO, 1991 cm1461).
50
50 Part I Fundamentals
Commission guidelines on market analysis and the assessment of significant market power under the
42
Community regulatory framework for electronic communications networks and services, OJ C 165/6, 11
July 2002.
43
Ofcom, ‘Strategic review of telecommunications Phase 2 Consultation’, 2004, at <http://webarchive.
nationalarchives.gov.uk/20160702162827/http:/stakeholders.ofcom.org.uk/consultations/telecoms_p2/>.
44
See further Chapter 7.
52
52 Part I Fundamentals
were auctioned. As in the previous auction award, four of these went to existing
mobile companies (EE, O2, Vodafone, and H3G) but the fifth was awarded to a new
player (in this case BT).45
In contrast, for fixed services, the main conclusion from the TSR was that al-
though liberalization had led to a number of downstream retail markets being
opened to competition, the strong presence of common costs and economies of
scale in fixed networks meant that most competitors still rely on upstream whole-
sale inputs provided by the incumbent, BT. Ofcom decided therefore that the
focus of economic regulation should be on opening access to the network elem-
ents subject to scale and scope economies to allow competition to be introduced
in related markets. It argued however that BT’s market power in the provision of
fixed infrastructure and its vertically integrated structure into the downstream
markets for which that infrastructure is a critical input meant that BT (given its
then current structure) would always have an incentive and the ability to engage
in discriminatory behaviour against its competitors. In particular, Ofcom high-
lighted that while price discrimination may be easier to detect, verify, and enforce,
non-price discrimination (such as delaying access to key inputs to competitors etc)
is not. Ofcom believed that its then current powers did not suffice to deal with the
problem. Consequently, it put forward a proposal to introduce a form of functional
separation and to strengthen the then current non-d iscrimination rules (from
model 3 to model 5 in Table 2.3 above). This manifested itself in what is termed
Equivalence of Input (EoI) supported by the organizational separation of BT and
the creation of Openreach as a functionally separate entity.46
In March 2015, ten years after the TSR, Ofcom launched a Strategic Review
of Digital Communications (DCR). Whilst Ofcom set out that the UK’s telecoms
users have enjoyed largely positive outcomes in the last decade, some concerns re-
main.47 In particular, Ofcom highlighted that given the increasing dependence on
communications, more needs to be done to make sure there is widespread avail-
ability of superfast fixed broadband and better mobile coverage. Moreover, more
generally Ofcom pointed out that there are continuing concerns about the quality
of service delivered by some providers and argued that as future demand for data
grows more network investment will be required to deliver it.
45
Ofcom Press Release, ‘Ofcom announces winners of the 4G mobile auction’, 2013, at <http://media.
ofcom.org.uk/news/2013/w inners-of-t he-4g-mobile-auction/>.
46
Ofcom, ‘A notice under Section 155(1) of the Enterprise Act 2002’, 2005, at <http://webarchive.
nationalarchives.gov.uk/2 0160702162827/http:/stakeholders.ofcom.org.uk/ binaries/c onsultations/s ec155/
summary/sec155.pdf>.
47
Ofcom, ‘Initial conclusions from the Strategic Review of Digital Communications’, 2016, at <https://
www.ofcom.org.uk/phones-telecoms-a nd-i nternet/i nformation-for-i ndustry/policy/d igital-comms-review/
conclusions-strategic-review-d igital-Communications>.
53
Ofcom, ‘Making communications work for everyone: Initial conclusions from the Strategic Review of
49
Digital Communications’, 2016, section 6 at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 016/50416/
dcr-statement.pdf>.
50
Ofcom Press Release, ‘BT agrees to legal separation of Openreach’, 2017, at <https://w ww.ofcom.org.uk/
about-ofcom/latest/media/media-releases/2017/bt-agrees-to-legal-separation-of-openreach>.
51
Ofcom, ‘Delivering a more independent Openreach: Update on BT’s voluntary notification under s.89C
Communications Act 2003 and consultation on releasing the BT undertakings pursuant to section 154 Enterprise Act
2002’, 2017 at <https://w ww.ofcom.org.uk/_ _data/assets/pdf_file/0035/98855/Openreach-consultation-2017.pdf>.
54
54 Part I Fundamentals
The discussion above provides a whistle-stop tour of the broad phases of liber-
alization and regulation in the UK since the privatization of BT. In the following
sections, we now turn to a more detailed discussion of the regulations that were
implemented at the retail and wholesale levels drawing on the economic concepts
discussed earlier.
In 1984, the first price cap was set at RPI –3 per cent and the basket of controlled
services covering approximately half of BT’s revenues was set for five years. In the
subsequent Price Review of 1988, Oftel tightened the price cap to RPI –4.5 per cent,
and added additional services to the basket increasing the coverage of the cap to
just over half of BT’s revenues. These arrangements were supposed to stand for
a duration of four years but the Duopoly Review in 1990 meant that these com-
mitments were jettisoned and new arrangements were put in place. These tight-
ened the price cap further from RPI –4.5 per cent to RPI –6.25 per cent. Moreover,
because routine monitoring of BT’s international calls showed profits were rising
sharply, out-going international call charges were added to the price cap basket.
Oftel’s next review of the BT price cap also tightened the cap further to RPI –7.5
per cent and as in the previous review, more services were added to the control. By
1997, nearly 70 per cent of BT’s revenues were covered by the control.
52
Source: Cave, M, ‘The Evolution of Telecommunications Regulation in the UK’ (1997) 41 European
Economic Review 691–699.
5
53
See Section 2.5 for a discussion about how the extent of competition in communications markets is exam-
ined for the purposes of deciding whether regulatory controls are required or not.
54
Wholesale Line Rental (WLR) is intended to stimulate competition by allowing suppliers to provide an
integrated service comprising calls and access, renting the exchange line from BT and sending customers a
single bill.
55
Ofcom, ‘Retail price controls: explanatory statement and proposals’, 2006, at <http://webarchive.
nationalarchives.gov.uk/2 0160702162827/ http:/s takeholders.ofcom.org.uk/ binaries/c onsultations/retail/
statement/r pcstatement.pdf>.
56
56 Part I Fundamentals
exercise choice and switch quickly without loss of service.56 In addition, it has put
in place a number of informational remedies targeted at consumers to help con-
sumers through the process of switching.
To further inform consumers of their available options, Ofcom has also priori-
tized publishing service quality performance data on all operators. This acts as a
bit of a ‘name and shame’ of poor performing operators and so provides a mech-
anism for consumers to be informed of alternative providers and an incentive for
operators to improve their quality of service.57 The incentive for operators to im-
prove their quality of service is also reinforced by Ofcom’s proposal to introduce
automatic compensation for consumers and smaller businesses when things go
wrong.58
Alongside, the demand-side remedies above, Ofcom has also been keeping
an eye on operators’ behaviour at the retail level. Even though retail regulation
had been removed in the UK, in 2016, prompted by concerns over rapidly rising
prices for standalone landline telephone services (ie the sale of telephone services
to those people who buy such services in a standalone contract and not as part
of a bundle with other services such as broadband or pay TV) Ofcom launched a
review of the retail market for standalone landline telephone services.59 In 2017,
it provisionally concluded that there is a distinct market for standalone landline
telephone services, with BT holding significant market power. To prevent BT from
using this market power against standalone landline telephone customers, Ofcom
proposed to regulate BT’s standalone telephony services through a retail price
control, with an initial price cut of between £5 and £7 in monthly line rental, and
a basket cap on prices of line rental and calls to limit future price increases to no
more than the rate of inflation. In addition, Ofcom also proposed to require BT to
work with it to trial—a nd, if appropriate, deliver—consumer information to en-
courage its standalone telephony customers to look for better value deals to pro-
mote competition. In response to these proposals, in October 2017, BT voluntarily
agreed to put these measures in place for a period of three years.60
56
Ofcom, ‘Consumer switching: A statement and consultation on the processes for switching fixed voice
and broadband providers on the Openreach copper network’, 2013, at <http://webarchive.nationalarchives.
gov.uk/2 0160702162827/ http://s takeholders.ofcom.org.uk/ b inaries/c onsultations/c onsumer-s witching-
review/summary/Consumer_Switching.pdf>.
57
Ofcom, ‘Comparing service quality’, 2017, at <https://w ww.ofcom.org.uk/phones-telecoms-a nd-
internet/advice-for-consumers/quality-of-service>.
58
Ofcom, ‘Automatic compensation’, 2017, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 030/
98706/automatic-compensation-consultation.pdf>.
59
Ofcom Press Release, ‘Landline prices review to protect elderly and vulnerable’, 2016, at <https://w ww.
ofcom.org.uk/about-ofcom/latest/features-a nd-news/landline-prices-review>.
60
Ofcom, ‘Review of the market for standalone landline telephone services: statement’, 2017, at <https://
www.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 015/107322/standalone-landline-statement.pdf>.
57
61
Where there is the potential for competition, NRAs need to define markets in a dynamic manner. They
should take account of the changing nature of the industry otherwise erroneous regulatory decisions could be
made which could distort the workings of the sector.
58
58 Part I Fundamentals
where two services are closely linked. Secondly and more importantly from a pol-
itical perspective, rebalancing can be difficult for politicians to sell to the public.
This is because rebalancing generally requires that the majority (ie the voters) pay
more whilst the better-off pay less.
Rebalancing became a prominent issue in the UK when BT made its first price
changes as a private company in the 1980s. It was keen to rebalance quickly as
having tariffs out of balance exposed it to competition targeted exclusively at the
high margin calling business. It argued that line rentals priced below cost ‘distorts
the market and encourages inefficient and misplaced investment’.63 BT managed
to carry out some rebalancing actions but these tended to favour large users over
smaller users and involved price cuts in areas where competition was prevalent
and price increases where it was not. (The incentives on BT for rebalancing were
therefore similar to the incentives produced from price regulation as discussed in
Section 2.7.)
Rebalancing, if pursued too vigorously, could undermine the liberalization pro-
cess. Oftel was concerned about this and so concluded in 1986 that no further re-
balancing between local and long-d istance charges should occur. The view was
that the liberalization process needed to be protected and if effective would lead to
natural cost reductions and rebalancing over a number of years.
An industry exhibiting economies of density is one where the more closely packed together the customers
64
60 Part I Fundamentals
imposing a USO is both social and economic. The social policy goal is to provide
individuals with access to communications facilities to avoid a gulf emerging be-
tween groups in society. The economic rationale, on the other hand, relates to the
presence of externalities, not taken into account by individuals in their private
decision-making. New customers joining a network not only benefit themselves,
but create extra opportunities for existing customers. There is, moreover, evidence
of dynamic benefits for the economy arising from the development of the commu-
nications sector.
The presence of these externalities and the social and political considerations
mentioned above thus create a case for imposing a USO sharing mechanism as
long as it can be proven that costs outweigh benefits. It should, however, be rec-
ognized that the main aim of the obligations will differ at different periods. At the
time of network build-out and mass-market take-up, the objective of universal
service obligations is likely to be primarily economic. Once the network is com-
pleted, however, the goal of universal service shifts to being primarily a social one.
In the former stage, it is desirable to keep installation prices low so as to stimulate
demand and to take account of the network externality. In the latter stage, the em-
phasis is likely to be upon targeting subsidies to ensure that the telephone is af-
fordable to all and adapted to those with special needs.
Traditionally in a monopoly environment the costs of the USO have been
covered by a cross-subsidy. When competition has been introduced, however,
the incumbent has asserted, as with rebalancing, that having a USO, exposes it to
competition targeted exclusively at the profitable business thereby resulting in it
having inadequate funds to cover the costs of serving unprofitable customers. As
a consequence, it claims that the costs of this should be shared amongst operators
to ensure competition on equal terms.
that can be served economically. Once this has been identified, it should then
be possible to derive and cost the shortfall or the cost of serving loss-making
customers. Once the avoidable costs of delivering universal service have been
quantified, any commercial benefits such as good public relations, perception
of marketplace ubiquity, reduced churn, simplified credit procedures arising
from serving remaining customers must be quantified. All incremental revenues
emanating from loss-making customers must also be included on the revenue cal-
culation. This should comprise call charges, line rental charges, as well as revenue
of incoming calls to loss-making customers. This is of key importance to the cal-
culation as this would be lost to the operator if the customer left the network. The
cost of delivering universal service is then the amount by which the cost of serving
loss-making customers exceeds the benefits and incremental revenues associated
with serving these same customers.
65
Oftel, Universal Telecommunications Services, Statement, 1997, at <http://webarchive.nationalarchives.
gov.uk/2 0040104233440/ http:/w ww.ofcom.org.uk/s tatic/a rchive/oftel/publications/1995_ 9 8/c onsumer/
univ_2 .htm>.
62
62 Part I Fundamentals
the delivery of the USO should not be funded by the industry as a whole. Ofcom
stated however that the costs and burden of universal service would be kept under
review.66
If an undue burden from the USO is found the funding methods outlined above
can be used. The efficacy of using a fund or virtual fund could, however, be im-
proved if alternative service providers are present in the market. For example,
permitting the incumbent to contract out the USO to the most efficient service
provider or equally, franchising the USO to the most efficient service provider and
then making the appropriate transfers could potentially improve the efficiency of
USO delivery.
In November 2015, the government set out its intention to introduce a broad-
band USO.67 This is intended to give everyone a right to a decent broadband con-
nection, with a download speed of 10Mbit/s, on request. The government proposed
introducing this USO in recognition of the increasingly important role broadband
plays in people’s lives. Ofcom was commissioned by the government to provide
technical analysis and advice to support the design of the broadband USO.68 In
December 2016, Ofcom published its advice for government on how it might se-
cure its overarching policy objectives.69 This comprised of, amongst other things,
a high-level consideration of the scope and level of the USO, the costs of delivering
the USO, options for USO designation which delivers value for money, views on
the extent of any market distortion arising from the implementation of a broad-
band USO, and mechanisms for funding the USO. In July 2017, BT made an offer
to provide the required infrastructure on a voluntary basis, avoiding the need for
regulation, and recouping the cost through customer bills. In December 2017, the
government rejected this offer and decided to press ahead with adoption of a regu-
latory USO, to provide everyone with minimum 10 Mbps line speeds by 2020.70
66
Notification of Proposals for the Designation of Universal Service Providers and Setting of Conditions,
Consultation Document, 12 March 2003. Under the Communications Act 2003, s 65 et seq, Ofcom have the
right to review such matters for the purpose of making a universal service designation. Document, at <http://
webarchive.nationalarchives.gov.uk/2 0040104233440/http:/w ww.ofcom.org.uk/static/a rchive/oftel/publi-
cations/eu_d irectives/2003/u so0303.htm>.
67
Government press release, ‘Government plans to make sure no-one is left behind on broadband access’,
2015, at <https://w ww.gov.uk/government/news/government-plans-to-m ake-s ure-no-one-i s-left-b ehind-
on-broadband-access>.
68
Letter from DCMS to Ofcom, March 2016, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 027/
53676/dcms_letter.pdf>.
69
Ofcom, ‘Achieving decent broadband connectivity for everyone: Technical advice to UK Government on
broadband universal service’, 2016, at <https://w ww.ofcom.org.uk/__data/a ssets/pdf_fi le/0 028/95581/fi nal-
report.pdf>.
70
See <https://w w w.gov.uk/g overnment/n ews/h igh-s peed-b roadband-t o-b ecome-a -l egal-r ight>.
The obligation will be made through a ‘universal service order’, issued by the Secretary of State under the
Communications Act 2003, s 65 (as amended by the Digital Economy Act 2017, s 1).
63
As detailed in Section 2.4, the value of belonging or being connected to the net-
work increases with the number of people on the network. This means that com-
petition between separate networks is unlikely to be sustainable, as the larger
any one network gets the greater becomes its advantage over the others. The
solution to this problem is for networks to interconnect, in effect forming one
single network. Therefore, regulators need to impose an obligation to intercon-
nect or to terminate calls to enable members of one network to call members of
another.
In addition to ‘interconnection’ services, regulators usually also demand that
incumbent operators provide a number of ‘access’ services72 to competitors at
a regulated price. As noted above, a number of call origination services are often
considered naturally monopolistic because of sunk costs and economies of scale.
Opening access to the network elements subject to scale and scope economies may
allow competition to be introduced in related markets. However, this requires regu-
latory intervention because without it, the incumbent would not have an incentive
to provide access to these network elements.
Whilst intervention to oblige operators to provide access and interconnection
services can be helpful, it provides no guarantee that access and interconnection
will be provided on reasonable terms.73 Interconnection and access charges are
key cost components of entrants’ tariffs.
If high interconnection and access charges are set, new entrants in evaluating
whether they should enter and provide service could conclude that business is
not viable and may not enter the market. Existing entrants’ investment behav-
iour could also be affected. If interconnect and access prices are kept at inflated
levels, existing entrants in the marketplace are likely to try to minimize costs by
bypassing the incumbent’s network and building their own network components
to compensate. However, if these network components have natural monopoly
properties,74 this duplication of network may be inefficient.
64 Part I Fundamentals
In contrast, if interconnect and access charges are set too low, this could inef-
ficiently encourage prospective competitors to buy access rather than build their
own networks. Moreover, if charge levels are too low, the incumbent may be at risk
of not being able to efficiently recover its network costs. If this is the case, it could
result in investor uncertainty and therefore a corresponding decrease in invest-
ment and innovation in the industry. As a result, future network build-out may
be less robust because capital funding that might otherwise have been used for
network and service construction may not readily be available. In addition, tech-
nology choices may be driven by a short-term focus of recovery of network costs ra-
ther than a long-term focus of over-a ll industry growth. This could therefore have
potentially irreversible consequences for service provision and the development
of competition in the industry.
Striking the appropriate balance of interconnect and access charge levels is
crucial to ensure the efficient development of competition and of the industry.
The NRA’s role in setting these charges is critical to establish a sustainable inter-
connect and access regime. To prevent the dominant operator from abusing its
position, the NRA must have the appropriate powers and penalty mechanisms to
control for this, both for creating the conditions for effective competition and also
for a system of minimum regulatory intervention.
75
Determining such costs has been subject to judicial consideration in a number of jurisdictions. See
eg Telecom Corporation of NZ Ltd v Clear Communications Ltd (1992) 4 NZBLC; Verizon v FCC 535 US 467,
122 S Ct 1646, and the CJEU decision in Case C-5 5/0 6, Arcor AG & Co. KG v Bundesrepublik Deutschland, 24
April 2008.
65
example, large parts of the local loop costs can be associated with the provision of
access to customers or the costs of an international switch can be associated with
the provision of international calls. There are, however, costs that could be viewed
as ‘common’ to a number of services that cannot be allocated on a causative basis.
For example, the costs associated with running regulatory departments, legal de-
partments, administration, human resources and the Chief Executive’s office. To
ensure recovery of these common costs, under the FAC method, these are nor-
mally allocated to the respective service on the basis of either output, gross rev-
enues, or the direct costs of each service.
Although FAC (for the purposes of interconnect and access charging) ensures
that all costs are recovered, the procedures for deriving and agreeing the costs is
complex. The process for calculating FAC is heavily reliant on information that
the incumbent dominant operator supplies the regulator. ‘Strategic’ cost alloca-
tions, by the incumbent operator, can allow it to raise competitors’ costs and so
keep them at a permanent cost disadvantage. These ‘strategic’ cost attribution
procedures can be very intricate and complex and so regulatory scrutiny of them
can be difficult. Another difficulty of using FAC is that many costs cannot be al-
located on a causative basis and so must be attributed using a particular rule ie
the output, gross revenues, direct costs of each service etc. However, changing
the rule can substantially change the results. Given the criticisms on FAC, many
have suggested that the use of FAC for interconnect and access services is ineffi-
cient. Consequently, in recent years discussions about the relevant cost standard
for setting interconnection and access charges has shifted towards incremental or
marginal costs.
66 Part I Fundamentals
There are two basic types of marginal cost: Short Run Marginal Cost (SRMC) and
Long Run Marginal Costs (LRMC):
• SRMC reflect the costs that occur when a unit of output is changed when only
some inputs can be varied. Those that cannot be varied therefore represent fixed
costs and are not included in the calculation of SRMC.
• LRMC reflect the costs that occur when a unit of output is changed when all in-
puts can be varied. Hence product specific costs that can be efficiently varied
with marginal changes in output over the long run are included in the calcula-
tion of LRMC.
LRMC is often considered a better measure than SRMC for regulatory purposes.
The reason is because if a company is producing at capacity, increasing output by
one unit could mean significant levels of SRMC, whereas when it is not producing
at capacity SRMC can be negligible. SRMC is quite ‘lumpy’ depending on when
the demand change is assumed to occur. In contrast, because LRMC reflects the
costs that occur when all inputs can be varied in response to a sustained demand
change, this ‘lumpiness’ can be smoothed out. As such, it can provide better sig-
nals to consumers and the market.
LRMC is however rather difficult to calculate in practice. Inputs (such as staff, ve-
hicles, and machines) can only be sensibly varied in larger discrete amounts in re-
sponse to variations in much larger volumes of outputs (and not just an additional
unit of output). Consequently, in practice, the increment of output considered is
often much larger and the cost calculated is long run incremental costs (LRIC).
This is defined as the cost of adding an increment of output. The size of the incre-
ment can either be small (perhaps a 5 per cent change in the volume of a service)
or large—consisting of a whole service or a group of services. So for example, the
incremental cost of long distance calls is the extra costs of providing all long dis-
tance calls given the availability of access expressed on a per unit basis. Thus with
the same output increment, marginal and incremental costs may be the same.
Marginal and incremental costs can be calculated in two ways. The first, known
as ‘bottom-up’ cost modelling involves the construction of an engineering/eco-
nomic model of an optimal telecommunications network. The idea is to design and
cost an efficient network which can meet any given set of demands.76 By changing
certain demand parameters for individual services, it is then possible to calculate
the marginal or incremental costs of that service. The alternative approach, known
as ‘top-down’ cost modelling, starts from the incumbent’s management accounts
76
The bottom-up model in designing an optimal communications network rather than the actual network
in place removes the margin of inefficiency implicit in most incumbent’s network thereby allowing only effi-
cient costs to be recovered through interconnect charges.
67
or fully allocated costs. The first step involves (where necessary) re-valuing the as-
sets on the basis of their replacement (or current) costs (see Section 2.7). The non-
incremental costs such as non-attributive ‘common’ costs are then removed from
the accounts and the remaining costs are used to calculate the incremental costs
of service on a per unit basis.
It is normal for both methods to be used in determining incremental or marginal
costs. The reason is that the two models act as an auditing mechanism for each
other. Each has strengths and weaknesses that ensure that the reconciliation be-
tween the two models allows for only relevant costs to be recovered. If the NRA re-
lied exclusively on the top-down model derived from the fully allocated costs of the
incumbent, the resulting incremental costs could lead to a skewed outcome in fa-
vour of the incumbent where inefficiencies in the incumbent’s cost structure could
be passed on to competitors in interconnection and access charges. The bottom-up
model has the important advantage of being derived in an open and transparent
way but because it is based on a theoretically efficient network rather than the ac-
tual network in place, the assumptions underpinning the model may be unreal-
istic. As a consequence, the use of both models yields advantages and allows the
regulator to scrutinize more closely the cost structure of the incumbent operator.
68 Part I Fundamentals
therefore, the highest mark-up for those services where customers are price
insensitive.
The Ramsey pricing formula for calculating the mark-up depends strongly on
the type of competition between the incumbent operator and its competitors, the
relative sizes of the firms, the differences in the costs of supplying the final output
and the cost of interconnection or access. The complexity of the information re-
quired to put Ramsey prices into practice means that at a practical level, it is diffi-
cult. The problems of estimating the various elasticities which are required in the
Ramsey approach are considerable particularly when dynamic effects have to be
factored in somehow. Therefore, whatever appeal the approach has at a theoretical
level, in practice, pure Ramsey pricing is not often used.
Given the practical difficulties with Ramsey pricing, the alternative is to use an
accounting rule to recover common costs. For example, if the common input was
used to produce two separate, regulated services, one simple rule would be to split
the common cost equally between the two services. An alternative rule is to re-
cover common costs in proportion to the incremental cost of the two services. This
method of allocating costs is known as the equal proportionate mark-up (EPMU)
and is an approach that is often used in the calculation of fully allocated costs to
cover ‘common’ costs. While these rules allow the firm to recover their common
costs, they lack any of the theoretical economic justifications which, despite the
practical problems associated with their implementation, Ramsey pricing at least
partly possesses. It means therefore that the NRA needs to have a level of scrutiny
to ensure that price signals are not distorted.
77
This policy is to be contrasted with the more radical policy in the US of structural separation of the RBOCs
from AT&T. See Chapter 5.
70
70 Part I Fundamentals
78
Network Charges From 1997, Oftel Consultative Document, December 1996, at <http://webarchive.
nationalarchives.gov.uk/20040104233440/http:/w ww.ofcom.org.uk/static/a rchive/oftel/publications/1995_
98/pricing/ncctitle.htm>.
79
See Sections 2.6 and 2.7 for a discussion on the methodology for setting price caps.
80
Oftel Statement, ‘Network charges from 1997’, May 1997, at <http://webarchive.nationalarchives.gov.uk/
20040104233440/http:/w ww.ofcom.org.uk/static/a rchive/oftel/publications/1995_ 98/pricing/ncct797.htm>.
71
81
Standalone costs refer to the costs that would be incurred by an efficient entrant if it were to decide to
produce only a specified set of commodities, e.g. access lines or call minutes. There are generally significant
common costs associated with access lines and minutes. These costs would be incurred regardless of whether
only one service is supplied. This means that the standalone costs of a particular service would be signifi-
cantly higher than the incremental cost of that same service.
82
For more information see, (1) Price Control Review— Possible Approaches for Future Retail and
Network Charge Controls, Consultation March 2000; (2) Price Control Review, Consultation October 2000;
(3) Proposals for Network Charge and Retail Price Controls from 2001. (February 2001). These are available
at <http://webarchive.nationalarchives.gov.uk/20040104233440/http:/w ww.ofcom.org.uk/static/a rchive/
oftel/publications/>.
72
72 Part I Fundamentals
2001, for BT to meet reasonable requests for unbundled access to the local loop.83
The aim of the Regulation was to address the lack of competition on the local net-
work where incumbent operators continued to dominate the market for voice tele
phony services and high-speed internet access. By allowing entrants access to the
incumbent’s local loop (rather than expecting them to build their own local loop),
the Commission believed that increased competition in this area would allow
higher bandwidth services such as high speed always on internet access and video
on demand to develop more rapidly. Further, they believed that increased com-
petitive pressure in this area could lead to a wider range of services for consumers
and better value for money.
In May 2000, Oftel published a consultation document proposing prices for op-
erators leasing unbundled loops.84 The key pricing principles were that the price
of the loop would be cost oriented, the starting charges would be geographically
averaged, and that BT should be able to recover the costs associated with setting
up co-location facilities. On 29 December 2000, Oftel published the final wholesale
prices to be applied until 31 March 2002 and suggested that in April 2002, it would
introduce an RPI –X cap on the charges.85 In March 2002, Oftel concluded that the
market for the provision of LLU services had not developed as quickly as origin-
ally anticipated. However, being aware of the forthcoming European Directives, it
decided to roll over the price controls from December 2000 and said that it would
review in early 2003.
83
Regulation (EC) No 2887/2000 of the European Parliament and of the Council of 18 December 2000 on
unbundled access to the local loop, OJ L 336/4, 30 December 2000.
84
Oftel, ‘Access to Bandwidth: Indicative prices and pricing principles’, 2006, at <http://webarchive.
nationalarchives.gov.uk/2 0040104233440/ h ttp:/w w w.ofcom.org.uk/s tatic/a rchive/o ftel/p ublications/
broadband/l lu/l lu0500.htm>.
85
Oftel, ‘Determination under Condition 83.16 of the Licence of British Telecommunications Plc relating
to the charges for the provision of metallic path facilities and associated internal tie circuits’, 2000, at <http://
webarchive.nationalarchives.gov.uk/2 0040104233440/http:/w ww.ofcom.org.uk/static/a rchive/oftel/publi-
cations/broadband/l lu/l lup1200.htm>.
86
Leased lines are permanently connected communications links that are used by business and other op-
erators for services such as voice and data traffic and internet access.
73
87
It maintained safeguard caps on analogue retail leased lines since the competitive pressures created by
its wholesale policy options were likely to stimulate sufficient retail competition to constrain retail prices for
all other services.
88
Oftel, ‘Phase 1 direction to resolve a dispute concerning the provision of partial private circuits’, 2002,
at <http://webarchive.nationalarchives.gov.uk/20040104233440/http:/w ww.ofcom.org.uk/static/a rchive/
oftel/publications/broadband/leased_l ines/ppcs0602.htm>.
89
Oftel, ‘Partial Private Circuits, Phase Two—a Direction to resolve a dispute concerning the provision of par-
tial private circuits’, 23 December 2002, at <http://webarchive.nationalarchives.gov.uk/20040104233440/http:/
www.ofcom.org.uk/static/a rchive/oftel/publications/broadband/leased_l ines/ppc1202/d irection.htm>.
74
74 Part I Fundamentals
leased lines that were considerably lower than BT charges (typically 50 per cent lower
for connection and 20 per cent lower for rental). Further, it backdated these charges
to the launch of the products, which meant that BT had to provide considerable re-
funds to operators. In addition, Oftel confirmed a number of improvements for BT to
make to its service level agreement. These improvements included BT paying com-
pensation to other operators in the event of late delivery.
90
See Chapter 4.
Oftel EU directive implementation, at <http://webarchive.nationalarchives.gov.uk/20040104233440/
91
between independent service providers and the service providers of the operators
with market power, while allowing for the regulated firms to set charges according
to their commercial judgment.
The TSR led to the implementation of two main interventions. The first in-
volved a renewed focus on and increased use of LLU. Given the scale and scope
economies in networks, there was a recognition that it would be difficult to get
multiple competing networks so the aim was to encourage the number of com-
petitors to BT in residential telecoms services via access regulation. By using
charge controls, this promoted market entry by scale competitors to BT who
invested in installing equipment and backhaul in local telephone exchanges,
while maintaining the opportunity for BT to make a fair return. The second
addressed the concerns about non-price discrimination. To ensure that com-
petitors were granted access to infrastructure on an equal basis, two parallel
interventions—equivalence of inputs and the organizational separation of BT—
were imposed.
More generally, the TSR laid out seven principles for regulation to address the
issues in the market at the time:92
92
See para 1.25 of Ofcom, ‘Strategic Review of Telecommunications Phase 2 Consultation’, 2004 (link
at n 43).
76
76 Part I Fundamentals
Following the TSR and the removal of regulatory controls in retail services, Ofcom
undertook market reviews following the economic principles outlined above.
These reviews essentially covered the following:
Below we consider each of them in turn and set out how economic regulation has
evolved in these markets since the TSR.93
93
See also Chapter 8, at Section 8.5.
94
Ofcom, ‘Explanatory Statement and Notification of decisions on BT’s SMP status and charge controls
in narrowband wholesale markets’, 2005, at <http://webarchive.nationalarchives.gov.uk/20160702162827/
http:/stakeholders.ofcom.org.uk/consultations/charge/statement/>.
7
95
Ofcom, ‘Review of BT’s network charge controls, statement’, 2009, at <http://webarchive.nationalarchives.
gov.uk/20160702162827/http:/stakeholders.ofcom.org.uk/consultations/review_bt_ ncc/statement/>.
96
Ofcom, ‘Review of the fixed narrowband services wholesale markets: Consultation on the proposed
markets, market power determinations and remedies’, 2009, at <http://webarchive.nationalarchives.gov.
uk/2 0160702162827/http:/stakeholders.ofcom.org.uk/binaries/c onsultations/review_w holesale/summary/
fnwm.pdf>.
97
WLR stands for Wholesale Line Rental. It is a facility which allows alternative providers to rent access
lines on wholesale terms from BT, and resell the lines to customers, providing a single bill that covers both
your line rental and calls.
98
This charge control comprised of three baskets: WLR Rental with a cap of RPI –7.3%; WLR transfer with
a cap of RPI and WLR new connection with a cap of RPI –10.2%. This control applied from on 1 April 2012 until
1 April 2014.
99
Ofcom, ‘Fixed access market reviews: wholesale local access, wholesale fixed analogue exchange lines,
ISDN2 and ISDN30: Consultation on the proposed markets, market power determinations and remedies’,
2013, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 033/76497/fi xed-access-markets.pdf>.
100
Ofcom, ‘Fixed access market reviews: Approach to setting LLU and WLR Charge Controls’, 2013, at
<https://w ww.ofcom.org.uk/consultations-a nd-statements/c ategory-1/l lu-w lr-cc-13>.
101
As set out in Section 2.14, pure LRIC is the cost of adding an increment of output. LRIC+, in contrast in-
cludes an EPMU mark-up for common cost recovery.
78
78 Part I Fundamentals
with effect from 1 January 2014. This meant that common costs should no longer
be required from wholesale call termination and instead operators would need to
recover common costs from other services (ie origination services). This followed
an EU Recommendation in 2009102 that termination charges should be based on
pure LRIC and also the fact that mobile call termination was based on pure LRIC.
102
Commission Recommendation (2009/ 396/
EC) on the Regulatory Treatment of Fixed and Mobile
Termination Rates in the EU, OJ L 124/67, 25 May 2009.
103
Ofcom, ‘Valuing copper access: Final statement’, 2005, at <http://webarchive.nationalarchives.gov.
uk/2 0160702162827/http:/stakeholders.ofcom.org.uk/ binaries/c onsultations/c opper/statement/statement.
pdf>.
104
Ofcom, ‘Local loop unbundling: setting the fully unbundled rental charge ceiling and minor amend-
ment to SMP conditions FA6 and FB6’, 2005, at <http://webarchive.nationalarchives.gov.uk/20160702162827/
http:/stakeholders.ofcom.org.uk/binaries/consultations/l lu/statement/l lu_ statement.pdf>.
105
Ofcom, ‘Review of the wholesale local access market: Statement on market definition, market power
determinations and remedies’, 2010, at <http://webarchive.nationalarchives.gov.uk/20160702162827/http:/
stakeholders.ofcom.org.uk/binaries/consultations/w la/statement/ W LA_ statement.pdf>.
79
This regulatory framework set out that VULA would likely be attractive for com-
munications providers where BT had already upgraded its local access network;
PIA would be attractive to companies wishing to address market opportunities in
advance of BT and may also be of interest to companies wishing to provide service
in locations which may be in receipt of public funding support. The remedies were
complemented by other measures such as Sub-loop Unbundling (SLU),106 charge
controls for LLU107 but greater freedom for BT in the pricing of VULA services.108
This greater freedom for BT was to account for the risk in investment and the initial
small scale of adoption of NGA services. These remedies were therefore designed
to promote access competition, protect customers, and balance the incentives for
companies facing what remained risky investments.
In 2014, Ofcom carried out a further review of the WLA market and concluded
that the core elements (set out above) continued to be important.109 However,
VULA was increasingly becoming an important input for CPs to provide NGA
services in competition with BT and so Ofcom placed a requirement on BT to
supply a VULA product to competitors who wanted it. Ofcom considered that,
in the absence of such a requirement, BT would have an incentive and ability to
refuse access at the wholesale level and so favour its own retail operations with
the effect of hindering sustainable competition in the downstream market, ul-
timately against the interests of end-u sers. Ofcom did not however implement
106
SLU allows originating communications providers (OCPs) to physically take over (or share) the part of
BT’s existing copper lines between a street cabinet and the customer premises. This remedy will allow OCPs to
deploy fibre to the cabinet technology where they consider this to be economic.
107
Ofcom, ‘Charge control review for LLU and WLR services’, 2012, at <http://webarchive.nationalarchives.
gov.uk/2 0160702162827/ http:/s takeholders.ofcom.org.uk/ binaries/c onsultations/w lr-c c-2 011/s tatement/
statementMarch12.pdf>.
108
Ofcom believed that by just controlling the prices of the copper remedies, this would act as a constraint
on BT’s pricing of VULA.
109
Ofcom, ‘Fixed access market reviews: wholesale local access, wholesale fixed analogue exchange lines,
ISDN2 and ISDN30 Volume 1: Statement on the markets, market power determinations and remedies’, 2014, at
<https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 032/78863/volume1.pdf>.
80
80 Part I Fundamentals
• those geographic areas covered by exchanges where KCOM is the only operator
(‘the Hull area’);
• those geographic areas covered by exchanges where BT is the only operator
(‘Market 1’);
110
Ofcom, ‘Fixed access market reviews: Approach to the VULA margin’, 2015, at <https://w ww.ofcom.org.
uk/__data/a ssets/pdf_fi le/0 015/72420/v ula_margin_fi nal_statement.pdf>.
111
Ofcom, ‘Review of the wholesale broadband access markets 2006/07’, 2007, at <http://stakeholders.
ofcom.org.uk/consultations/w bamr07/summary>.
81
• those geographic areas covered by exchanges where there are two or three prin-
cipal operators AND exchanges where there are four or more principal operators
but where the exchange serves fewer than 10,000 premises (‘Market 2’); and
• those geographic areas covered by exchanges where there are four or more
principal operators and where the exchange serves 10,000 or more premises
(‘Market 3’).
Ofcom found KCOM had SMP in the Hull area112 and that BT had SMP in Market 1
and, separately, in Market 2. However, because of the rapidly changing competi-
tive conditions Ofcom found that no operator had SMP in Market 3 on a forward-
looking basis. In light of its SMP assessment, Ofcom directed BT to provide access
on non-d iscriminatory terms and to publish a reference offer. Separate to this re-
view, BT also made certain pricing commitments to the industry and Ofcom. In
particular, it committed to reducing the price of its wholesale broadband services,
in all parts of the UK, year-on-year until the end of 2010. BT also committed to
supply wholesale broadband services and to not unduly discriminate, in all parts
of the UK, until the end of 2008 and it committed to provide a period of stability for
LLU by not introducing geographically targeted reductions, below a certain level,
to its wholesale broadband prices.
In 2010, Ofcom completed another review of the wholesale broadband access
market.113 Ofcom found that there was effective competition in almost 80 per cent of
the UK. However, in just over one-fifth of the UK—covered by what it called Market
1 and Market 2—it concluded that there was not sufficient competition. Market 1
was made up of exchange areas in which BT was the only provider of wholesale
broadband services, whereas Market 2 comprised of exchange areas with two sig-
nificant providers or with three significant providers where BT’s market share was
50 per cent or more. For Market 1, Ofcom decided that BT should be subject to a
charge control.114 The charge control was imposed on the main product used by
competitors and so Ofcom believed that charge controlling this product directly
protected most consumers in Market 1 and constrained BT from excessive charging
on the other products available in Market 1. The charge control took the form of
RPI –12.00 per cent with a duration until 31 March 2014. In addition, Ofcom set a
number of RPI –0 per cent sub-caps for a number of services within the basket, to
112
Hull is an area in the UK, which is not served by BT but instead is served by KCOM Group (formerly
known as Kingston Communications).
113
Ofcom, ‘Review of the wholesale broadband access markets: Statement on market definition, market
power determinations and remedies’, 2010, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 028/
37666/w bastatement.pdf>.
114
Ofcom, ‘WBA charge control: Charge control framework for WBA Market 1 services’, 2011, at <http://
stakeholders.ofcom.org.uk/binaries/consultations/823069/statement/statement.pdf>.
82
82 Part I Fundamentals
ensure that charges for these services did not increase in real terms over the charge
control period.
In 2014, Ofcom completed the next review of the wholesale broadband access
market.115 Taking account of market and competitive developments, it defined
three distinct markets: Market A—where no more than two operators are present
or forecast to be present, which accounts for 9.5 per cent of UK premises; Market
B—i n which there is effective competition, accounting for 89.8 per cent of prem-
ises; and the Hull Area—0.7 per cent of UK premises, where KCOM is the only sig-
nificant provider. According to Ofcom, Market A tends to be in the most rural and
remote parts of the country. As Ofcom found effective competition in Market B, it
did not impose regulation in that market and removed regulation in those parts of
Market B where there was currently regulation—approximately 12 per cent of UK
premises. In Market A, where it found BT to have SMP, it implemented a charge
control at a level of CPI-10.7 per cent until 31 March 2017.116
115
Ofcom, ‘Review of the wholesale broadband access markets: Statement on market definition, market
power determinations and remedies’, 2014, at <http://stakeholders.ofcom.org.uk/binaries/consultations/
review-w ba-markets/statement/ W BA-Statement.pdf>.
116
In January 2013, the Office of National Statistics (ONS) announced the outcome of its October 2012 con-
sultation on RPI. The ONS concluded that the RPI ‘does not meet international standards . . .’. In light of this,
Ofcom has decided to use CPI as the standard measure of inflation in its charge controls.
117
Transmission of voice and of data and data transmission is symmetrical when upload speeds are the
same as download speeds.
118
Ofcom, ‘Leased lines charge control—Statement’, 2009, at <http://webarchive.nationalarchives.gov.uk/
20160702162827/http://stakeholders.ofcom.org.uk/consultations/l lcc/statement/>.
83
ones expired. However, due to some accounting amendments that BT made to its
regulatory accounts, which required detailed independent scrutiny ahead of setting
the new charge controls, it had to delay the start of the charge controls. As such,
whilst the review of BT’s accounts was taking place, it sought a commitment from BT
that it would not increase prices in nominal terms and that the charge control would
be backdated to 1 October 2008. In 2009, Ofcom set a charge control comprising of
six baskets with a number of sub-caps and other safeguards to reduce the likelihood
of undue price discrimination. These charge controls ran until 2012 at which point,
Ofcom conducted another BCMR.119 The main difference between the 2013 review
and that carried out in 2008 was:
• There were separate markets identified for regional and national TI trunk con-
nectivity. In the previous review of the market Ofcom defined a single TI trunk
market;
• Ofcom defined a wholesale multiple interface (MI) market which included any
service faster than 1Gbit/s and any service delivered with wavelength-d ivision
multiplex (WDM)120 equipment at the customers’ premises, irrespective of band-
width and interface; and
• Ofcom determined that separate geographic markets existed (i) in the Hull area
for all wholesale leased lines, and (ii) in a defined area of London and including
Slough (the Western, Eastern, and Central London Area, or WECLA) for all the
defined wholesale symmetric broadband origination product markets other
than the low bandwidth (up to and including 8Mbit/s) and very high bandwidth
(622Mbit/s) TISBO markets.
Based on these revised market definitions, Ofcom found that BT had SMP in the
AI, TI, and MI markets. It found though that the WECLA and very high band-
width TISBOs were competitive. Ofcom also found BT to have SMP in regional
trunk TI segments. In response to these SMP findings, Ofcom put in place a
charge control with a duration of three years until 2016 with several sub-c aps
and safeguard caps comprising of two separate service baskets for wholesale
services:
• TI at RPI + 2.25 per cent—covering low, medium, and high bandwidth services
outside the WECLA, low bandwidth services within the WECLA, and regional
trunk services at all bandwidths; and
119
Ofcom, ‘Business Connectivity Market Review’, 2013, at <http://stakeholders.ofcom.org.uk/consult-
ations/business-connectivity-m r/fi nal-statement/>.
120
This technology can multiply by several times the bandwidth transmissible in an optical fibre. WDM
equipment allows providers to aggregate traffic from different services and to use optical fibres efficiently in
the core of their networks as demand for bandwidth continues to increase.
84
84 Part I Fundamentals
121
Active products include the physical elements of the network (ie duct, access to poles, copper, fibre) and
the electronic equipment to provide service.
122
ie Ofcom would now consider whether regulation should be based on passive products (just the physical
elements of the network: duct, poles, copper, fibre) because in doing so, it may encourage competitors to invest
in building competing networks to BT.
123
Ofcom, ‘Narrowband Market Review: Consultation on the proposed markets, market power determin-
ations and remedies for wholesale call termination, wholesale call origination and wholesale narrowband
access markets’, 2016, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 016/95011/Narrowband-
Market-Review.pdf>.
85
Wholesale local access (WLA) market review Ofcom carried out a market re-
view of WLA between March 2017 and March 2018. As in the previous review, Ofcom
found BT to have SMP and so required BT to continue to provide access to LLU and
VULA (amongst other services).125 It additionally decided to include a direction-
making power enabling Ofcom to set appropriate quality of service standards on BT.
It imposed a cost-based charge control on the main form of LLU126 (MPF) and the
supporting services used by BT’s competitors (referred to as ancillary services) but
it removed the specific network access obligation and charge control on SMPF. In
setting a cost-based control, it said that it would seek to allow BT the opportunity
to recover the costs of network deployment, to the extent such costs are efficiently
incurred. In other words, if there were costs incurred in network expansion that pro-
vide customers with an improved quality of broadband service, then these should be
considered in setting those controls.
Ofcom recognized in the review that the PIA remedy it had imposed in 2010 suf-
fered from some limitations. This meant that there had been limited take-up of PIA
to date in the UK. To make it easier and more cost effective for telecoms providers
to invest in advanced, competing infrastructure (in line with the conclusion of the
DCR), in December 2016 and April 2017 Ofcom published proposals to develop an
effective remedy for access to BT’s ducts and telegraph poles, which were finalized
in February 2018.127, 128, 129 These proposals are aimed to address concerns from BT’s
124
Ofcom, ‘Narrowband Market Review’, 30 November 2017, at <https://w ww.ofcom.org.uk/_ _data/a ssets/
pdf_ fi le/0 020/108353/fi nal-statement-narrowband-market-review.pdf>.
125
Ofcom, ‘Wholesale Local Access Market Review—Statement—Volume 1—Markets, market power de-
terminations and remedies’, 2018, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 020/112475/w la-
statement-vol-1.pdf>.
126
There are two forms of LLU—Metallic Path Facility (MPF) and Shared Metallic Path Facility (SMPF). MPF
allows providers to offer both voice and broadband services. SMPF allows providers to offer only broadband
services over the copper network. This means that one provider can provide broadband services to the cus-
tomer while another provider supplies voice services on the same line.
127
Ofcom, ‘Wholesale Local Access Market Review: Initial proposals to develop an effective PIA remedy’, 2016, at
<https://www.ofcom.org.uk/__data/assets/pdf_file/0024/95109/Wholesale-Local-Access-Market-Review.pdf>.
128
Ofcom, ‘Wholesale Local Access Market Review: Consultation on duct and pole access remedies’, 2017, at
<https://www.ofcom.org.uk/__data/assets/pdf_file/0008/101051/duct-pole-access-remedies-consultation.pdf>.
129
Ofcom, ‘Wholesale Local Access Market Review: Statement—Volume 3: Physical infrastructure access
remedy’, 2018, at < https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 023/112469/w la-statement-vol-3.pdf>.
86
86 Part I Fundamentals
competitors about the absolute costs and time required to build ultrafast broadband
networks at scale. Following consultation of the proposals, a final decision was pub-
lished in March 2018.
As regards VULA, while in previous reviews, there had been no explicit pricing
controls (to encourage investment), this time around, Ofcom said that the controls
of standard broadband services were unlikely to sufficiently constrain BT’s
superfast broadband prices over the period of this market review.130 Consequently,
it said that there was a significant risk that retail competition would be weaker and
consumers would face considerably higher prices if there was no control on VULA
pricing. In striking a balance between protecting consumers and competition in
the short term while encouraging network investment, it concluded therefore that
for the lower bandwidth VULA product, BT’s prices should be subject to a charge
control rather than the VULA margin test (it set in 2015—see WLA discussion in
Section 2.15.1.6). However, it stated that BT would continue to have pricing flexi-
bility on other higher bandwidth variants of VULA but because there are controls
on the lower bandwidth service, it should provide sufficient protection to superfast
broadband customers from the risk of higher prices, while allowing other telecoms
providers to compete with BT for those customers as well as preserving BT’s incen-
tives to invest.
Wholesale broadband access market review As set out above, the wholesale
broadband access market sits between the retail broadband market and the WLA
market. Given that remedies in the WLA market are still under consultation, it be-
came clear that given the linkages between the two markets, there would be delays
in implementing new controls in WBA from April 2017 (when the controls expire).
As such, Ofcom asked BT to make a voluntary price commitment to cover the
period between the expiry of the current controls and the commencement of the
new controls. In August 2016, BT committed to keep prices in Market A to a level
of CPI-CPI to 31 December 2017. In June 2017, Ofcom issued provisional conclu-
sions.131 As in the previous review, it identified two markets: Market A where no
more than two operators are present and Market B in which there is effective com-
petition. Ofcom said that the size of Market A, where BT has SMP should reduce
to 2 per cent of UK premises from the previous 9.5 per cent. Given these findings,
Ofcom proposed not to put a charge control in place on any WBA services as it
130
Ofcom, ‘Wholesale Local Access Market Review: Statement—Volume 3: Physical infrastructure access
remedy’, 2018, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 023/112487/w la-statement-vol-2 .pdf>.
131
Ofcom, ‘Wholesale Broadband Access Market Review: Consultation on market definition, market power
determinations and remedies’, 2017, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 013/103180/
wba-consultation.pdf>.
87
considers that BT’s retail national pricing and the level of competition in the rest of
the country acts as a constraint to prevent consumers facing excessive retail prices
in Market A.
Based on these SMP findings, unlike in previous reviews, Ofcom decided that they
would put in place two remedies to operate concurrently to promote competition
in the provision of leased lines:
132
Ofcom, ‘Business Connectivity Market Review: Final Statement’, 2016, at <https://w ww.ofcom.org.uk/
consultations-a nd-statements/c ategory-1/business-connectivity-market-review-2015>.
133
WDM allows a single fibre to carry several leased line services simultaneously.
8
88 Part I Fundamentals
In previous reviews and following the conclusion from the TSR, reliance was pre-
dominantly based on active remedies but the view was taken in the 2016 BCMR
that there should be a transition to passive remedies to provide incentives for ef-
ficient investment for BT and for rival infrastructure operators (in line with the
conclusion from the DCR). In implementing these two remedies, Ofcom recog-
nized that a dark fibre remedy would carry some risks relative to an actives-only
remedies package. These include the potential for inefficient entry incentivized
by regulatory arbitrage opportunities, which could result from any inconsisten-
cies between the pricing of active and dark fibre products. Given this, Ofcom de-
termined that BT should provide dark fibre at a price consistent with its 1Gbit/s
wholesale Ethernet leased line services. More specifically, Ofcom specified that
BT, from 1 October 2017, will be required to provide dark fibre at the same price as
the 1Gbit/s active service, minus the long run incremental costs of the active elem-
ents of that 1Gbit/s service—called the ‘active-m inus’ pricing approach.
Ofcom considered that this approach results in a charge consistent with the de-
sign of the active controls which it was imposing on BT (described below) and so
would provide incentives for efficient investment for BT and for rival infrastruc-
ture operators. They argued that it should incentivize use of dark fibre where it
provides benefits relative to active remedies and it should ensure that BT will con-
tinue to have a fair opportunity to recover its efficiently incurred costs.
For the active remedy, as in previous reviews, Ofcom put in place a charge con-
trol with a duration of three years until 2019 with several sub-caps and safeguard
caps comprising of two separate service baskets for wholesale services comprising
of a TI service basket (based on the product definition above) at CPI-3.50 per cent
and an Ethernet service basket (based on the product definition above) at CPI-
13.50 per cent. In addition, Ofcom proposed significant one-off charge reductions
to both BT’s Ethernet and TI charges to reflect that BT’s returns in these markets
were significantly more than its cost of capital.
In response, BT appealed Ofcom’s decision and alleged errors concerning
market definition and alleged errors concerning the remedies imposed. In July
2017, the Competition Appeals Tribunal (CAT) issued a short statement quashing
Ofcom’s decision in relation to its definition of the market.134 It has since provided
its reasoning and has remitted matters back to Ofcom for reconsideration.135
134
British Telecommunications plc v Office of Communications (Market definition Ruling) [2017] CAT 17.
135
British Telecommunications plc v Office of Communications (Judgment Market Definition) [2017] CAT 25.
89
Prominent features have however been the pattern of ‘rolling back’ regulation as
competition takes hold and targeting regulatory controls where competition is
ineffective.
Since the TSR, many operators and in particular BT (in 2009) have started to adopt
and invest in communications networks with the capability to provide superfast
broadband. This has been driven by consumer demand for bandwidth. Investment
in superfast broadband technologies is risky because of cost and demand uncer-
tainty. Given the riskiness of this investment, regulators have had to strike an ap-
propriate balance in ensuring investment incentives, promoting competition, and
protecting consumers where competition is not effective or sustainable.
The approach taken in the UK to ensuring investment incentives has involved
what is termed ‘a fair bet’ approach. Under this approach, if, at the time of invest-
ment, the expected return is equal to the cost of capital, the firm should be allowed
to enjoy some of the upside risk when demand turns out to be higher than expected
(ie it allows returns higher than the cost of capital) to balance the fact that the firm
will earn returns below the cost of capital if demand turns out to be low. In theory,
the ‘fair bet’ approach should not undermine investment incentives and should
provide the firm with a fair opportunity to recover its investment. Essentially this
provides regulatory certainty to firms and means that investors can commit funds
for investment with confidence that the regulator will not act in a way which would
lead to the investor not having the opportunity to recover its costs.
To encourage competition, the UK (since the TSR) has placed considerable focus
on the provision of active access products136 designed to give other communications
providers the ability to compete effectively downstream with BT. Further, Ofcom has
continued to focus on ensuring equivalence of input and functional separation to
ensure that competitors are treated in a non-discriminatory manner. To further en-
courage investment, Ofcom has also given BT some pricing freedom in setting the
wholesale price for VULA to account for the risk in investment and the initial small
scale of adoption. However, it has done so because there is an ongoing constraint
from current generation copper-based broadband services (which are price regu-
lated). By continuing to have price controls on current generation copper-based
broadband services, it reduces the risk of consumer detriment by constraining BT’s
ability to charge excessive prices on superfast broadband. It also protects consumers
during the change to superfast broadband and means that consumers of existing
136
The focus on ‘active’ wholesale products reflects Ofcom’s assessment in the TSR that investment in infra-
structure by other network providers was unlikely. However, to safeguard the opportunity for further competition
based on physical infrastructure access it implemented passive remedies (poles and duct access) and mandated
sub-loop unbundling (a type of unbundled access whereby a sub-section of the local loop is unbundled. In prac-
tice this often means the competitor placing a small street cabinet with a DSLAM, next to a telco local copper ag-
gregation cabinet using a ‘tie cable’ to connect to the last part of the local loop into customers’ homes).
90
90 Part I Fundamentals
services are not made worse off by the adoption of new technology, and the price of
these basic services provides a competitive constraint to the pricing of new services
which are not price controlled.
Whilst Ofcom’s overarching strategy has focused on active access products to
promote competition in superfast broadband, following the DCR, it is now con-
sciously testing the market to see if there is further appetite for investment in fibre
through passive infrastructure access (PIA). In residential markets, Ofcom already
requires BT Openreach to allow operators to deploy NGA networks in the physical
infrastructure of its access network (ie via ducts and poles). This allows other op-
erators to deploy their own fibre to serve customers on their own networks—a n
alternative to VULA to deliver superfast broadband. However, to date there has
been no interest in using PIA by other communications providers unless PIA is
also extended to the business market.
The 2016 BCMR did consider whether the PIA remedy should be extended to the
business market but the conclusion was to impose dark fibre and not duct access. The
main reason was because most of the benefits of passive remedies could be achieved
via dark fibre and a dark fibre remedy would allow Ofcom to manage the implemen-
tation risks during a transitional period whilst active remedies and passive remedies
coexist. In contrast, with a PIA remedy, it would be more difficult to manage prices
at different levels in the value chain to avoid creating incentives for inefficient entry
while active remedies are an important part of the remedy package. Ofcom did how-
ever say that once competition based on dark fibre proves effective and, active rem-
edies can be removed, the pricing of dark fibre and duct access could be made more
compatible. It appears therefore that the intention following the DCR is that much
more emphasis will be placed in the future on passive remedies (comprising of dark
fibre and duct access) but the UK will need to go through a transition phase to get
there, which involves running active and passive remedies concurrently.
has no choice but to call the network to which the called party has subscribed.
This means that mobile operators, in common with other network operators, are
able to set charges for call termination, without reference to significant competi-
tive pressures. Given this, Oftel initiated an investigation. The main preliminary
finding from this work was that BT’s prices for calls to Vodafone and Cellnet cus-
tomers were too high which was mainly caused by Vodafone’s and Cellnet’s high
termination charges. Oftel had the option to impose price controls but it recog-
nized that such action would have a significant impact on the whole of the mo-
bile market. This is especially so, given that the commercial strategy of most UK
mobile operators was to subsidize handsets to encourage take-up of service. Any
potential price control on termination rates would have had a knock-on impact on
the pricing structure for handsets and calls from mobile networks. Given this, in
March 1998, Oftel referred the issue of prices of calls to Vodafone and Cellnet to the
then Monopolies and Mergers Commission (MMC).137
In December 1998, the MMC completed its investigation and concluded that
there was insufficient competitive constraint on termination charges.138 It con-
sidered that the only effective means of remedying or preventing any adverse
effects would be to impose a price control on termination. It thus proposed that
Cellnet and Vodafone should reduce their weighted average termination charges
by RPI –9 per cent until 2001/02.
In February 2001, Oftel carried out a review of the price controls, noting that
although the market had grown rapidly and at a rate much greater than that
predicted, there was still an incentive for each of the mobile network operators
(MNOs) to charge termination rates above the competitive price.139 In light of this,
Oftel concluded that controls on termination charges on the four main mobile net-
works were needed to protect consumers and proposed a charge control of RPI –12
per cent each year for the four years until March 2006.140 The MNOs objected to this
proposal, stating that it was inappropriate to view call termination as a separate
137
Prices of Calls to Mobiles Statement, March 1998, at <http://webarchive.nationalarchives.gov.uk/
20040104233440/ http://w ww.ofcom.org.uk/s tatic/a rchive/oftel/publications/1995_ 9 8/pricing/c tm0398.
htm>. The MMC was first replaced by the Competition Commission (CC), which has since been replaced by
the Competition and Markets Authority (CMA).
138
Reports on references under section 13 of the Telecommunications Act 1984 on the charges made by
Cellnet and Vodafone for terminating calls from fixed-l ine networks, at <http://webarchive.nationalarchives.
gov.uk/2 0040104233440/ h ttp://w w w.ofcom.org.uk/s tatic/a rchive/o ftel/p ublications/1995_ 9 8/p ricing/
cmmc1298.htm>.
139
Review of the Price Control on Calls to Mobiles, February 2001, at <http://webarchive.nationalarchives.
gov.uk/2 0040104233440/ h ttp://w ww.ofcom.org.uk/s tatic/a rchive/o ftel/p ublications/m obile/c tom0201.
htm>.
140
Review of the Charge Control on Calls to Mobiles, 26 September 2001, at <http://webarchive.
nationalarchives.gov.uk/2 0040104233440/ h ttp://w w w.ofcom.org.uk/s tatic/a rchive/o ftel/p ublications/
mobile/c tm0901.htm>.
92
92 Part I Fundamentals
141
Reports on references under section 13 of the Telecommunications Act 1984 on the charges made by
Vodafone, O2, Orange, and T-Mobile for terminating calls from fixed and mobile networks, at <http://
webarchive.nationalarchives.gov.uk/ 2 0040104233440/ h ttp:// w w w.ofcom.org.uk/s tatic/a rchive/o ftel/
publications/mobile/c tm_ 2003/i ndex.htm>.
142
See further Section 2.15 and Chapter 4.
143
Review of mobile wholesale voice call termination markets—E U Market Review, at <http://webarchive.
nationalarchives.gov.uk/2 0040104233440/ http://w ww.ofcom.org.uk/s tatic/a rchive/oftel/publications/eu_
directives/2003/c tm/c tm0503.pdf> and Wholesale Mobile Voice Call Termination: Proposals for the identi-
fication and analysis of markets, determination of market power and setting of SMP Explanatory Statement
and Notification, 19 December 2003, conditions available at <http://webarchive.nationalarchives.gov.uk/
20160702162827/http://stakeholders.ofcom.org.uk/consultations/mobile_c all_termination/>.
93
services, that they should (a) provide network access for the purposes of 2G call
termination; (b) not unduly discriminate in the provision of such access; (c) pub-
lish a Reference Offer; (d) give prior notification of price changes; and (e) reduce
termination charges in line with the proposed charge controls by the CC. In re-
spect of 3G voice call termination services, it recommended that there should be
no ex- ante regulation although H3G was required to give advance notification of
price changes and provide Ofcom with details of call volumes.
Given that Ofcom effectively designated all five MNOs as having significant
market power, H3G subsequently appealed its SMP designation to the Competition
Appeals Tribunal (CAT) on the grounds, among others, that Ofcom did not carry out
sufficient analysis of prices to entitle it to come to a decision that H3G had signifi-
cant market power and, failed to take account or sufficient account, of the ability of
BT to restrain pricing, in reaching its conclusions.144 The CAT, in November 2005,
found that Ofcom erred in its SMP determination since it did not conduct a full
assessment of the extent to which BT had countervailing buyer power. As such the
CAT remitted the decision back to Ofcom to reconsider.
In March 2007, Ofcom published its assessment and concluded that there are
separate markets for the provision of wholesale mobile voice call termination in
the UK to other communications providers and that each of the five MNOs has
SMP in the market for termination of voice calls on its network.145 On this basis,
Ofcom determined that charge controls (applying for four years from 1 April
2007146) should be imposed on the supply of mobile call termination by each of the
five MNOs, and those controls should apply without distinction to voice call ter-
mination whether on 2G or 3G networks.
Both BT and H3G appealed Ofcom’s MCT Statement. H3G appealed Ofcom’s de-
termination that H3G has SMP and the price control; while BT appealed the level
of the price control only. In May 2008, the CAT upheld Ofcom’s finding of SMP
for H3G, dismissing the non-price control matters arising in H3G’s appeal.147 That
judgment was appealed to the Court of Appeal, which found in favour of Ofcom
uk/20160702162827/http://stakeholders.ofcom.org.uk/binaries/consultations/mobile_c all_term/statement/
statement.pdf> and Assessment of whether 3G holds a position of SMP in the market for wholesale mobile voice
call termination on its network, March 2007, at <http://webarchive.nationalarchives.gov.uk/20160702162827/
http://stakeholders.ofcom.org.uk/binaries/consultations/h 3gsmp/statement/statement.pdf>.
146
Given that the then existing charge controls were due to expire less than one week after publication of
the Ofcom statement, Ofcom decided to impose new controls from 1 April 2007 but to adjust the level of the
year-one (1 April 2007 to 31 March 2008) controls by weighting them as though they applied for only 10 of the
12 months of the year one control and as though for two of the 12 months the present average charges applied.
147
See Hutchison 3G (UK) Limited v Office of Communications (Mobile Call Termination) [2008] CAT 11.
94
94 Part I Fundamentals
and the interveners, BT and T-Mobile, and upheld the CAT’s decision rejecting
H3G’s challenge.148
On 18 March 2008, the CAT referred various ‘price control matters’ to the CC. In
January 2009, the CC issued its determination on mobile call termination charges.
This resulted in the charges being reduced even further than Ofcom’s original
2007 statement. Table 2.5 shows the CC’s determination of charges in real 2006/07
prices (with the original charges set in the 2007 MCT Statement shown in brackets).
Vodafone & O2 5.2 (5.5) 4.7 (5.4) 4.4 (5.2) 4.0 (5.1)
T-Mobile & Orange 5.7 (6.0) 5.0 (5.7) 4.5 (5.4) 4.0 (5.1)
H3G 8.9 (8.9) 6.8 (7.5) 5.5 (6.7) 4.3 (5.9)
148
See Hutchison 3G (UK) Limited v Office of Communications [2009] EWCA Civ 683.
149
‘On-net’ refers to traffic within the same mobile network ie between customers on the H3G network.
‘Off-net’ is when traffic crosses to another network eg a call from the H3G network to a Vodafone customer.
95
large amount of off-net traffic and therefore a net outflow of traffic from the smaller
network. Thus if MTRs are symmetric, this disadvantages the smaller operator.
H3G therefore argued that a move to Bill and Keep would level the playing field,
but suggested that in the transition to Bill and Keep smaller operators should re-
ceive higher MTRs to counter the impact of the outflows.
The final Commission Recommendation on termination suggested a pure LRIC
cost methodology.150 In essence, the Commission suggested recovering elements
of common costs not from termination, but from the competitive retail side of
the mobile market. This approach would reduce the headline rate of termination
charges, particularly MTRs, then currently in place across the EU, potentially by
a significant amount. It was recognized however that such a shift could affect mo-
bile retail prices, as MNOs would seek to recover costs from their retail customers
that were no longer recoverable from call termination charges.
In the context of all this debate and in anticipation of the mobile call termin-
ation charges expiring in March 2011, Ofcom published a consultation document
in May 2009,151 which considered the different approaches that may be taken to-
wards setting MTRs. It acknowledged that in arriving at a decision on the best
approach, it required to take utmost account of the EC Recommendation but it
must do so in the context of considering the effects on all market participants. For
this reason, it considered a much broader set of options than that set out by the
Commission.
Based on the responses to the May 2009 consultation and a second consultation
in April 2010,152 Ofcom, in April 2011153 proposed the use of the pure LRIC method
to set regulated rates. It proposed that after a single-year transitional period, a
symmetric rate would apply across the four mobile networks.
In 2014, Ofcom launched its consultation for MTRs for 2015/18.154 Based on re-
sponses to this consultation, in March 2015155 Ofcom concluded that it would set
150
Commission Recommendation on regulatory treatment of fixed and mobile termination rates in the EU
C(2009) 3359 final, at <http://ec.europa.eu/smart-regulation/i mpact/ia_c arried_out/docs/ia_2009/c _2009_
3359_en.pdf>.
151
Ofcom, ‘Wholesale mobile voice call termination: Preliminary consultation on future regulation’, 2009,
at <http://webarchive.nationalarchives.gov.uk/20160702162827/http://stakeholders.ofcom.org.uk/binaries/
consultations/mobilecallterm/summary/mobile_c all_term.pdf>.
152
Ofcom, ‘Wholesale mobile call termination review (second consultation)’, 2010, at <http://webarchive.
nationalarchives.gov.uk/20160702162827/http://stakeholders.ofcom.org.uk/consultations/w mctr/>.
153
Ofcom, ‘Mobile termination review statement’, 2011, at <http://webarchive.nationalarchives.gov.uk/
20160702162827/http://stakeholders.ofcom.org.uk/consultations/mtr/statement>.
154
Ofcom, ‘Mobile call termination market review 2015–18’, 2014, at <http://stakeholders.ofcom.org.uk/
binaries/consultations/mobile-c all-termination-14/summary/MCT_Consultation.pdf>.
155
Ofcom, ‘Mobile call termination market review 2015–18: Statement on the markets, market power de-
terminations and remedies’, 2015, at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 029/76385/mct_
final_statement.pdf>.
96
96 Part I Fundamentals
a single MTR cap for all mobile networks with SMP and to set the MTRs with ref-
erence to the LRIC in each and every year of the cap. This represented a change
from the previous market review where the charge control only applied to the four
largest mobile networks and smaller mobile networks were subject to an obliga-
tion to provide network access on fair and reasonable (F&R) terms and conditions,
including charges. Ofcom’s reasoning for this change was that imposing a charge
control on all mobile networks with SMP will be more effective than the F&R ap-
proach in remedying the harm caused by MTRs set above the efficient cost bench-
mark. In March 2018, Ofcom set out its decision on the regulation of the wholesale
MCT market for the period 2018 to 2021.156 As in the previous review, Ofcom im-
posed a single maximum cap on MTRs based on LRIC for all mobile providers
with SMP.
156
Ofcom, ‘Mobile call termination market review 2018–21: Final Statement’, March 2018, at <https://
www.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 021/112458/Final-Statement-Mobile-Call-Termination-Market-
Review-2018-2021.pdf>.
97
98 Part I Fundamentals
example, passive remedies such as access to ducts or dark fibre may become more
important. In addition, as the UK Government prepares to leave the European
Union, a key consideration will be whether the European legal frameworks
governing communications in the UK will need replicated or replaced in UK legis-
lation. All these issues will play out over the next few years. The forces of competi-
tion and technological developments, alongside the emergence of new innovative
economic regulatory policies mean that the future of communications continues
to be very exciting!
9
Part II
1
See generally Standage, T, The Victorian Internet (London: Phoenix, 1999).
012
Between 1835 and 1844 the electric telegraph was introduced,2 and elec-
tric telegraphs made it possible to send letter-like objects in non-material form
and this technology made it possible to send ‘telegrams’ on a commercial and
affordable basis.
The railways had a continuing demand for telegraphs and their expanding na-
tional networks gave them the physical wayleaves over which to convey messages
for others, as well as obligations to allow others to enter onto their land and build
telegraph lines.3 Private parties that build such systems were also required to
provide access to their telegraph services to ‘all persons alike, without Favour or
Preference’, an early version of net neutrality.4
The years of telecommunications had begun.
• The Acts were founded on the principle that telegrams were letters over which
the GPO had a monopoly.
2
Wheatstone and Cooke patented their invention in 1837 based on electromagnetic impulses travelling
over wires.
3
The Railway Regulation Act 1844, s XII. 4
Ibid, at s XIII. See further Chapter 15.
5
As an institution, the ‘General Post Office’ was established by Oliver Cromwell in 1657.
013
• Companies and individuals were allowed to run their own telegraph systems
on their own land for their own internal purposes but not to provide services to
others.
• There were arrangements for licensing companies and individuals to run their
own telegraph systems.
• There were rules about the conduct of telegraphs including, for example, the
confidentiality of what was conveyed, interference with telegraphs.
• There were provisions about the construction and installation of telegraphs (es-
pecially telegraph poles), compulsory acquisition of land, arrangement for dig-
ging up streets, provision for running wires over private land, rights to cross
railways and canals, etc.
6
See generally BT Archives, available at <http://w ww.btplc.com/Thegroup/BTsHistory/BTgrouparchives/
index.htm>.
7
AG v Edison Telephone Company of London (1880) 6 QBD 244.
014
to operate telephone systems under certain conditions, for example they were re-
stricted to the areas in which they were already operating. The licences required
the payment of a royalty to the PMG and gave the PMG an option to purchase the
telephone undertaking at the end of a specified term.
This policy was further relaxed in 1882 when the PMG decided to grant licences
to operate telephone systems to all responsible persons who applied for them,
even where a Post Office system was already established. This was a reversal of the
previous policy on the ground that ‘it would not be in the interest of the public to
create a monopoly in relation to the supply of telephonic communication’.8
There was a further change in the position of the PMG when he realized that
the developing telephone systems were seriously affecting the revenue of the tele-
graph service. There were also complaints about the quality of the NTC’s service
and the accumulation of its overhead wires in towns. In 1892, the government de-
cided that the trunk line system should be owned by the State and in 1896 the PMG
took over the trunk lines of the NTC. The NTC was restricted to providing service
in local exchange areas and it was decided that no further national licences would
be issued. Intercommunications were established between exchange customers of
the Post Office in one area and those of the NTC in another.
In 1905, the PMG and the NTC agreed conditions for the transfer of the NTC’s
undertakings to the Post Office. From this time the Post Office and the NTC began
to work towards the unification of their two systems. Intercommunication was
possible between subscribers to both systems in the same local area throughout
most of the country. On 1 January 1912, the PMG took over the system of the NTC
and from this date the Post Office became the monopoly supplier of telephone
services throughout most of Britain, with a few exceptions. The first statutory rec-
ognition of telephones as a distinct business from the telegraph was the Telephone
Act 1951, which granted the PMG the power to make regulations governing ‘the
terms and conditions on which the use of means of telephonic communication
provided by him (whether through the medium of the public telephone system
under his control or otherwise) will be permitted and for the general conduct of
telephonic business carried on under his control’.
8
Henry Fawcett, Postmaster-G eneral, HC Deb 17 July 1882 vol 272 cc711-2 .
015
money for the establishment of local telephone systems under licence from the
PMG. The PMG maintained the right to purchase any local authority system after
a period of years. Thirteen authorities took out licences but only six set up tele-
phone systems. One of these authorities was Hull Corporation (the forerunner of
Hull City Council), which was granted its licence on 8 August 1902. This licence
was conditional upon it embracing the same exchange area as that covered by
the NTC.
For a number of reasons, 9 all of these licences with the exception of the one
granted to the Hull Corporation lapsed within a few years. Hull Corporation’s
licence was due to expire in 1911 together with that granted to the NTC. By this
time the Post Office network was so small within the Hull area that the Post
Office had limited local commercial interest within the area and was content
to grant a new licence to Hull Corporation on the condition that it acquired
the plant and equipment used in the NTC network. This occurred and a new
licence was finally issued in 1917 expiring on 31 December 1932. A succes-
sion of licences followed. The only substantial change was the replacement of
Hull Corporation by Hull City Council in 1974. In 1984, the City Council was
granted a licence under the 1984 Act. The licence was transferred to Kingston
Communications (Hull) PLC10 in 1987, a company wholly owned by the City
Council.11
This rather unique situation in Hull has been very important to telecommunica-
tions regulation in a number of ways, including:
• it showed that a small operation that did not enjoy economies of scale could pro-
vide an efficient and cost-effective service; and
• Kingston Communications had a working interconnect which enabled messages
sent via one operator’s system (ie Hull’s) to be conveyed by another operator’s
system (ie BT’s) and this provided the critical precedent for the Mercury/BT
interconnect.12
9
eg some local telephone users discovered that competition sometimes meant having to rent two
telephones.
10
The name changed to KCOM Group PLC in 2007.
11
The Group was partially floated in 1999 with the City Council retaining a 44.9% stake. The City Council
sold its remaining shares in 2007 and the company changed its name to KCOM.
12
See further Chapter 8.
016
The 1969 Act was innovative in attempting to define what was meant by the
running of a telecommunication system. This definition was replicated in section
017
4(1) of the Telecommunications Act 1984, while the concept of running a telecom-
munication system remained the foundation of the regulatory system until the
Communications Act 2003.
The next logical step would have been for the government to take over the li-
censing function. However the government did not do this; instead it surrendered
all licensing powers to the new Post Office. The Post Office was given powers to
license other telecommunication systems.
The labour-intensive, low-tech, and traditional postal business had little in
common with the high-tech, capital-intensive, and dynamically expanding tele-
coms business. The common ground was a shared history, common pool of em-
ployees, and vested interests in protecting its position.
In 1977, the Carter Committee report recommended a further separation of
the postal and telecommunications services of the Post Office and their reloca-
tion under two individual corporations.13 This led to the renaming of the Post
Office Telecommunications as British Telecom in 1980 and to the introduction of
the British Telecommunications Act 1981. While postal and telecommunications
services remain separate, the Post Office now offers its own range of communica-
tion services,14 while the regulator for both is Ofcom, having assumed responsi-
bility for the postal sector in 2011.15
<http://w ww.postoffice.co.uk/broadband-phone>.
14 15
Postal Services Act 2011, s 28(1).
018
on their behalf. Paging was also authorized. All the licensed operators were on
a small scale and the GPO ran the main national radio and telephone networks.
Some of these radio licences permitted connections into the public switched
telephone network, but these were normally indirect connections through human
operators to private ‘call handling’ services. By 1979, the GPO woke up to the fact
that these small operators were threatening its own operations and started to de-
velop its own telephone systems. The GPO systems began to enjoy the economies
of scale by using facilities installed for ‘telegraphs’ and as a result the competition
struggled.
During the last two decades of the twentieth century, the global telecommuni-
cations market experienced a period of unprecedented growth and the telecom-
munications industry changed almost beyond recognition. Some of the main
developments during this period are highlighted below.
The government sold all of its shares in Cable & Wireless in three stages between 1981 and 1999.
16
019
connecting directly only a limited number of large business users and attracting
smaller users to its network for long-d istance and international calls only. It
showed little motivation to invest in a national network and by 1991 Mercury had
secured only three per cent of the market.
Communications Liberalisation in the UK, March 2001, Department of Trade and Industry.
17
A small number of franchises remained independent—i ncluding Wight Cable covering the Isle of Wight
18
not launch their analogue services until 1985, creating another duopoly. The
first mobile telephone call in the UK was made on 1 January 1985. Towards the
end of 2006, mobile was the most prevalent telecoms technology with the pro-
portion of households with access to a mobile phone overtaking the proportion
of households with a fixed line. 21
There were initially restrictions on the retailing of mobile airtime by the mobile
operators directly to the public. This led to a growth in the importance of High
Street retailers, including Dixons, Currys, and the Carphone Warehouse. These
dealers were primarily sellers of mobile hardware and did not themselves offer
airtime contracts to their customers. However, because of the expectations of cus-
tomers buying a mobile phone to complete all the necessary contractual arrange-
ments at the same time, it became the dealer who ‘arranged’ the airtime contract.
Mobile operators became increasingly reliant on the High Street chains to market
their services.
In 1993, two further licences were granted to Orange and Mercury One-2-One22
allowing these companies to operate 2nd Generation (2G) or GSM (global system
for mobile communications) networks. To allow the mobile operators to compete
on an equal basis Vodafone and Cellnet were granted reissued licences which were
modified in order that they could provide their services via 2G networks.
In 2000, the government held an auction for licences to operate 3rd Generation
(3G) spectrum. 3G networks supported higher speed call services and mobile data
services. The auction process resulted in five 3G licences being granted in 2000,
with one to a new entrant TIW UMTS (UK) Ltd (now operating as 3). The other
licences were issued to Vodafone, One2One, Orange, and BT. The process raised
over £22 billion. Those who bid successfully were required to provide a 3G network
that would cover at least 80 per cent of the UK population by 2007.
In addition to the mobile network operators a number of established retail
brands, such as Tesco and Virgin Media, entered the market as MVNOs (mo-
bile virtual network operators). There are currently over seventy such MVNOs in
the UK.
The next generation of 4G broadband cellular network services were first
launched in the UK in October 2012 by Everything Everywhere, but are now en-
abled on two-t hirds of mobile subscriptions.23 The next iteration, 5G services are
expected to be launched by 2020.
21
Ofcom consultation, ‘Mostly mobile’, 8 July 2009, at 2.12.
22
One-2-One was purchased by Deutsche Telekom in 1999 and became T-Mobile, which merged with the
Orange UK business in 2010, to form Everything Everywhere, which was sold to BT in 2016.
23
Ofcom, ‘Communications Market Report 2017’, 3 August 2017.
1
24
Ofcom Statement, ‘Review of the Wholesale Broadband Access Markets’, 26 June 2014, at 1.20.
25
The cable operator Virgin Media is not deemed to be nationally dominant and so the regulator cannot
force them to open up their network to other service providers in the same way that BT has had to offer whole-
sale products to players such as TalkTalk and Sky.
26
See further Chapter 9, at Section 9.2.1.
27
Terminal equipment is customer premises apparatus. For regulatory purposes the boundary has been
drawn at the socket, or test jack frame, where a connection can be made between the chain of apparatus on
a customer’s premises and the chain of apparatus back to the telephone exchange and beyond the telephone
networks.
12
The British Telecommunications Act 1981 separated the Post Office’s func-
tions of telecommunications and postal carrier and BT was created.28 The 1981
Act granted BT an exclusive privilege of ‘running telecommunication systems’
(s 12), but also recognized certain classes of act that did not infringe the privilege,
such as internal business systems (s 13). The government became the licensing au-
thority for telecommunications operators (s 15), but unfortunately the Act did not
include a power to limit BT’s exclusive privilege.
Mercury was granted a licence as the first competitor to BT. This licence gave
Mercury the right to provide every form of digital telecommunications service,
including leased circuits, switched services to business and domestic premises, and
the full range of international services. Mercury was not, however, allowed to lease
elements of BT’s infrastructure (except for interconnection for call termination).
This was in line with the government’s policy to encourage infrastructure-based
competition. In its 1983 Duopoly Statement29 the government made it clear that they
did not intend to license operators other than BT or Mercury to provide the basic
telecommunications service of converting messages over fixed links, whether cable,
radio, or satellite, both domestically or internationally, for seven years. In return for
its protection, Mercury undertook some mild obligations to expand its network.
It soon became clear that the 1981 Act was not a suitable vehicle to promote com-
petition. It did not give Mercury powers to dig up the streets or to erect telegraph
poles. It included licensing provisions which were seriously flawed, for example BT
had to be consulted about all licences and could therefore find out about competi-
tors’ plans. It had no provisions to force BT to connect Mercury’s network and BT
initially refused to agree to do this, proposing that Mercury should build an overlay
network with every customer having two phone points and phone lines, one BT and
one Mercury. When an agreement to interconnect was finally reached, in November
1982, the Post Office Engineering Union then ordered its members not carry out any
such works, in order to preserve jobs and oppose BT’s privatization.30 Overall the
issues had not been thought through from the perspective of a competitor.
28
The formal separation occurred on 1 October 1981.
29
Government Statement of 17 November 1983 by Kenneth Baker MP, Minister for Information Technology
to the Standing Committee on the Telecommunications Bill.
30
See Mercury Communications Ltd v Scott-Garner & ors [1983] 3 WLR 914.
13
31
The November 1984 share offer was oversubscribed by 3.2 times with shares being issued to applicants
on a pro rata basis.
32
A second share issue took place on 21 November 1991, reducing the government’s stake to 21.8%. A fur-
ther issue followed in July 1993, with the government selling off virtually all of its remaining shares. In July
1997 the government relinquished its ‘golden share’, which allowed it to block a take-over of the company and
to appoint two non-executive directors to the Board.
14
The 1981 Act had not granted the rights that operators required to construct in-
frastructure and therefore prevented effective competition in this respect. The
Telecommunications Code (generally referred to as ‘Code Powers’) contained in
the Telecommunications Act 1984, section 10 and Schedule 2, allowed operators
to install apparatus under or over the street, dig up the street, and open sewers34
among other works. Where apparatus was constructed to the height of three
metres or more a landowner could object to the installation where it affected their
enjoyment of the land. This matter could be dealt with in a number of ways: com-
pensation could be paid to the landowner, the apparatus could be required to be
modified, or a court could declare that the landowner’s agreement be dispensed
with. It was imperative for cable companies, as well as public telecommunications
operators (PTOs), to be granted Code Powers so that they could operate and com-
pete in the marketplace.
The 1984 Act also introduced an independent regulator known as the DGT. The
DGT was appointed by the Secretary of State for Trade and Industry and was an
unelected, independent position. He was appointed for a fixed but renewable term
of office and could be removed from office only as provided for in his contract. The
DGT was head of a non-m inisterial government department, known as the Office
of Telecommunications (Oftel). Oftel was subject to treasury control so far as ex-
penditure was concerned and accountable to Parliament like any other govern-
ment department.
The 1984 Act split regulatory competence between the Secretary of State and the
DGT. It imposed primary duties on the Secretary of State and the DGT to exercise
their functions with a view to ensuring:
Schedule 2 para 9. Schedule 2 of the 1984 Act has been incorporated into the Communications Act 2003,
34
under s 106, and renamed the ‘electronic communications code’. See further Chapter 6.
15
• that the persons responsible for providing telecommunications services are able
to finance the provision of those services.35
Subject to these overriding duties, the Secretary of State and the DGT
were required to exercise their functions in the manner they considered best
calculated:
There were also other duties designed to encourage investment, promote inter-
national transit services, and enable providers of telecommunication services and
producers of the telecommunication apparatus to compete overseas.37
In addition to these functions the DGT also had a duty to consider com-
plaints under section 49 and had the power to make competition references to
the MMC. 38
The Secretary of State was responsible for issuing licences.39 In practical terms
this meant the licensing process was handled by an executive agency of the
Department of Trade and Industry,40 the Radiocommunications Agency.
The DGT was responsible for enforcement of the licences. The DGT was also
obliged to enforce the observance of conditions included in licences granted to
telecommunications operators, by making orders under sections 16 to 18 of the
1984 Act. Where the DGT was satisfied that a licence holder was, had or was likely
to contravene the conditions of their licence, he was obliged to make such provi-
sion as was requisite to secure compliance with the condition. In the early days,
these sections were rarely used as the threat of enforcement seemed to have the
desired effect. More orders were made in the 1990s. Any person who suffered a loss
or damage as a result of a breach of a final or confirmed provisional order could
bring an action for damages against the licensee.41 Failure to comply with a final
order could also result in the revocation of a licence.42
The DGT also had the power to modify the conditions included in a tele-
communication licence. The Act set out two mechanisms for implementing
changes: neither of which was particularly efficient. The first was through a
voluntary agreement of licensees under section 12. This meant that in the case
of a class licence, all licensees had to agree to a proposed modification.43 The
second mechanism was through a compulsory modification against the wishes
of the licensee under section 15 following an MMC investigation. The role of
the MMC was therefore to act as an appeal body. For example mobile termin-
ation rates were investigated by the MMC in 1998.44 The difficulties in pursuing
licence modifications under either of these mechanisms led to a revision of
section 12.45
Another important function was to give directions and determinations in rela-
tion to matters reserved for the DGT’s decision under licences granted to telecom-
munication operators. This power, derived under subsections 7(5) and (6) of the
1984 Act, was used extensively in relation to interconnect.
The DGT was given a large amount of discretion in making decisions and had
a high level of autonomy, although his decisions were open to judicial review by
the High Court. Judicial review does not however allow the court to carry out a
review of the merits of the decision itself. In practice it was very difficult to chal-
lenge any decision where the DGT could argue that he had exercised a judgment
under section 3 ‘in a manner which he considers is best calculated’ to undertake
his duties. There were some challenges to his decisions but the courts generally al-
lowed the DGT wide discretion, and so successful claims against him were rare.46
There was a general feeling that the courts in any event would be unwilling to
interfere with the decision of a sector specific regulator.47
The 1984 Act continued in force for almost 20 years until it was replaced by the
Communications Act 2003. Significant changes did occur in the intervening period,
primarily reflecting evolving European Union law.48 Successive governments
41
Section 18(5). 42
Section 18(6).
43
The cumbersome nature of this process meant that in practice class licences were rarely if ever modified.
Instead they tended to be revoked and reissued in a modified form by the Secretary of State.
44
See <http://webarchive.nationalarchives.gov.uk/20111202172129/http://w ww.competition-commission.
org. uk/rep_pub/reports/1999/422bt.htm#full>.
45
Electronic Communications Act 2000, ss 11–12.
46
R v Director General of Telecommunications, ex p Cellcom Ltd and others [1999] ECC 314.
47
Graham, G, Regulating Public Utilities; A constitutional approach (Oxford: Hart, 2000).
48
See further Chapter 4.
17
chose to transpose the various EU Directives into UK law through statutory in-
struments made under the European Communities Act 1972, section 2(2), rather
than through amendments to the 1984 Act. This secondary regulation imposed a
broad range of new functions and duties upon the Secretary of State and DGT in
areas such as interconnection, licensing, and voice telephony.49
• to end the duopoly policy. This meant that anyone who could show a need for a
licence and who had the necessary financial resources could become a public
telecommunications operator;
• to introduce equal access as soon as possible;
• to strengthen the arrangements for interconnection in operators’ licences
including provisions to extend the DGT’s power of direction to cover all aspects
49
eg Telecommunications (Licensing) Regulations 1997, SI 1997/2930.
50
See OECD Reviews of Regulatory Reform, Regulatory Reform in the UK. 51
Ibid.
52
Cm1303. 53
Cm1461, March 1991.
18
54
In 1996 the government licensed an initial batch of 44 companies to provide international telecommu-
nications services on any route they choose over their own facilities. At the same time the restrictions which
limited international simple resale to certain routes were removed.
55
See further Chapter 2.
56
Ofcom, Strategic Review of Telecommunications, Phase 1 consultation document, Research Annexes,
Annex G, Review of Regulatory Policy in the telecoms sector.
19
3.3.6 Convergence
In July 1998, the DTI published a Green Paper, Regulating communications: ap-
proaching convergence in the Information Age.57 The Green Paper recognized that
digital technologies were already changing the way that services were delivered,
blurring the boundaries between types of service operation and means of delivery
and eroding the technological distinctions.
Broadcasting, both content and delivery, had been regulated by the ITC58 and
the BBC. Spectrum was regulated by the Radiocommunications Agency. Bodies
like the Broadcasting Standards Commission59 and the Radio Authority 60 also had
regulatory powers and obligations.
Following a consultation period, in 1999 the government published its report
suggesting a number of options including the possibility of creating a radically new
regulatory structure to avoid barriers to competitiveness. However, the government
decided to take the evolutionary approach and let the existing structures stay for the
present but required them to work closer together.
It was following EU moves for reform, which eventually resulted in the
Framework Directive (2002/21/EC), and an appreciation that convergence was
gathering pace, that the government introduced a White Paper, A New Future for
Communications61 (the Communications White Paper) to create a new regulatory
structure. The proposal was that a new regulator, the Office of Communications
(Ofcom), would be created.62
57
Cm 4022. The Green Paper was a response to the EC’s own Green Paper on Convergence of the
Telecommunications, Media and Information Technology Sectors, and the implications for regulation to-
wards an information society approach (EC COM (97) 623, 3 December 1997.)
58
The Independent Television Commission (ITC) was the statutory body which licensed and regulated in-
dependent television services in the UK. Under the Broadcasting Acts 1990 and 1996, their responsibilities
included setting and maintaining the standards for programmes, economic regulation, public service obliga-
tions, research, TV advertising regulation, and technical quality.
59
The Broadcasting Standards Commission had statutory responsibilities for standards and fairness in
broadcasting. It had three main tasks: to produce codes of conduct relating to standards and fairness; to con-
sider and adjudicate on complaints; and to monitor, research, and report on standards and fairness in broad-
casting (Broadcasting Act 1996, Part V).
60
The Radio Authority was the statutory body responsible for regulating independent radio broadcasting
in the UK (ie non-BBC radio services). Their responsibilities included frequency planning, the awarding of
licences, the regulation of programming and radio advertising, and the supervision of the radio ownership
system.
61
Cm 5010 published on 12 December 2000.
62
Office of Communications Act 2002. The Act enabled the government to formally establish Ofcom before
the Communications Act came into force and placed the existing regulators under a duty to assist Ofcom to
prepare (s 4).
210
63
See further Chapter 4.
64
A consolidated version of the general conditions, as at 28 May 2015, is available at <http://w ww.ofcom.
org.uk>.
65
And subsequently the Postal Services Commission and the BBC Trust.
66
The Office of Communications Act 2002, s 1. 67
Section 3(1).
21
Ofcom has a duty to act in accordance with the six Community requirements set
out in the Framework Directive.69 These are:
• to promote competition;
• to ensure that Ofcom’s activities contribute to the development of the European
internal market;
• to promote the interests of all persons who are citizens of the European Union;
• to take account of the desirability of carrying out their functions in a manner
which, so far as practicable, does not favour one form of network, service, or as-
sociated facility, or one means of providing or making available such a network,
service, or facility over another;
• to encourage the provision of network access and service interoperability; and
• to encourage compliance with international standards to the extent neces-
sary to facilitate service interoperability, and to secure a freedom of choice for
customers.
Details of the 2003 Act are discussed elsewhere in this book, but Ofcom has
been given improved duties to make its processes more transparent and efficient
and to encourage deregulation as the sector becomes more competitive. Since
the creation of Ofcom the breadth of its responsibilities has increased, such a
duty to report to the Secretary of State for Culture, Media and Sport on the state
of the UK’s communications infrastructure.70 Other additional duties imposed on
Ofcom include: to plan and manage spectrum for the London 2012 Olympics and
Paralympic Games, to reduce the scale of illegal peer to peer file sharing, and to
increase digital participation.71 The Secretary of State has the power to give Ofcom
Section 3(2).
68 69
Section 4.
Sections 134A–134B, inserted by the Digital Economy Act 2010, s 1, and amended by the Digital Economy
70
Act 2017, s 82. For the Ofcom reports, see <https://w ww.ofcom.org.uk/research-a nd-data/multi-sector-re-
search/i nfrastructure-research>.
71
National Audit Office, ‘The effectiveness of converged regulation’, November 2010.
21
general and specific directions;72 and can also set out ‘strategic priorities’, which
Ofcom has a duty to have regard to.73
72
Communications Act, s 5(2). For example, the Wireless Telegraphy Act 2006 (Direction to Ofcom) Order
2010, SI 2010/3024.
73
Communications Act 2003, s 2B, inserted by the Digital Economy Act 2017.
74
Communications Act, s 185.
75
Ofcom, ‘Enforcement guidelines for Competition Act investigations’, 28 June 2017. See further Chapter 10.
76
Communications Act, s 188(5), except in exceptional circumstances.
77
Ofcom, ‘Dispute resolution guidelines’, 7 June 2011. 78
<http://w ww.offta.org.uk/i ndex.htm>.
79
Communications Act, s 193.
213
The 2003 Act provides that an appeal to the CAT can be on the grounds that
the decision was based on an error of fact, was wrong in law, or both, or against
the exercise of discretion by Ofcom, the government, the CMA or another person
(s 192(6)). Appeals from decisions of the CAT are on points of law only and are to
the Court of Appeal (s 196(2)). Such appeals are only allowed with the permission
of the CAT or Court of Appeal.80
A widely held criticism of the previous appeals system was that the courts were
less well equipped than a specialist regulatory body to understand complex tech-
nical and economic issues and consequently were often reluctant to overturn the
decision of an industry-specific regulator. Under the old appeals system no deci-
sion taken by the DGT was ever successfully challenged in the UK courts. In con-
trast the CAT is generally seen to be a more effective appeals tribunal than the
courts. The first case appealed to the CAT in the telecommunications sector (the
Freeserve 81 case) resulted in part of the DGT’s decision being struck down. Since
then there have been numerous cases referred to the CAT.82
Until 2017, the CAT was required to decide such appeals ‘on the merits’, which
was considered to be in compliance with EU requirements.83 However, a merits-
based review is a complex and time-consuming process, as well as encouraging
operators to appeal against Ofcom decisions as a means of delaying regulatory
decision-making. To try and reduce the burden for the CAT, as well as strengthen
the position of Ofcom, the Communications Act was amended to alter the review
standard to that of judicial review.84 Whether this change will have a significant
impact may depend on the attitude of the CAT to how ‘flexible’ the judicial review
standard should be viewed, since a broad interpretation could end up being not so
dissimilar from that of a ‘merits’ review.
Ibid, s 196(4).
80 81
[2003] CAT 5.
For copies of the judgments see <http://w ww.catribunal.org.uk/2 38/a ll/2/Judgments.html>.
82
83
See the Explanatory Notes to the Digital Economy Act 2017, at paras 45–49. The CAT had previously dis-
tinguished a ‘merits’ review from a de novo hearing, since only pleaded errors of fact or law are examined (BT
v Ofcom [2010] CAT 17, at para 76).
84
Communications Act, s 194A.
124
Some of the key issues from the results of Phase 186 of the review were:
Ofcom concluded that regulation had failed to overcome the problems of enduring
bottlenecks combined with lack of access to those parts of the network. It acknow-
ledged that those who relied on BT to provide access have experienced 20 years
of slow product development; inferior quality wholesale products; poor transac-
tional processes; and a general lack of transparency.
Ofcom’s proposal was that it should focus regulation to deliver equality of ac-
cess beyond the levels of infrastructure where competition will be effective and
sustainable. Ofcom’s preferred approach to deliver this was what they referred to
as ‘equality of access’. There were two elements to this:
• Equivalence at the product level. This meant that in parts of the network where
BT had SMP and which are enduring bottlenecks, BT must offer the same or
similar wholesale products to wholesale customers as it offers to itself, at the
same prices and using the same or similar transactional processes; and
85
See <http://w ww.Ofcom.org.uk/static/telecoms_ review/i ndex.htm> for the strategic review consult-
ation documents and statements.
86
The review did cover mobile but the results of Phase 1 showed that in terms of competitive market struc-
tures mobile was strong with five competitive operators and more virtual network operators. Ofcom felt that
in all respects the mobile sector displayed the features of a competitive market.
215
Ofcom wanted a new regulatory contract with BT: by which it meant a settle-
ment with the industry to try to obtain real equality of access. It wanted the rules
setting out such an approach to be legally enforceable. It found the means to
achieve this under the Enterprise Act 2002. The Enterprise Act gives regulators,
including Ofcom, the power to make a referral of a market to the CMA where there
are reasonable grounds for suspecting,
that any feature, or combination of features of a market in the UK for goods or
services prevents, restricts or distorts competition in connection with the supply
or acquisition of any goods or services in the UK . . . .87
Any such investigation would be wide ranging and the CMA would have the power
to order structural separation of BT’s wholesale network operations and its retail
service provision. The Enterprise Act also provides that instead of making a market
investigation reference to the CMA, a regulator may accept undertakings to take
such action as it considers appropriate.88 These undertakings must be for the pur-
pose of remedying, mitigating, or preventing any adverse effect on the competi-
tion concerned.
90
Numerous amendments and exemptions to the undertakings have been granted since 2005. See <https://
www.ofcom.org.uk/phones-telecoms-a nd-i nternet/i nformation-for-i ndustry/policy/bt-u ndertakings>.
91
As from 31 July 2006, all retail price controls on line rental and call charges were removed: Ofcom, Retail
Price Controls—explanatory statement, 19 July 2006.
92
Section 5 of the Undertakings.
216
Under the concept of EOI, BT’s wholesale customers should be able to use exactly
the same set of regulated products, at the same prices and using the same systems
and transactional processes as BT’s own retail activities.
The Undertakings also subjected BT Wholesale to a number of obligations.
For example, it was required to establish two separate internal divisions for the
product management of SMP products not supplied by Openreach and non-SMP
products of significance to competing providers (such as wholesale calls and
IPStream). BT was required to implement organizational separation between its
downstream and upstream (other than Openreach) divisions. In particular, it had
to maintain a strong organizational separation of people, commercial informa-
tion, and sales functions.
The Oftel approach had been to try to ensure that wholesale products specific-
ally designed by BT under regulatory pressure were as close to being fit for pur-
pose as possible. Differences in product and processes were tolerated as long as no
material difference in overall outcome. There were however a number of problems
with this approach, including the fact that BT had little incentive to produce prod-
ucts for its competitors which it had no interest in using itself.
In theory the advantage of the Ofcom approach is that it provides an immediate
incentive on BT to remedy any deficiencies, as it is required to offer exactly the
same products to its wholesale customers as to its own retail division.
93
Undertakings, section 5.46. There are some exceptions to this section.
94
Undertakings, section 2.1.
217
In terms of BT, Ofcom concluded that despite the EAB and related processes, it
continued to have both the incentive and ability to favour its own business to the
95
Undertakings, section 10. See <http://w ww.btplc.com/Thegroup/Theboard/Boardcommittees/E quality
ofAccessBoard/E qualityofAccessBoard.htm>.
96
EAB Annual Report, 2009.
97
Ofcom, ‘Making communications work for everyone: Initial conclusions from the Strategic Review of
Digital Communications’, 25 February 2016.
218
98
Ofcom, ‘Strengthening Openreach’s strategic and operational independence’, 26 July 2016.
99
Ofcom, ‘Delivering a more independent Openreach’, 17 March 2017.
100
From 2000–02, both Cellnet and Vodafone were designated as having ‘market influence’, which was de-
fined in Condition 56 of the standard mobile PTO licence as ‘the ability to raise prices above the competitive
level for a non-t ransitory period without losing sales to such a degree as to make this unprofitable.’ Designation
triggered obligations to provide airtime to qualifying service providers and not to show undue preference or
discrimination in the provision of services. Cellnet and Vodafone were also designated as having ‘SMP’ for the
purposes of the Interconnection Directive.
101
The relationship between the networks and their service providers came increasingly under strain from
the beginning of the 1990s for mainly commercial reasons. The difficulties came to a head in 1992 when an
independent service provider, Talkland International, made a complaint alleging unfair cross subsidy of the
service providers owned by or closely linked to operators of the mobile networks. On 17 May 17 1994 the DGT
219
As already mentioned, in 1993, two further licences were granted to Orange and
Mercury One-2-One. These licences did not include this prohibition on the direct
retailing of airtime to the public, although they were obliged to sell wholesale
to service providers on request.102 The same freedom was extended to Vodafone
and Cellnet when they received new licences in December 1993 and March 1994
respectively.
produced a statement, ‘Fair Competition in Mobile Service Provision’, setting out his conclusions and the
measures to remedy the situation.
102
This condition was removed from the licences of One-2-One and Orange in April 1997: Oftel, ‘Fair
Trading in the Mobile Telephony Market’.
103
Oftel, ‘Review of the mobile market: Statement’, July 1999.
104
MMC, ‘Cellnet and Vodafone: Reports on references under section 13 of the Telecommunications Act 1984
on the charges made by Cellnet and Vodafone for terminating calls from fixed line networks’ January 1999.
310
one and has been subject to a high degree of scrutiny by Ofcom, the CAT, and
the CMA.
In 2004 charge controls were imposed in respect of mobile voice call termin-
ation charges in respect of Vodafone, O2, Orange, and T-Mobile.105 At this time, the
price control was imposed on termination charges using the 2G spectrum. There
was no price control on termination using 3G spectrum and although Ofcom con-
sidered that all of the mobile operators including 3 had SMP106 in the market for
call termination, the regulatory obligations on 3 did not include charge controls
either for termination on its 3G spectrum or via its roaming arrangements on 2G
spectrum. In 2007 Ofcom decided to extend these charge controls to 3. It said that
the controls should apply to each of the five network operators without distinction
to voice call termination on 2G or 3G networks.107 This decision was subject to a
number of appeals, although the decision to apply charge controls to 3 was ultim-
ately upheld.
The latest charge controls imposed by Ofcom apply from 1 April 2018 until 31
March 2021. Ofcom has decided to:
105
Ofcom statement, ‘Wholesale Mobile Voice Call Termination’, June 2004.
106
3 appealed against the decision that it has SMP. The CAT found that Ofcom had erred in its decision
making process and remitted the case back to Ofcom (Hutchison 3G(UK) Ltd v Ofcom [2005] CAT 39). On 27
March 2007 Ofcom published a reassessment and confirmed its earlier conclusion that 3 has SMP. See Ofcom
mobile call termination statement, March 2007.
107
See March 2007 Ofcom statement.
108
Ofcom, ‘Mobile Call Termination Market Review 2018–2021’, 28 March 2018. In the previous market re-
view, Ofcom identified seventy-t wo MCPs (March 2015) and before that thirty-t wo MCPs (April 2011).
31
Ofcom has introduced a series of measures with the aim of de-regulating spec-
trum. This includes reducing the number of restrictions both in terms of who can
use spectrum and what it can be used for. This new approach is being implemented
primarily through:
• spectrum trading,115
• spectrum liberalization, and
• prompt release of unused spectrum into the market allowing maximum flexi-
bility as to subsequent use.116
112
Professor Martin Cave, ‘Review of Radio Spectrum Management: An independent review for Department
of Trade and Industry and HM Treasury’, March 2002.
113
Ofcom, ‘A Statement on Spectrum Trading—I mplementation in 2004 and beyond’, 6 August 2004.
114
Ofcom, ‘Progress on key spectrum initiatives’, April 2008.
115
See the Wireless Telegraphy (Spectrum Trading) Regulations 2012, SI 2012/2187. See further Chapter 7.
116
See generally <https://w ww.ofcom.org.uk/spectrum/spectrum-management>.
31
3.4 K E Y ISSUE S
In this section, attention is given to a number of key regulatory issues that have
arisen over recent decades, some of which are still proving challenges for Ofcom
and the industry, such as number portability and universal service. We have
also included some of the important regulatory interventions, such as local loop
unbundling, the introduction of wholesale line rental, as well as how Ofcom is ad-
dressing the challenges of new technologies in terms of internet telephony and
Next Generation Networks. Finally, we examine the potential implications of the
UK’s departure from the European Union.
117
Inquiry by the Monopolies and Mergers Commission into Telephone Number Portability: Explanatory
Statement from the Director General of Telecommunications, 27 April 1995.
118
The Numbering Directive 98/61/EC, OJ L 268/37, 3 October 1998, amended the Interconnection Directive
97/33/EC.
314
the right to require the operator to provide number portability. The Numbering
Directive required that subscribers or customers should be able, if they wish, to
keep their telephone numbers when they change the operator providing their fixed
telecommunications services ie telephone services using geographic numbers.
Services using non-geographic numbers, eg mobile, freephone numbers (080),
local and national rate numbers (0845 and 0870), and premium rate numbers (090)
were not covered by the Directive.
Where telephone companies had chosen to market or offer their services they
were required to provide number portability within the normal timescale of pro-
viding a basic service to customers. In such circumstances operators and ser-
vice providers would need to provide number portability within five to eight
working days.
Number portability for mobile services became available in January 1999119 by
conditions inserted into the mobile operators’ licences requiring them to provide
portability to other operators. By 31 March 2008 lead times for porting mobile
numbers were required to have been reduced to two working days.120
In November 2007, Ofcom, still concerned that the current porting practices
were discouraging consumers from switching, issued proposals to reduce the
time taken to port mobile numbers from two days to two hours. This decision was
appealed to the CAT by Vodafone and others and was remitted back to Ofcom.121
In its reconsidered decision Ofcom set out rules requiring providers to give a
Porting Authorization Code (PAC) to consumers who ask for it either immediately
over the phone or within two hours by SMS.122 For fixed numbers, port activation
must take place within one working day from when a subscriber’s new provider
requests activation from the subscriber’s existing provider. This is after the neces-
sary consumer protection measures and any physical line provisioning have been
completed.123
The amended Universal Service Directive also provided that communications
providers must provide compensation to subscribers in the event of delay or
fault with the porting process (Art 30(4)). Communications providers must put in
place schemes which give reasonable compensation to subscribers following any
porting delay or abuse.124 Communications providers are able to design the details
of the schemes themselves, following guidance produced by Ofcom on the oper-
ation of such schemes.
119
Oftel, ‘Number portability in the Mobile Market’, July 1997.
120
Ofcom, ‘Arrangements for porting phone numbers when consumers switch supplier—a review of
General Condition 18’, 17 July 2007.
121
Vodafone Limited v Office of Communications [2008] CAT 22.
122
Ofcom, ‘Changes to the mobile number porting process: Final statement’, 8 July 2010.
123
See further Chapter 9, at Section 9.2.5. 124
General Conditions of Entitlement, at 18.9.
315
Since the introduction of number portability in the UK, both fixed and mobile
networks have used a system referred to as onward routing to route calls to ported
numbers. With onward routing, calls are first routed to the network to which the
customer originally subscribed, known as the donor network. The donor is respon-
sible for routing the call onward to the network to which the consumer now sub-
scribes. There are a number of weaknesses in onward routing. For example, the
dependence on the donor network leaves consumers exposed in the event of net-
work failure by the donor network.125 There are also issues of inefficiency associ-
ated with the additional capacity required for the onward routing of calls.
In November 2007, Ofcom set out its decision that industry should co-operate
to establish a common database (CDB) to allow direct routing of calls to fixed
and mobile ported numbers. As mentioned, that decision was set aside by the
CAT following an appeal by Vodafone and was remitted back to Ofcom. Ofcom
reconsidered its decision and in particular assessed the costs and benefits of
implementing direct routing and its conclusion was that regulatory intervention
would not be appropriate.126
125
In 2001 the insolvency of Atlantic Telecom resulted in Atlantic customers and customers of other com-
munications providers who had ported numbers from Atlantic losing service on those numbers.
126
Ofcom, ‘Statement on Routing Calls to Ported Numbers’, 1 April 2010.
127
Oftel, ‘Universal telecommunication services’, July 1999. In respect of EU policy and law on universal
service, see further Chapter 4, at Section 4.8.
128
Oftel, ‘Designation of BT and Kingston as universal service providers, and the specific universal service
conditions’, 22 July 2003.
316
129
See Oftel, n 127 above, at para 18.
130
Section 65. See also the Electronic Communications (Universal Service) Order 2003, SI 2003/1904 as
amended.
131
Communications Act 2003, s 65(2B), as amended by the Digital Economy Act 2017. Ofcom can be asked
to review and report on the implementation of any requirement concerning broadband connections and
services (s 72A) and must be directed to carry out such a review where the minimum speed is less than 30
Mbps (s 72B).
132
As specified in the Universal Services Directive, Art 4(2).
317
The scope and delivery of the USO has changed only slightly since 1984. The
current USO still falls on BT and KCOM and is implemented by a number of spe-
cific conditions on these two organizations. In addition, a number of General
Conditions impose obligations on all publicly available telephone service pro-
viders. For example, General Condition 8 requires the provision of directory in-
formation and General Condition 18 requires the provision of certain facilities for
end users with disabilities.133
Ofcom looked at questions surrounding universal service as part of its Strategic
Review and it carried out a further review of the USO in 2006134 to ensure that it
continues to meet the needs of customers. In particular it looked at the question
of whether the USO should be extended to include broadband and mobile. It con-
cluded that neither was currently appropriate. It considered that exclusion from
broadband services did not currently result in social exclusion sufficient to war-
rant universal service measures being introduced and that in addition imposing a
USO for broadband would be inappropriate whilst the market for broadband is still
developing. It did feel that it would not be appropriate to rule out the possibility of
imposing a USO for broadband at a future date. It also concluded that increased
fixed mobile convergence means that at some stage mobile may replace fixed as
the primary means of connection and that a universal service obligation defined
in terms of access to voice services could be delivered via a mobile connection, ra-
ther than the imposition of a separate mobile USO.
Another recurring issue relating to universal service relates to funding. BT and
KCOM bears the cost of the USO because Ofcom has determined that the net cost
of provision does not constitute an ‘unfair burden’,135 taking into account factors
such as the positive brand value from being considered the USO provider and the
likelihood that uneconomic customers would remain with BT when they became
‘economic customers’. Were Ofcom to decide that the USO had become an ‘un-
fair burden’ it has the power to put in place alternative methods of provision or
funding.136 This fund could be implemented by a number of means including:
133
See further Chapter 7. 134
Ofcom, ‘Review of Universal Service Obligation’, 14 March 2006.
135
The concept of ‘unfair burden’ originates from the Universal Services Directive, at Art 12(1).
136
Communications Act 2003, s 71.
137
’Strategic Review of Telecommunications—Phase 2’ consultation document.
318
As part of its 2006 review, Ofcom set out indicative estimates for 2003/04 costs
to BT at £56–74 million and £59–64 million for benefits. It felt that these estimates
did not warrant a full scale review and suggested that currently there was not an
unfair burden on BT.
138
See also Chapter 8. 139
This decision was taken in advance of EU Regulation EC 2887/2000.
140
LLU is the process where the incumbent operator makes its local network (the copper cables that run
from customers’ premises to the nearest telephone exchange) available to other companies, who are then able
to upgrade individual lines using DSL technology to offer competing services, such as always-on high speed
internet access. See further Chapter 8.
141
Select Committee on Trade and Industry, Sixth Report, ‘Local Loop Unbundling’, HC 90, 20 March 2001.
142
See also Chapter 8, at Section 8.3.4.4.
319
After a notoriously shaky start the process of LLU is generally seen as having
been a success. Between 2005 and October 2011 the number of unbundled lines in-
creased from 123,000 to 7.77 million and by the end of December 2017 had reached
9.87 million.143
143
See <www.offta.org.uk>. 144
Ofcom, ‘Communications market report’, 3 August 2017.
145
The main obligations which apply to PATS are discussed further in Chapter 6.
146
Ofcom, ‘New voice services—A consultation and interim guidance’, 6 September 2004.
410
sale and at the point of use so that both consumers and users are fully aware of any
limitations of the service.
This view was made pending clarification from the European Commission on its
position. At that time, Ofcom’s understanding of the Commission’s position was
that providers should be able to choose whether or not they were providing PATS
even if they offered the four core elements of PATS, including access to emergency
services. This would mean that a service provider could offer access to emergency
services, without being subject to the full rigour of the PATS obligations.
It was not until March 2007 and following a period of heavy lobbying by some
parts of the industry that Ofcom issued its final statement.147 The main decisions in
the March statement were:
When Ofcom set out its forbearance policy it was partly on the basis that this re-
flected the view of the Commission. The Commission did however clarify its view
and stated that where a service meets all the gating criteria it automatically be-
comes and must be regulated as a PATS.148
Ofcom issued a consultation over the summer and in December 2007149 an-
nounced that as from 8 September 2008:
• VoIP Out services which allow customers to make calls over the internet to the
PSTN but not receive calls from the PSTN; and
• VOIP In and Out services which allow customers to make calls over the internet
to the PSTN and receive calls from the internet over the PSTN
147
’Regulation of VoIP Services—Statement and publication of statutory notifications under section 48(1) of
the Communications Act 2003 modifying General Conditions 14 and 18’.
148
Expert Group on Emergency Access working document, ‘Regulatory clarification of ECS/PATS and
Fixed/Non-Fixed’, 23 May 2006.
149
’Regulation of VOIP services: Access to Emergency Services’, December 2007.
150
See further Chapter 4, at Section 4.2, for a discussion of ongoing reforms around this issue.
14
• Next generation core networks (NGNs)—t hese are internet protocol based core
or backbone networks which can support a variety of existing and new services.
Typically they replace multiple legacy core networks with a single IP based net-
work for the provision of all services. For example, BT has announced that it
intends to switch off its legacy PSTN core network by 2025; and
• Next generation access networks—t hese are broadband access networks which
connect the user to a core network capable of a bandwidth quantity and quality
significantly in excess of current levels. For example, the provision of ‘fibre to
the curb’ and ‘fibre to the home’. This enables new services such as HDTV over
broadband to be delivered to end users.
It is the provisioning of the latter that has raised the more significant policy and
regulatory issues.
There are a number of issues associated with these new networks and in par-
ticular with the access networks.153 For current access network owners, next
Video-on-demand services, such as Netflix and BBC’s iPlayer, have increased the strain on existing
152
networks.
153
Broadband Stakeholder Group, ‘Pipe Dreams? Prospects for next generation broadband deployment in
the UK’, 16 April 2007.
412
generation access deployments offer the prospect of very high end customer
bandwidths, but they require significant investment in infrastructure. For market
competitors and new entrants, these new networks may result in changes to the
wholesale products and services that they can purchase.
Ofcom has recognized that the development of the new networks presents it
with the opportunity to influence the competitive landscape in years to come.154
The challenge for governments and regulators is to encourage investment in these
new networks and at the same time maintain the benefits of competition avoiding
the issues of regulatory bottlenecks which have plagued the industry over the
last 30 years. The current regulatory framework was designed to open up access
to a legacy infrastructure (essentially paid for by the tax payer and inherited by
BT). This may not be appropriate to regulate one designed to enable investment
in a whole new infrastructure and at the same time to maintain the benefits of
competition.
The government views broadband investment as a key policy issue, due to its
perceived economic and social benefits for the UK. There has also been concern
that the UK lags behind some of its competitors in the deployment of fibre access
networks, for example Japan and South Korea. A number of reasons have been
put forward for this, including the higher cost of deployment in the UK and the
continuing capabilities of the legacy copper access network infrastructure.155 The
government has established a dedicated unit, Broadband Delivery UK, within
the Department of Culture Media and Sport, to oversee a range of different pro-
grammes, such as the ‘Superfast Broadband Programme’, which involves sub-
stantial public investment.156 The current target is to ensure that 95 per cent of UK
premises have access to a minimum line speed of 24 Mbps, with the remaining
5 per cent having at least 2Mbps.157 In addition, there are various other schemes
designed to support local authority investment, take-up by SMEs and improving
mobile infrastructure.158 There has been criticism that the vast majority of the state
aid to date has gone to BT, the ex-i ncumbent, although it is required to pay back
some of the funding it receives if customer take-up reaches a certain threshold, as
well as share its profits for seven years after the network becomes operational.159
154
See Ofcom, ‘Future Broadband, Policy approach to next generation access’, 29 September 2007.
155
BT’s G.fast technology enables speeds up to 330 Mbps over copper.
156
eg the Superfast programme involves three phrases, with some £800m of central government funding,
which should be matched by private investment.
157
DCMS, ‘UK Next Generation Network Infrastructure Deployment Plan’, March 2015.
158
See generally <https://w ww.gov.uk/g uidance/broadband-delivery-u k>.
159
The Telegraph, ‘BT defends £1.7bn rural broadband scheme as rivals question whether taxpayer cash was
needed’, 16 June 2016.
143
3.4.6 Brexit
At the time of writing, the UK is scheduled to leave the EU in 2019, subject to
any transitional period. Our departure is clearly going to have significant im-
plications for most, if not all, sectors of industry, telecommunications included.
This section considers some of the implications of Brexit for UK telecommunica-
tions law, although the reality will likely vary depending on whether the Brexit is
‘hard’ or ‘soft’ and emerging government policy.160 As noted above, Part 2 of the
Communications Act 2003 transposes four of the key EU Directives that com-
prise the New Regulatory Framework. These measures are to be recast into the
European Electronic Communications Code (EECC), although the transposition
date will fall after the UK’s scheduled departure. As such, the UK will not be ob-
liged to implement into domestic law. Since much of the EECC is a recasting of ex-
isting rules, whether the government decides to follow or refrain may not result in
a substantially different regime. One key issue, however, is the scope of application
of the new rules, with the EECC adopting a functional approach that will embrace
OTT services that currently lie outside the regulatory sphere.161 While there are
strong consumer protection arguments in favour of the UK doing the same, the
government may consider a divergent approach to have net benefits for the UK
economy.
One reason why taking a divergent approach from the EU may confer only in-
cremental advantages for the UK, however, is another factor that may reduce the
overall impact of Brexit on operators, ie the fact that the provision of telecommu-
nications networks and services are not subject to a ‘country of origin’ approach
to EU harmonization. Operators have to conduct themselves on a distinctly na-
tional basis, ensuring compliance in every state in which they provide services,
and subject to oversight and intervention by national regulatory authorities, such
as Ofcom, who have significant freedom to govern the market. As a consequence,
an operator may gain little regulatory advantage from locating in the UK, but pro-
viding services into the other Member States.
Another reason for expecting less change post-Brexit is because responsibility
for key areas of the telecommunications sector have been retained under national
control, rather than handed over to the EU. With respect to wireless communica-
tion services (eg mobile and satellite), control and management of spectrum has
remained the near exclusive (and valuable) property of the state. While spectrum
usage requires co-ordination and co-operation, this occurs at an international
160
As at January 2018, the DCMS was in the process of recruiting a policy adviser on ‘Telecoms EU Exit
Policy and Legislation’ (advertised on Civil Service Jobs).
161
See further Chapter 4, at Section 4.2.
41
162
See Chapter 16, at Section 16.3. 163
See Chapter 7, at Section 7.5.
164
eg Town and Country Planning (General Permitted Development) (England) Order 2015/596, Schedule
1, Part 16 Communications.
165
See further Chapter 6, at Section 6.4.
166
Council Regulation 139/2004 on the control of concentrations between undertaking, OJ L 24/1, 29
January 2004, at Art 1. See further Chapter 10.
167
Case M.7612, Hutchison 3G UK/Telefónica UK, Commission decision declaring a concentration incom-
patible with the internal market, OJ C 357/15, 29 September 2016.
168
eg Case COMP/M.1946, Bellsouth/SBC/J V, OJ C 202/5, 15 July 2000.
169
Regulation 531/2012 on roaming on public mobile communications networks within the Union, OJ L
172/10, 30 June 2012.
170
eg <http://telecoms.com/2 33002/smaller-european-operators-oppose-eu-roaming-d irective/>.
145
As with all developed nations, the UK’s telecommunications market has changed
dramatically over the past 30 or more years. While the liberalization process has
not always been smooth, the UK can rightly lay claim to having been a leader
in embracing a competitive market. Although BT remains a dominant player in
many markets, full privatization has exposed it to a fiercer competitive environ-
ment than many of its fellow ex-i ncumbents in continental Europe.
171
Rosas, M, ‘Is the abolition of EU roaming charges an opportunity or threat for operators’, 26 August
2016, at <http://w ww.vanillaplus.com/2016/0 8/26/20824-is-t he-abolition-of-eu-roaming-charges-a n-
opportunity-or-t hreat-for-operators/>.
172
FCC, In the matter of Restoring Internet Freedom, Declaratory Ruling, Report and Order, 14 December
2017 (FCC 17-166).
416
147
4.1 INTRODUC TION
The past thirty years and more has seen an extraordinary level of policy, legal,
and regulatory activity in the telecommunications sector within the European
Union (EU); with well over 100 different directives, decisions, regulations, recom-
mendations, and resolutions, relating to every aspect of the industry, having been
adopted since 1984.1 Such activity is a clear illustration that market liberalization
should not be confused with concepts of market deregulation. While from a UK
perspective, initial EU regulatory intervention in the telecommunications sector
seldom impinged on the wider public consciousness, largely due to developments
already commenced domestically,2 some Member States experienced significant
political fall-out from Commission initiatives in the area, such as public sector in-
dustrial action.
1
Council Recommendation (84/5 49/E EC) concerning the implementation of harmonization in the field of
telecommunications, OJ L 298/49, 16 November 1984.
2
See further Chapter 3.
418
The chapter is broadly divided in two: the first part reviews the historical de-
velopment and key components of the EU regulatory framework; the second part
examines particular elements addressed by the framework. It is not the objective
of this chapter to provide a detailed analysis of every legal instrument in the field,
in part because such a treatment would require a complete book on its own; but
also because many aspects are examined in depth in other chapters of the book.
Rather this chapter is designed to place the mass of EU laws, decisions, and regu-
lations into a comprehensible contextual framework.
In the second phase, from 1993 to 2002, full market liberalization became pol-
itically acceptable as concerns about the impact of liberalization failed to materi-
alize. The key commitment to liberalization came on 22 December 1994, when the
Council of Ministers committed themselves to the target date of 1 January 1998
for full liberalization of the voice telephony monopoly and telecommunications
infrastructure in the majority of Member States.4 The fact that such a fundamental
3
Commission, ‘On the Development of the Common Market for Telecommunications Services and
Equipment’, COM(87) 290 final of 30 June 1987. See also Commission, ‘On the Way to a Competitive
Community-W ide Telecommunications Market in the Year 1992’, COM(88) 48 final of 9 February 1988.
4
Council Resolution of 22 December 1994 on the principles and timetable for the liberalization of telecom-
munications infrastructures, OJ C 379/4, 31 December 1994.
419
change in the legal framework governing a market was undertaken and substan-
tially achieved in a relatively short period of time illustrates the considerable
degree of consensus between Member States, the Community institutions, and
industry itself. However, the reality of a fully competitive market, as well as the
establishment of a single European market, is taking considerably longer, as the
divergent interests involved emerge and are fully expressed during the process of
implementation.
A third phase of EU telecommunications policy commenced on 25 July 2003,
when the new ‘Framework Directive’5 and the specific measures came into force,
the ‘New Regulatory Framework’ (NRF):
5
Directive 2002/21/EC on a common regulatory framework for electronic communications networks and
services, OJ L 108/33, 24 April 2002.
6
Directive 2002/20/EC on the authorization of electronic communications networks and services, OJ L
108/21, 24 April 2002. See further Chapter 6.
7
Directive 2002/19/EC on access to, and interconnection of, electronic communications networks and
associated facilities, OJ L 108/7, 24 April 2002. See further Chapter 8.
8
Directive 2002/22/EC on universal service and users’ rights relating to electronic communications net-
works and services, OJ L 108/7, 24 April 2002. See further Chapter 9.
9
Directive 2002/58/EC concerning the processing of personal data and the protection of privacy in the
electronic communications sector, OJ L 201/37, 31 July 2002. See further Chapter 13.
10
See Commission Communication, ‘Towards a new framework for Electronic Communications infra-
structure and associated services: The 1999 Communications Review’, COM(1999)539, 10 November 1999;
at p vi.
11
See further Chapter 16, at Section 16.4.
12
The Commission has produced reports on the implementation of the regulatory framework since 1998;
the most recent was published in 2015; available at <https://ec.europa.eu/d igital-single-market/en/news/
implementation-eu-regulatory-f ramework-electronic-communications-2015>.
510
As with any regulatory regime, definitions constitute the boundaries that deter-
mine what falls within and outside the regulated sphere. Such definitions attempt
to reflect, not describe, the marketplace to which they apply, since an undertaking’s
13
For the purpose of this Chapter, the phrase ex ante (‘before the fact’) is used in respect of regulatory meas-
ures that proactively control the manner in which entities operate going forward; while ex post (‘after the fact’)
refers to measures that arise in reaction to the decisions and activities of entities.
14
This measure has now been codified in Directive 2015/1535/E U laying down a procedure for the provision
of information in the field of technical regulations and of rules on Information Society services, OJ L 241/1, 17
September 2017.
51
15
Directive 2010/13/E U ‘on the coordination of certain provisions laid down by law, regulation or admin-
istrative action in Member States concerning the provision of audiovisual media services’, OJ L 95/1, 14 April
2010. Note that a proposal to reform the current regime was agreed by the EU institutions on 6 June 2018. See
Commission Proposal COM(2016) 287 final of 25 May 2016. See further Chapter 14, at Section 14.2.5.
16
Directive 2000/31/EC on certain legal aspects of information society services, in particular electronic
commerce, in the Internal Market, OJ L 178/1, 17 July 2000.
17
The Audiovisual Media Services Directive employs this criterion for determining whether an ‘on-
demand’ audiovisual media service is ‘television-l ike’; see Recital 24.
512
18
See Commission Staff Working Document on The Treatment of Voice over Internet Protocol (VoIP) under
the EU Regulatory Framework, 14 June 2004, at 3.
19
eg ETNO response to the public consultation on the evaluation and review of the regulatory frame-
work for electronic communication networks and services available at <https://e tno.eu/d atas/p ositions-
papers/2 015/ E xpert_ c ontributions/ E TNO_ R esponse_Telecoms_ F ramework _ R ev iew_ c onsultation_
071215.pdf>.
20
See in Germany, Ruling of 11 November 2015, Administrative Court of Cologne, Ref 21 K 450/15,
Google v BNetzA. See also Case C-518/11, UPC Nederland v Gemeente Hilversum, 7 November 2013, at
paras 35–47.
21
As defined at Art 2(2) of Regulation 2015/2120 laying down measures concerning open internet access, OJ
L 310/1, 26 November 2015 (Open Internet Access Regulation).
22
Proposal for a Directive establishing the European Electronic Communications Code, COM(2016) 590
final of 14 September 2016 (‘2016 Proposal’), at Art 2(5).
513
23
See Proposal for a Regulation concerning the respect for private life and the protection of personal data
in electronic communications, COM(2017) 10 final of 10 January 2107, at Recita1 11.
24
Proposal for a Regulation concerning the respect for private life and the protection of personal data
in electronic communications, COM(2017) 10 final of 10 January 2017, at Art 4(2). See further Chapter 13, at
Section 13.2.
25
Universal Services Directive, Article 22(3). 26
See further Chapter 15.
27
Framework Directive, Art (1)4. See further Section 4.4.3.
514
referred to as the ‘interface’, which meant either the ‘network termination point’
for fixed network access or the ‘air interface’ for wireless access,28 although this
concept has since been removed. Certain end-user equipment may also contain
components that are categorized as an ‘associated service’ under the NRF:
those services associated with an electronic communications network and/or an
electronic communications service which enable and/or support the provision
of services via that network and/or service or have the potential to do so and in-
clude, inter alia, number translation or systems offering equivalent functionality,
conditional access systems and electronic programme guides, as well as other
services such as identity, location and presence service. (Framework Directive,
Article 2(ea))
28
Directive 99/5/EC, at Art 2(e).
29
Framework Directive, Art 2(f). See further Chapter 8, at Section 8.3.4.5 and Chapter 14, at Section 14.3.3.2.
30
Ibid, at Art 2(o). 31
Ibid, at Art 2(d).
32
Oftel, Guidelines for the interconnection of public electronic communications networks, 23 May 2003, at
para 6.1 et seq; as endorsed in Ofcom’s Guidelines on the General Conditions of Entitlement, see <http://w ww.
ofcom.org.uk/telecoms/ioi/g _ a _ regime/gce/gcoe/>.
33
eg Framework Directive, at Art 25.
51
was operating successfully, with only relatively minor amendments and improve-
ments being proposed.34 In November 2007, the Commission published a series
of legislative proposals to amend the NRF, key areas for reform being in respect
of spectrum management and the procedural burden in respect of the market re-
views and the resultant ex ante remedies.35 Adoption of the final texts occurred
in November 2009 (the ‘2009 Reforms’): The ‘Citizens’ Rights’ Directive36 and the
‘Better Regulation’ Directive.37 Member States were required to transpose these
amendments into national law by May 2011.
A second review was commenced in 2015 and led to the publication, in September
2016, of the Commission’s proposal to establish the ‘Electronic Communications
Code’ (‘Code’).38 The Code is part of the REFIT programme to simplify EU laws,
which in this case involves recasting four of the five NRF measures into a single
code. It does not radically depart from the NRF, so cannot be considered to repre-
sent a new generation of EU telecommunications law, but is intended to consoli-
date the existing instruments.39 On 5 June 2018, political agreement on the new
Code was reached between the Commission, Parliament, and Council.
The basis for Community involvement in the telecommunications market has pri-
marily been founded on two different strands of European Treaty law: competi-
tion law (Articles 101–109) and the establishment of the ‘Internal Market’ (Article
114).40 The former articles have been primarily used to open up national markets to
34
Communication from the Commission to the Council, the European Parliament, the European Economic
and Social Committee and the Committee of the Regions on the Review of the EU Regulatory Framework for
electronic communications networks and services, COM(2006) 334 final (28 June 2006); and accompanying
Commission Staff Working Document, SEC(2006) 816.
35
See Commission Communication, Report on the outcome of the Review of the EU regulatory frame-
work for electronic communications networks and services in accordance with Directive 2002/21/EC and
Summary of the 2007 Reform Proposals, COM(2007)696 rev 1 (‘2007 Reform Proposals’).
36
Directive 2009/136/EC amending Directive 2002/22/EC on universal service and users’ rights relating to
electronic communications networks and services, Directive 2002/58/EC concerning the processing of per-
sonal data and the protection of privacy in the electronic communications sector, and Regulation (EC) No
2006/2004 on cooperation between national authorities responsible for the enforcement of consumer protec-
tion laws, OJ L 337/11, 18 December 2009.
37
Directive 2009/140/EC amending Directives 2002/21/EC on a common regulatory framework for elec-
tronic communications networks and services, 2002/19/EC on access to, and interconnection of, electronic
communications networks and associated facilities, and 2002/20/EC on the authorisation of electronic com-
munications networks and services, OJ L 337/37, 18 December 2009.
38
2016 Proposal. 39
See further below and Chapters 6, 7, 8, and 9.
40
Treaty on the Functioning of the European Union (‘TFEU’), OJ C 83/47, 30 March 2010.
516
competition, whilst the latter has primarily addressed competition issues between
national markets, through harmonization measures.
Surprisingly, however, the telecommunications market has not been subject to
harmonization measures under the freedom to provide services provisions of the
Treaty (Articles 56–62). As a consequence, the provision of ‘electronic communi-
cation services’ is not subject to the ‘country of origin’ principle, whereby busi-
nesses established in one Member State are free to supply services into the other
twenty-seven Member States without further authorization or regulatory control
from the recipient state (except in limited and procedurally controlled circum-
stances).41 This is in stark contrast to the provision of other closely related services,
specifically ‘information society services’ and ‘audiovisual media services’. The
Commission has repeatedly sought to address this apparent anomaly, with the
adoption of a ‘one-stop shopping procedure’,42 a proposal to establish a European
Communications Markets Authority,43 and a 2013 proposal for a single authoriza-
tion regime;44 each of which has either been disregarded or rejected by the Member
States. While the current approach was initially seen as reflecting perceptions that
communication services were intimately tied to the physical networks over which
they operated; the continued intransigence of the Member States must be viewed
as being deeply rooted in a range of political imperatives and the ‘public interest’
nature of telecommunications.
Initiatives within each area have been the responsibility of different depart-
ments of the European Commission; harmonization measures originating within
DG Connect and liberalization issues residing primarily with the DG Competition.
The role of DG Competition in the development of EU policy in the telecommuni-
cations sector has been very considerable. Indeed, the manner in which EU com-
petition law has been applied to the telecommunications sector provides a case
study of the significance of competition law within the acquis communautaire. In
particular, Article 106(3) of the Treaty of the Functioning of the European Union
(TFEU) bestows a supervisory function upon the Commission, supported by spe-
cial law-making powers:
3. The Commission shall ensure the application of the provisions of this Article
and shall, where necessary, address appropriate directives or decisions to Member
States.
41
See <http://ec.europa.eu/i nternal_market/services/docs/services-d ir/g uides/cop_en.pdf>.
42
Directive 97/13/EC on a common framework for general authorisations and individual licences in the
field of telecommunications services, OJ L 117, 7 May 1997, Art 13.
43
See further Section 4.7.2.
44
Proposal for a Regulation laying down measures concerning the European single market for electronic
communications and to achieve a Connected Continent, COM(2013) 627 final of 11 September 2013.
517
45
The relevant Treaty provision at the time was 90(3).
46
Directive 88/301/E EC, OJ L 131/73, 27 May 1988 and Directive 90/388/E EC, OJ L 192/10, 24 July 1990.
47
Case C-202/8 8: France v Commission [1992] 5 CMLR 552; and Case C-271/9 0 Spain v Commission [1992]
ECR I-5833.
48
TEC, Art 249.
49
Regulation 2887/2000 of the European Parliament and Council of 18 December 2000 on unbundled ac-
cess to the local loop, OJ L 336/4, 30 December 2000; which was repealed by the Better Regulation Directive
(Art 4).
50
See Regulation (EC) 717/2007 on roaming on public mobile telephone networks within the Community,
OJ L 171/32, 29 June 2007; repealed by Regulation (EU) 531/2012 on roaming on public mobile communica-
tions networks within the Union, OJ L 172/10, 30 June 2012.
51
See n 22.
52
Regulation 1211/2009 establishing the Body of European Regulators for Electronic Communications
(BEREC) and the Office, OJ L 337/1, 18 December 2009. See Section 4.7.
53
It was repealed by the Better Regulation Directive (n 37), at Art 4.
518
profile issue with the general public, who experienced high roaming charges
when travelling within Europe, as well as being politically symbolic of the desire
to promote greater European integration. The justification for a Regulation was the
‘urgency and persistence of the problem’,54 and followed on a sectoral inquiry car-
ried out by DG Competition in 2000,55 and investigative raids carried out against
nine European mobile operators based in the UK and Germany.56 The initial 2007
Regulation was first amended in 2009, then replaced in 2012 by a measure, which
was then amended again in 2015 leading to the abolition of roaming charges by
15 June 2017.57 Such direct ‘state’ intervention in the market, mandating the retail
price of a service, is not just prompted by competition and consumer protection
concerns, but is another indication of the ‘public interest’ nature of telecommu-
nications,58 in this case, the desire to promote the concept of the EU as a single
market. The Open Internet Access Regulation addresses the issue of ‘net neu-
trality’, constraining the ability of operators to discriminate certain types of con-
tent, application, or service (see further Chapter 15). Taken together, it becomes
apparent that the choice of a Regulation as the legislative instrument correlates to
the extent of public consciousness and debate around the applicable issue.
The Commission has also made extensive use of ‘soft law’ measures, both
formal Recommendations59 and informal guidelines and notices.60 Under the
Framework Directive, the Commission can issue a recommendation to address
divergences in the implementation by NRAs of regulated tasks under the NRF,61 to
which NRAs must ‘take the utmost account of’,62 but which have no binding legal
force.63 Such documents are used both to further harmonization among Member
States, providing a benchmark of good practice for national regulatory authorities,
54
COM(2006) 382 final, 12 July 2006, at p 8.
55
See <http://ec.europa.eu/comm/competition/sectors/telecommunications/a rchive/i nquiries/roaming/
index.html>.
56
Commission Press Release, ‘Statement on inquiry regarding mobile roaming’, MEMO/ 01/262, 11
July 2001.
57
Regulation 531/2012, as amended, at Art 6a. 58
See Chapter 1, at Section 1.7.
59
eg Commission Recommendation 2005/698/EC ‘on accounting separation and cost accounting sys-
tems under the regulatory framework for electronic communications’, OJ L 266/6 4, 11 October 2005 and
Recommendation 2010/572/E U ‘on regulated access to Next Generation Access Networks’, OJ L 251/35, 25
September 2010.
60
eg Guidelines on the Application of EEC Competition Rules in the Telecommunications Sector, OJ C 233/
2, 6 September 1991.
61
Framework Directive, Art 19(1). 62
Ibid, Art 19(2).
63
TFEU, Art 288. However, national courts are bound to take recommendations into consideration when
deciding disputes, especially where they ‘cast light on the interpretation of national measures adopted
in order to implement them or where they are designed to supplement binding EU provisions’. (Case C-2 8/
15, Koninklijke KPN BV v ACM, 15 September 2016, at para 41). The Framework Directive also provides the
Commission with the power to adopt a binding decision on a matter, two years after issuing a recommenda-
tion (Art 19(3)(a)).
519
particularly in respect of the complex but critical areas of pricing and cost ac-
counting, as well as providing assistance to undertakings, both market players
and potential entrants, about how the Commission views particular matters, par-
ticularly in terms of competition analysis.
However, while ‘soft law’ has been used by the Commission to pursue its competi-
tion agenda, it is pertinent to note that the NRF does not expressly acknowledge the
use of self-regulation as an element of the regulatory regime; while co-regulation is
only referred to once in connection with the enhancement of service quality.64 This is
in contrast to the position adopted in some Member States, such as the UK,65 where
industry self-regulation is expressly referred to as a means of moving towards de-
regulation as competitive markets become established. The technical complexity of
the telecommunications market has always meant that much of the input on certain
issues, such as interconnection, primarily consisted of the convening and oversight
of particular industry groups; intervening only in the event of impasse. As regulators
reduce or withdraw from ex ante intervention in the market, as they are obliged to do
under the NRF,66 then increasing reliance is likely to be made upon industry to regu-
late itself. This silence about the role of self-regulation runs counter to general EU
policy reflected in an Interinstitutional Agreement on Better Law-making, which ex-
pressly acknowledges the potential role of self-regulation,67 as do measures in related
areas, specifically the provision of audiovisual media services.68
DG Internal Market has also been responsible for some initiatives relating dir-
ectly or indirectly to the telecommunications sector. It is responsible for elec-
tronic commerce issues, including regulating the provision of ‘information society
services’,69 which will generally be offered by telecommunication services pro-
viders. DG Internal Market was also responsible for data protection issues, which
included sectoral measures imposing special obligations in the telecommunica-
tions sector; although the responsibility has subsequently transferred to the DG
Justice.70
The CJEU has inevitably played a role in the development of European tele-
communications law as the ultimate arbiter of European legal instruments.
Proceedings have come before the Court based on one of four legal grounds pro-
vided for under the TFEU:71
64
Universal Services Directive, Recital 48.
65
ie The Communications Act 2004, s 6(2), requires Ofcom to have regard to whether policy could be
achieved through ‘effective self-regulation’, which is further defined at s 6(3).
66
See Framework Directive, at Art 16(3). 67
OJ C 321/1, 31 December 2003, at paras 22–2 3.
68
Audiovisual Media Services Directive at Art 4(7). 69
See n 14. 70
See further Chapter 13.
71
See generally Commission, Guide to the Case Law of the Court of Justice of the European Union in the field
of Telecommunications (January 2010), available at <https://ec.europa.eu/d igital-single-market/sites/d igital-
agenda/fi les/g uidetocaselaw2010en_0.pdf>.
610
Finally, it should be noted that the WTO agreements addressing the telecom-
munications sector, such as the Annex of Telecommunications and the Reference
Paper, comprise a potential source of EU law in terms of interpretation and appli-
cation, if not a basis for initiating proceedings before the Court of Justice.78
As noted, the basis for the liberalization of Member State markets was the appli-
cation of European competition law. The first indication of the potential impact
of these rules arose in a Commission decision against the UK incumbent, British
72
eg Case C-411/02, ECJ, 16 March 2004 (Austria); Case C-500/01, OJ C 47/6, 21 February 2004 (Spain); Case
C-97/01, OJ C 184/4, 2 August 2003 (Luxembourg); Case C-221/1, OJ C 274/14, 9 November 2002 (Belgium); Case
C-146/0 0, OJ C 84/2 3, OJ 6 April 2002 (France); Case C-396/99 [2001] ECR I-7577 (Greece); Case C-429/99, OJ C
369/3, 22 December 2001 (Portugal).
73
eg Case T-310/0 0, MCI Inc v Commission and France [2004] 5 CMLR 26, against the Commission’s decision
to prohibit the merger of MCI WorldCom/Sprint.
74
max.mobil Telekommunikation Service GmbH v Commission [2002] 4 CMLR 32.
75
eg France Télécom SA v Commission [2007] 4 CMLR 21, and Deutsche Telekom AG v Commission, CFI
Judgment, 10 April 2008.
76
eg Case C-18/8 8 RTT v GB- Inno-BM SA [1991] ECR I-5941; Case C-79/0 0 Telefónica de España SA
v Administración General del Estado [2002] 4 CMLR 22; and Case C-369/0 4, Hutchison 3G (UK) Ltd & ors v
Commissioners of Customs and Excise, 26 June 2007 re: payment of VAT on spectrum auction transactions.
77
eg Cases C-152/07 and C-154/07, Arcor AG & Co. KG and others v Bundesrepublik Deutschland, 17 July 2008.
78
See further Chapter 16, at Section 16.4.4.
16
81
Case 311/8 4 Centre Belge d’Etudes de Marché- Télé-
Marketing v Compagnie Luxembourgeoise de
Telediffusion SA and Information Publicite Benelux SA [1986] 2 CMLR 558.
82
See also Chapter 10.
612
joint ventures, merger activities, and even state aid83 in every aspect of the sector.
Such regulatory intervention has extended to alliances and mergers between na-
tional incumbents;84 in the mobile sector;85 concerning internet infrastructure;86
and with providers of content services.87 In all these cases, the Commission has
been concerned to protect the interests of European consumers and industry
against the inevitable commercial pressures created by the developing global
economy. The Commission, as competition authority, has also fined undertakings
for abusive practices in the market, including Deutsche Telekom AG,88 Wanadoo
Interactive,89 and Telefónica SA.90
During the initial phases of telecommunications liberalization in Europe, the
process was underpinned by two legal phrases that were key elements of the ex
ante legislative measures adopted by the Commission, that of ‘special or exclusive
rights’ and ‘essential requirements’.
83
eg France Télécom [2003] OJ C 57/5, 12 March 2003. On 20 July 2004, the Commission ordered France
Télécom to repay up to £1.1bn in back taxes, estimated savings that the firm had made from the granting of
exemptions from local taxes that constituted a form of state aid. Mobilcom [2003] OJ C 80/5, 3 April 2003 and
[2003] OJ C 210/4, 5 September 2003.
84
eg France Télécom and Deutsche Telekom (Case No IV/35.337—Atlas; OJ L 239/2 3, 19 September 1996);
Telia and Telenor (Case IV/M.1439; OJ L 40/1, 9 February 2001).
85
eg Vodafone Airtouch and Mannesmann (Case No Comp/M.1795; OJ C 141/19, 19 May 2000).
86
eg WorldCom and MCI (Case IV/M.1069; OJ L 116/1, 4 May 1999).
87
eg AOL and Time Warner (Case No COMP/M.1845; OJ L 268/2 8, 9 October 2001).
88
OJ L 263/9, 14 October 2003, imposing a fine of €12.6m.
89
Decision of 16 July 2004, imposing a fine of €10.35m.
90
Decision of 4 July 2007, imposing a fine of €151m.
613
Member States challenged both directives before the CJEU.91 The Court found
in the Commission’s favour in respect of the withdrawal of exclusive rights, but
upheld the claims of the Member States in respect of the limitation imposed on
the granting of special rights, on the grounds that the Directives failed to specify
what ‘special rights’ were or the reasons that such rights were contrary to the pro-
visions of the Treaty. Such provisions were therefore void. As a consequence, the
Commission amended the Services Directive to clarify the distinction between
‘exclusive rights’ and ‘special rights’,92 which the CJEU subsequently endorsed.93
‘Special rights’ would include powers of compulsory purchase and derogations
from laws on town and country planning 94 that are granted to undertakings ‘other-
wise than according to objective, proportional and non-d iscriminatory criteria’.95
During the liberalization process, the procurement practices of telecommunica-
tion operators that were public undertakings and operating under special or ex-
clusive rights were also subjected to regulatory controls.96 Such rules are now only
applicable to the purchasing of telecommunication systems and services, rather
than the provision of such services.97
Despite full market liberalization, Article 106(3) may continue to be relevant to
the European telecommunications market. First, in a number of Member States
the incumbent operator continues to be a ‘public undertaking’, through full or
partial state ownership, and as such could be subject to state measures which in-
fringe EU competition law. Second, where an operator has been granted ‘special
or exclusive’ rights in a different sector of activity, such as broadcasting or water
supply, the exercise or existence of such rights might be perceived as distorting
the competition in the telecommunications market.98 As a consequence, ex ante
controls may be imposed on such undertakings, to ensure structural separation
between the activities.99
91
See n 35.
92
See Art 2(1) of Commission Directive (94/4 6/EC) of 13 October 1994 amending Directive 88/301/E EC and
Directive 90/388/E EC in particular with regard to satellite communications, OJ L 268/15, 19 October 1994.
93
Case C-302/94, R v Secretary of State for Trade and Industry, ex parte British Telecommunications plc, ECR
I-6 417, at para 34.
94
Ibid, at Recital 11.
95
Commission Directive 2002/77/EC on competition in the markets for electronic communications net-
works and services, OJ L 249/21, 17 September 2002.
96
Directive 93/38/E EC coordinating the procurement procedures of entities operating in the water, energy,
transport and telecommunication sectors, OJ L 199/8 4, 9 August 93, now repealed.
97
Directive 2014/2 4/E U of 26 February 2014 on public procurement and repealing Directive 2004/18/EC,
OJ L 94/65, 28 March 2014, Art 8.
98
For the application of Art 106 to the broadcasting sector see Case C-260/89, Elliniki Radiophonia Tileorassi
(1991) ECR I-2925. In the UK, Ofcom has the power to impose ‘privileged supplier’ conditions on an operator in
such circumstances (Communications Act 2003, s 77).
99
Framework Directive, Art 13.
614
Over the years public policy concerns broadened to encompass the protection of
personal data and environmental issues, impacting on the building of network in-
frastructure, such as mobile transmitters and digging-up streets to lay cable.
100
Case C-18/8 8, Régie des Télégraphes et des Téléphones v GB-Inno-BM SA [1991] ECR I-5941.
101
As defined at Art 3 of Directive 2014/53/E U of 16 April 2014 on the harmonisation of the laws of the
Member States relating to the making available on the market of radio equipment, OJ L 153/62, 22 May 2014.
102
As defined by Art 1(1) of Directive 90/388 (as amended by Directive 96/19/EC) and Art 2(6) of Directive 90/
387.
615
In the first stages of liberalization, much concern was directed towards the im-
pact on the ‘national’ (ie incumbent) network of new operators connecting ‘un-
regulated’ telecommunications equipment and generating substantial volumes
of additional traffic. The network, as a strategic component of Member State
economies, was viewed as being vulnerable in a competitive environment. Over
time such concerns for the ‘national’ network have generally proven to be largely
overstated.
Incumbent operators have, however, continued to use the terminology of the
‘essential requirements’ as grounds for imposing restrictive conditions on new en-
trants. In the UK, for example, BT has used concerns about ‘network security’ as a
justification for requiring separate co-location rooms for operators implementing
ASDL at BT’s local exchanges, which impacted on operators’ timescales and costs
for the introduction of competing services. At times, new entrants have expressed
concern that national regulatory authorities did not always scrutinize fully the
evidence for some of these ‘essential requirement’ claims.103
While the concept of ‘essential requirements’ continues to be utilized in respect
of telecommunications equipment (see Section 4.4.3), its use as a distinct regu-
latory concept in respect of telecommunication networks and services has dis-
appeared; although some of the elements that comprise the concept continue to be
specific EU regulatory objectives under the Framework Directive,104 and all of the
elements comprise conditions that may be attached to an authorization granted
by a Member State under the Authorisation Directive.105
103
See Commission Communication, ‘Sixth Report on the Implementation of the Telecommunications
Regulatory Package’, COM(2000) 814, 7 December 2000, at p 16 et seq.
104
eg Art 8(3)(b) ‘interoperability of services’, Art 8(4)(f) ‘ensuring that the integrity and security of public
communications networks are maintained’.
105
See Annex at A. ‘Conditions which may be attached to a general authorization’ and B. ‘Conditions that
may be attached to rights of use for radio frequencies’. See further Chapters 7 and 8.
106
Council Directive of 24 July 1986 on the initial stage of the mutual recognition of type approval for tele-
communications terminal equipment, 86/361/E EC; OJ L 217/21, 5 August 1986.
61
the emergence of strong global players, such as Nokia and Ericsson, accompanied
by relatively light regulatory intervention.
At the outset, liberalization of the telecommunications terminal equipment
market primarily focused on the application of the principle of the free movement
of goods, under Articles 34–37 of the TEC. In 1985, for example, the Commission
intervened on the basis of Article 37 against Germany in respect of a proposed
regulation extending the Bundespost’s monopoly over telecommunications
equipment to cordless telephones.107 As with other product areas, mutual recogni-
tion was the initial vehicle for the achievement of a ‘Single Market’. The first legis-
lative initiative was a Council Directive in 1986 that called upon Member States to
implement mutual recognition in respect of conformity tests carried out on mass-
produced terminal equipment.108
A more comprehensive, and controversial, measure was taken by the
Commission in 1988 when it adopted a directive, under Article 106(3) (then Article
86), calling upon Member States to withdraw any ‘special or exclusive’ rights that
may have been granted to undertakings relating to telecommunications terminal
equipment.109 The Directive stated that the only grounds upon which a Member
State could restrict or regulate economic operators from importing, marketing,
operating, and maintaining terminal equipment was where such equipment
could either be shown to have failed to satisfy the ‘essential requirements’ or the
economic operator failed to possess the necessary technical qualifications in rela-
tion to the equipment.110
The mutual recognition process, first established under the 1986 Directive and
extended under a series of measures addressing terminal equipment,111 comprised
a number of inter-l inked principles and procedures, which continue to be largely
applicable:
107
Re Cordless telephones in Germany [1985] 2 CMLR 397. See also Case C-18/8 8, Régie des télégraphes et des
téléphones v GB-Inno-BM SA (1991) ECR I-5941, where it was held that Article 30 of the Treaty precludes an
undertaking from having the power to approve telephone equipment for connection to the public network
without being susceptible to legal challenge.
108
See n 112.
109
Commission Directive of 16 May 1988 on competition in the markets of telecommunications terminal
equipment, 88/301/E EC; OJ L131/73, 27 May 1988, Art 2.
110
Ibid, at Art 3.
111
eg Council Directive 91/263/EC on the approximation of the laws of the Member States concerning tele-
communications terminal equipment including the mutual recognition of their conformity, OJ L128/1, 23
May 1991 (repealing 86/361); and Directive 98/13/EC relating to telecommunications terminal and satellite
earth station equipment, including mutual conformity recognition, OJ L 74, 12 March 1998 (repealing 91/263).
617
112
Such designation is now governed by Chapter IV of Directive 2014/53/E U (see n 101). In the UK, there are
nine such bodies authorized in respect of ‘radio equipment’. For a complete listing, see <http://ec.europa.eu/
growth/tools-databases/nando/i ndex.cfm>.
113
The use of the CE marking is now primarily governed by Chapter IV of Regulation 765/2008/EC of 9 July
2008 setting out the requirements for accreditation and market surveillance relating to the marketing of prod-
ucts, OJ L 218/30, 13 August 2008.
114
Directive 1999/5/EC, see n 86, at Recital 7.
618
the subsequent measures amending it.115 In addition, under the 2009 Reforms
‘consumer premises terminal equipment’ was brought partially within the NRF,
specifically in respect of measures designed to improve access to and use of such
equipment by disabled users, such as text relay services.116
The ‘type approval’ regime was again reformed in 2014, with fixed-line equip-
ment being removed from the sectoral regime and placed under generic measures
governing all electrical equipment.117 ‘Radio equipment’ remains subject to a sec-
toral regime designed primarily to ensure the efficient use of spectrum and the
avoidance of harmful interference, but extending to the other ‘essential require-
ments’ outlined in the previous section.118 In the age of the smartphone, end-users
have greater capabilities to modify their devices through the installation of soft-
ware ‘apps’. Concerns that such apps could modify the device and compromise
the ‘essential requirements’ has resulted in a new requirement that users or third
parties should only be capable of loading software on to the radio equipment that
are demonstrably compliant with the ‘essential requirements’.119 Overall, however,
the key elements of the type approval regime remain the same, with the ‘manufac-
turer’ being the primary actor responsible for compliance.120
115
Commission Directive 2008/6 3/EC of 20 June 2008 on competition in the markets in telecommunica-
tions terminal equipment, OJ L 162/20, 21 June 2008.
116
Framework, Art 1(1) and the Universal Services Directive, at Art 23a(2).
117
Directive 2014/35/E U of 26 February 2014 on the harmonisation of the laws of Member States relating to
the making available on the market of electrical equipment designed for use within certain voltage limits, OJ
L 96/357, 29 March 2014; and Directive 2014/30/E U of 26 February 2014 on the harmonisation of the laws of the
Member States relating to electromagnetic compatibility, OJ L 96/79, 29 March 2014.
118
See n 107 at Art 3. It came into effect on 13 June 2016, although subject to a one-year transitional phase
(Art 48).
119
Ibid, at Art 3(3)(i) and Recital 16.
120
ie ‘any natural or legal person who manufactures radio equipment or has radio equipment designed
or manufactured, and markets that equipment under his name or trade mark’ (ibid, at Art 2(1)(12)). See also
Art 10.
121
Commission Directive 90/388/E EC on competition in the markets for telecommunications services, OJ
L192/10, 24 July 1990.
619
122
See further Chapters 2 and 8.
123
Commission Directive 94/4 6/EC amending Directive 88/301/E EC and Directive 90/388/E EC in par-
ticular with regard to satellite communications, OJ L268/15, 19 October 1994.
124
Commission Directive 95/51/EC amending Commission Directive 90/388/E EC with regard to the aboli-
tion of the restrictions on the use of cable television networks for the provision of already liberalized telecom-
munication services, OJ L256/49, 26 October 1995.
125
Commission Directive 96/2/EC amending Directive 90/388/E EC with regard to mobile and personal
communications, OJ L20/59, 21 November 1996.
126
Commission Directive 96/19/EC amending Commission Directive 90/388/E EC regarding the imple-
mentation of full competition in telecommunications services, OJ L74/13, 22 March 1996.
127
Council Decision (97/8 38/EC) of 28 November 1997 concerning the conclusion on behalf of the European
Community, as regards matters within its competence, of the results of the WTO negotiations on basic tele-
communications services, OJ L 347/45, 18 December 1997. See further Chapter 16, at Section 16.4.
710
128
See n 101. 129
See n 2.
130
CEPT is a body comprising some 48 postal and telecommunications ‘administrations’ of European
Countries, not limited to the European Union: <http://w ww.cept.org>.
131
eg Council Resolution of 27 April 1989 on standardization in the field of information technology and tele-
communications, OJ C 117/1, 11 May 1989.
132
Council Resolution of 19 November 1992 on the promotion of Europe-w ide cooperation on numbering of
telecommunications services, OJ C 318/2 , 4 December 1992.
133
See <http://w ww.ero.dk/etns>.
134
Council Decision (91/396/E EC) of 29 July 1991 on the introduction of a single European emergency call
number, OJ L 217/31, 6 August 1991; Council Decision (92/264/E EC) of 11 May 1992 on the introduction of a
standard international telephone access code in the Community, OJ L 137/21, 20 May 1992. Both measures
71
the provision of services of social value, such as hotlines and helplines.135 It was en-
visaged that further Europe-wide numbers would enable companies to utilize non-
geographic European codes for the provision of pan-European services, such as the
provision of mobile services. To date, such schemes have failed to materialize, and
the Commission proposed its removal from the NRF.136 This proposal was rejected,
however, and the ETNS was retained in the Universal Services Directive, at Article
27(2), and the Commission was tasked with establishing a legal entity to manage
and promote the ETNS, similar to that adopted for the ‘.eu’ domain.137 However, due
to lack of demand, the Commission has proposed its removal from the proposed
Code.138
In the mobile sector, the development of European-w ide services has been
pursued through the adoption of a series of legislative measures reserving
common frequency bands within Member States, most importantly in respect of
2G, 3G, and 4G spectrum.139 The initiative on 3G can be seen as a particular suc-
cess story for the EU, facilitating the take-up of GSM as the de facto worldwide
standard and placing European telecommunications companies at the forefront of
the global mobile industry. The GSM measure has since been amended to enable
UMTS services to also use the 900MHz band reserved for GSM, as well as future
generations of mobile telephony.140
In parallel with the Commission’s ‘Services Directive’ in 1990, the Council
adopted a directive, under Article 95 of the TEC, establishing the concept of ‘Open
Network Provision’ (ONP). The so-called ‘ONP framework’ programme was con-
ceived to provide the regulatory basis for imposing harmonization:
have been repealed under Framework Directive, at Art 26, and are consolidated under the Universal Services
Directive at Art 26 and Art 27 respectively.
135
Commission Decision 2007/116/EC on reserving the national numbering range beginning with ‘116’ for
harmonized numbers for harmonized services of social value, OJ L 49/30, 17 February 2007; subsequently
amended by Decision 2007/698/EC, OJ L 284/31, 30 October 2007.
136
Staff Document, see n 28, at 8.2.
137
See Directive 2009/136/EC, at Recital 42. See also Regulation 733/2002/EC on the implementation of the
.eu Top Level Domain, OJ L 113/1, 30 April 2002.
138
See n 23 at p 19.
139
eg Council Directive 87/372/E EC on the frequency bands to be reserved for the coordinated introduc-
tion of public pan-European cellular digital land-based mobile communications in the Community, OJ L 196/
85, 17 July 1987; Council Decision 128/1999/EC on the coordinated introduction of a third-generation mobile
and wireless communications system (UMTS) in the Community, OJ L 17/1, 22 January 1999, and Commission
Implementing Decision 2012/6 88/E U on the harmonisation of the frequency bands 1 920-1 980 MHz and 2
110-2 170 MHz for terrestrial systems capable of providing electronic communications services in the Union,
OJ L 307/8 4, 7 November 2012.
140
Directive 2009/114/EC of the European Parliament and of the Council of 16 September 2009 amending
Council Directive 87/372/E EC on the frequency bands to be reserved for the coordinated introduction of
public pan-European cellular digital land-based mobile communications in the Community; OJ L 274/25, 20
October 2009.
712
This Directive concerns the harmonisation of conditions for open and efficient
access to and use of public telecommunications networks and, where applicable,
public telecommunications services.141
Reflecting the liberalization process, the scope of the ONP programme was initially
limited to issues of access to the network infrastructure and ‘reserved services’
provided by the incumbent operator. As such, the harmonization framework en-
visaged the drafting of proposals on ONP conditions across a range of issues of
concern to providers of non-reserved services:
Such conditions were subject to basic principles concerning the use of objective
criteria, transparency, and non-d iscrimination, whilst any restrictions placed
on access would be limited to reasons based on the ‘essential requirements’.
Subsequent ONP measures were adopted in a number of areas, including the pro-
vision of leased lines; packet-switched data services;142 Integrated Services Digital
Networks (ISDN);143 voice telephony, and interconnection;144 and universal service.
In 1995, the ONP framework was applied to voice telephony.145 Under this
measure, the national regulatory authorities were given a broad range of obliga-
tions to ensure that the provision of ‘fixed’ voice telephony to users, which included
residential customers as well as competing service providers, was under harmon-
ized conditions. Such conditions included the connection of terminal equipment;
targets for supply time and quality of service; service termination; user contracts;
and the provision of advanced facilities, such as calling-line identification (CLI).
141
Directive 90/387/E EC on the establishment of the internal market for telecommunications services
through the implementation of open network provision; OJ L192/1, 24 July 1990.
142
Recommendation 92/382/E EC on the harmonized provision of a minimum set of packet-s witched data
services (PSDS) in accordance with open network provision (ONP) principles; OJ L200/1, 18.7.1992.
143
Recommendation 92/383/E EC on the provision of harmonized integrated services digital network
(ISDN) access arrangements and a minimum set of ISDN offerings in accordance with open network provi-
sion (ONP) principles; OJ L/200/10, 18 July 1992.
144
Directive 97/ 33/EC of the European Parliament and of the Council on Interconnection in
Telecommunications with regard to ensuring Universal Service and Interoperability through Application of
the Principles of Open Network Provision, OJ L 199/32, 26 July 1997.
145
Directive 95/62/EC on the application of open network provision to voice telephony, OJ L321/6, 30
December 1995.
713
Further market liberalization led to the replacement of the voice telephony dir-
ective in 1998, extending certain provisions to mobile voice telephony.146
Harmonization between Member State markets has inevitably involved greater
complexity and detailed regulatory intervention than that required for the liber-
alization of national markets. Such detail arises both from the scope of the issues
addressed, as well as the imposition of asymmetric obligations on market partici-
pants. One feature of the harmonization process is the key role played by the NRAs
in implementing and complying with the principles contained in the harmoniza-
tion measures. Such reliance on NRAs generated, in some instances, new areas of
divergence between market conditions and practices in the Member States.147 This
is reflected, in part, by the fact that the Commission pursued considerably more
infringement proceedings against Member States under Article 258 of the Treaty,
in respect of the harmonization directives, as compared with the liberalization
directives.
NRAs were required to notify the Commission that organizations had been so des-
ignated.150 They also had the discretion to determine that an organization on either
side of the 25 per cent figure fell outside the presumption, based on factors such
146
Directive 98/10/EC on the application of open network provision (ONP) to voice telephony and on uni-
versal service for telecommunications in a competitive environment, OJ L 101/2 4, 1 April 1998.
147
See generally the Sixth Implementation Report, see n 103.
148
Directive 97/51/EC of the European Parliament and of the Council of 6 October 1997 amending Council
Directives 90/387/E EC and 92/4 4/E EC for the purpose of adaptation to a competitive environment in
Telecommunications, OJ L 295/2 3, 29 October 1997.
149
Ibid, at Art 2(3). 150
Ibid, at Art 11(1a).
174
Justifying the lower threshold, the Commission argued that traditional competi-
tion law principles are not adequate to deal with some of the unique features of
the telecommunications market; whilst the trigger also reduced the burden upon
national regulatory authorities to assess ‘dominance’ on a case-by-case basis.153
However, as a result of the Commission’s desire to further deregulate the sector,
as well as addressing legitimacy concerns and the EU’s commitments under the
WTO Reference Paper, the NRF redefines the concept of an operator with ‘signifi-
cant market power’ in the following terms, based on CJEU jurisprudence,154
An undertaking shall be deemed to have significant market power if, either in-
dividually or jointly with others, it enjoys a position equivalent to dominance,
that is to say a position of economic strength affording it the power to behave to
151
See further Chapter 10.
152
Letter from Dr Sidel, German Economic Ministry to Mr Cockborne, DG-X III, dated 13 July 1998; quoted
in Tarrant, A, ‘Significant market power and dominance in the regulation of telecommunications markets’,
(2000) 21(7) European Competition Law Review 320–325.
153
Ibid. 154
eg Case 322/81 Michelin BV v Commission [1983] ECR 3461, para 6.
715
In addition, recognizing the peculiar nature of ‘network’ industries and the oli-
gopolistic structure of various telecommunications markets, such as mobile, ex-
press reference was made to the possibility of two or more undertakings being in
a ‘joint dominant position in a market’, a complex and developing area of EU com-
petition law.156
As under the previous regime, NRAs are required to designate operators as having
‘significant market power’ (Article 14(1)). However, to address the concern about di-
vergent approaches being taken by Member States, the designation procedure is
subject to certain harmonization provisions at each stage of the process: market def-
inition, market analysis, and remedies (ie imposition of ex ante obligations).
First, the Commission issued a Recommendation on the 18 product and service
markets, present at either a retail or wholesale level, in which it considered ‘ex ante
regulation may be warranted’, and to which NRAs are required to give ‘utmost
account’ when defining their national markets (Article 15(3)).157 This was subse-
quently revised in December 2007, reducing the number of markets to seven,158
and again in October 2014, down to four markets,159 illustrating the progress made
towards liberalization.
Second, when NRAs analyse the defined markets to establish whether any par-
ticipant has SMP, they should also give ‘utmost account’ to guidelines concerning
the analysis procedure issued by the Commission (Article 16).160 The intention be-
hind the ‘utmost account’ provisions is clear; however, the enforceability of such
provisions is less certain. When an NRA carries out a market definition, it must do
so ‘in accordance with the principles of competition law’. The Recommendation
sets out three criteria which it considers central to such an analysis:
eg Commission decision: Case IV/M. 1524 Airtours/First Choice [2000] OJ L 93/01 and Court of First
156
Where the criteria are not shown to be present, the application of ex ante regu-
lation would be considered inappropriate. Following such an analysis, were an
NRA to identify a particular market and then to vary that definition to align with
a market defined in the Recommendation, it is arguable that the validity of the
NRA’s final determination could be judicially reviewed.162
Once a designation has been made, the NRA must then determine whether to
maintain, amend, or withdraw existing obligations (Article 16(2)) or which obliga-
tions to impose on the SMP operator to remedy the identified problems. Primacy
is given to the wholesale remedies detailed in the Access Directive (Articles 9–
13b163), with the possibility of imposing remedies at a retail level, where neces-
sary, under the Universal Services Directive (Article 17).164 Where an SMP finding
has been made, an NRA is required to impose at least one of the ex ante remedies
(Article 16(4)). To ensure harmonization at this stage of the process, the European
Regulators Group (ERG), in conjunction with the Commission, adopted a Common
Position ‘on the approach to appropriate remedies in the new regulatory frame-
work’.165 This elaborated a typology of 27 potential competition problems based
around four market scenarios:
• Vertical leveraging: This occurs where a dominant firm seeks to extend its
market power from a wholesale market to a vertically related wholesale or re-
tail market.
• Horizontal leveraging: This applies where an SMP operator seeks to extend its
market power to another market that is not vertically related.
• Single market dominance: The problems which may occur within the context
of a single market are entry deterrence, exploitative pricing practices, and pro-
ductive inefficiencies.
• Termination (Two-way access): This relates to the link between price set-
ting in termination markets and in the related retail markets that may be
competitive.
Once the competition problem(s) has been identified, the NRAs should follow
certain principles in determining the appropriate remedy. First, the decision must
be adequately reasoned, with full consideration of alternatives and representing
the least burdensome option. Second, where infrastructure competition is not
162
Under the Communications Act 2003, OFCOM is only required to ‘take due account’ of the Commission’s
Recommendation and Guidelines (s 79(2)).
163
ie transparency (Art 9), non-d iscrimination (Art 10), accounting separation (Art 11), access to network
facilities (Art 12), price control and cost-accounting (Art 13), functional separation (Art 13a), and voluntary
separation (Art 13b).
164
eg retail price caps. 165
ERG (03) 30rev1 (April 2004).
71
feasible, sufficient access to wholesale inputs should be ensured. Third, where in-
frastructure replication is feasible, the remedies should assist transition to such
a situation, for example through investment incentives. The final principle is that
remedies should be ‘incentive compatible’, in terms of compliance by the desig-
nated SMP operator rather than evasion.
The fourth harmonization element in the Framework Directive concerns the no-
tification regime, whereby an NRA is required to notify the Commission, BEREC,
and the other Member State NRAs about measures it makes in respect of the SMP
process. Two distinct procedures exist: the first applicable to NRA decisions on
market definitions and whether to designate an operator as having SMP (Article 7);
the second concerning decisions on the imposition of remedies (Article 7a), intro-
duced under the 2009 Reforms. In both cases, comments may be submitted to the
notifying NRA on the draft measures, which the NRA is obliged to take ‘utmost
account’ of (Articles 7(7) and 7a(1) respectively). The legal nature of such com-
ments has been subject to challenge by operators dissatisfied with the impact they
have had, specifically those of the Commission, on subsequent NRA decisions. The
Court of First Instance ruled that such comments did not have a binding effect
and, therefore, could not be challenged under Article 263 of the TFEU.166 In add-
ition, both procedures grant the Commission an exclusive right to issue a stand-
still notification in respect of a draft measure (Articles 7(4) and 7a(1)), of two and
three months duration respectively.
The significant distinction between the two procedures lies in the power of the
Commission to require an NRA to subsequently amend or withdraw a decision,
where it is considered to create a barrier to the single market or be incompatible
with Community law. The Commission has such a veto power in respect of market
definition and designation decisions (Article 7(5)(a) and (6)), but not in respect of
decisions regarding the imposition of remedies (Article 7a(7)). The reason for the
differential treatment lies in the lack of competence that the Commission has to
interfere with remedies under national law. With regard to the former procedure,
the Commission has to date exercised its veto power on only thirteen occasions;
although NRAs generally withdraw decisions that have been challenged by the
Commission rather than have them formally vetoed.167 The Commission may only
veto a draft NRA decision where it considers that it would ‘create a barrier to the
single market or if it has serious doubts as to its compatibility with Community
law’ (Article 7(4)). As such, Commission approval is also confined to the absence
166
Case T-109/0 6—Vodafone (12 December 2008) and Case T- 0 6 Base NV v Commission (22
295-
February 2008).
167
There have been 105 withdrawals. For an overview of notifications, as of January 2018, see <https://
circabc.europa.eu/faces/jsp/extension/w ai/navigation/container.jsp>.
718
168
eg British Telecommunications plc v Ofcom [2017] CAT 17.
169
Commission Recommendation 2008/850/EC ‘on notifications, time limits and consultations provided
for in Article 7 of Directive 2002/21/EC’, OJ L 301/2 3, 12 November 2008, which replaced Recommendation
2003/561/EC (OJ 190/13, 30 July 2003).
170
Since 1 May 2004, jurisdiction is shared with Member States: Council Regulation No 1/2003 on the im-
plementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ L1/1, 4 January 2003.
719
One of the central features present in the Member States prior to liberalization
of the telecommunications market was the fact that the regulatory institution re-
sponsible for regulating the market, often a Ministry of Communications, was
usually also responsible for controlling the commercial activities of the incum-
bent operator. It was recognized that such merged functions would not be appro-
priate in a competitive market and that independent regulatory authorities for the
sector would need to be established.
The ‘Services Directive’ reiterated the need for Member States to ensure that ‘a
body independent of the telecommunications organisations’ carried out the regu-
latory functions.173 What this formulation does not adequately address is the issue
171
Commission Directive (88/301/E EC) of 16 May 1988 on competition in the markets of telecommunica-
tions terminal equipment; OJ L131/73, 27 May 1988, at Art 6. This position had previously been taken by the
Court of Justice in GB-Inno-BM, see n 100.
172
See Thierry Tranchant and Téléphone Store SARL [1995] Case C-91/94, ECR I-3911, [OJ 96/16/6]. See also
Procureur du Roi v Lagauche & Others, Evrard [1993] Cases C-4 6/9 0 and C-93/91, ECR I-5267, [OJ 93/C316/3];
Ministere Public v Decoster [1993] Case C-69/91, ECR I-5335, [OJ 93/C332/7]; Ministere Public v Taillandier-Neny
[1993] Case C-92/91, ECR I-5383, [OJ 93/C338/6].
173
Commission Directive (90/388/E EC) of 28 June 1990 on competition in the markets for telecommunica-
tions services; OJ L192/10, 24 July 1990, at Art 7.
810
174
Council Directive 92/4 4/E EC, of 5 June 1992, on the application of open network provision to leased
lines, OJ L165/27, 19 June 1992, at Recital 14.
175
See Council Directive 95/62/EC, of 13 December 1995, on the application of open network provision to
voice telephony, OJ L321/6, 30 December 1995, at Art 2(2).
176
Directive 97/51/EC of the European Parliament and of the Council of 6 October 1997 amending Council
Directives 90/387/E EC and 92/4 4/E EC for the purpose of adaptation to a competitive environment in
Telecommunications, OJ L 295/2 3, 29 October 1997: at Art 1(6), inserting Art 5a into Directive 90/387/E EC.
81
operators and are involved in the granting of rights of way.177 In the UK, such a pro-
vision would have been applicable to Hull City Council, which had a controlling
shareholding in Kingston Communications until 2007.
Another aspect of the position of any regulatory authority is that such a body
must be given the resources to carry out its assigned tasks. The effectiveness of
a regulator depends to a considerable degree on the resources made available to
it. This issue was indirectly addressed through the recitals of some of the ONP
measures. Initially reference is simply made to an authority having ‘the necessary
means to carry out these tasks fully’;178 although this was subsequently elaborated,
whereas the national regulatory authorities should be in possession of all the re-
sources necessary, in terms of staffing, expertise, and financial means, for the per-
formance of their functions.179
To meet this objective, the NRA must either look to government or the regulated
industry for the necessary resources. In an era of public sector spending restraint,
sufficient resources from government must always appear doubtful. In terms
of the providers of telecommunications networks, equipment, or services, one
source of income is through the operation of the licensing regime. However, under
the Authorisation Directive, NRAs are only permitted to charge fees that cover ‘the
administrative costs which will be incurred in the management, control and en-
forcement of the general authorisation scheme’ and related matters (Article 12(1)
(a)), effectively a form of cost-accounting obligation placed on the regulator rather
than the regulated, which clearly emphasizes the need to minimize the costs of
regulation.
Member States have adopted a diversity of models in establishing regulatory
institutions, some granting regulatory tasks to the national legislature,180 while
others disperse regulatory tasks among a number of separate institutions, which
is seen as significantly weakening the exercise of such powers. Regulatory de-
pendency on the incumbent for the provision of information, as well as expertise,
continues to be perceived as a problem by some new entrants in a number of jur-
isdictions. In terms of resources, the main reported problem is the retention of
staff in such a fast moving well-remunerated employment market, which can lead
to over-reliance on seconded personnel from operators including the incumbent.
180
See Case C-389/08, Base NV and others v Ministerraad (6 October 2010), where it was held that a determin-
ation that the provision of universal service was an ‘unfair burden’ for a designated undertaking (see further
Section 4.8) could be made by the national legislature, provided it met the ‘requirements of competence, in-
dependence, impartiality and transparency’ stipulated in the Framework and Universal Services Directives.
812
181
See ‘The 1999 Communications Review’, see n 10, at section 4.8.3.
182
Framework Directive, at Art 3(2).
183
Ibid, at Chapter III, ‘Tasks of National Regulatory Authorities’. In Case C-424/07, Commission v Germany
(3 December 2009), it was held that German law that excluded certain ‘new’ markets from regulation was an
unlawful limitation of the NRA’s discretion.
184
eg see 13th Implementation Report, at p 10 et seq.
185
Framework Directive, at Art 3(3). In Case C-2 40/15, AGC v ISTAT (28 July 2016), it was held that this ob-
ligation does ‘not preclude . . . provisions for limiting and streamlining the spending of public administrative
authorities’.
186
Framework Directive, at Art 3(3a). In Case C-560/15, Europa Way v AGCOM (26 July 2017), it was held that
annulment by the Italian legislature of a selection procedure for radio frequencies being carried out by the
NRA was precluded.
187
Framework Directive, at Art 3(3a). In Case C-424/15, Garai v Administración del Estado (19 October 2016),
it was held that Art 3(3a) precluded dismissals from the NRA that resulted from a merger of regulators without
rules designed to protect the independence of the NRA.
188
Framework Directive, at Art 3(4).
813
189
Communications Act 2003, s 370 (functions under Part 4 of the Enterprise Act 2002) and s 371 (functions
under the Competition Act 1998).
190
ie Privacy and Electronic Communications (EC Directive) Regulations 2003, at r 33. See further
Chapter 13.
191
Framework Directive, Art 4. 192
2006 Review, Staff Document, see n 28 at 5.3.2.
193
Proposed Directive amending the Framework Directive, at Art 2(4).
194
See, for example, Order of the President of the Court of First Instance of 30 April 1999 [1999] ECR II-1427.
195
See R v Director General of Telecommunications (Respondent), ex parte Cellcom, [1999] ECC 314, with re-
spect to reconciling the principles contained in the Telecommunications Act 1984, s 3(2).
814
A final aspect of NRA responsibility concerns their role in intervening and re-
solving disputes between market participants. Under pre-NRF, NRAs were re-
quired to make decisions in respect of disputes between undertakings, such as
interconnection arrangements. However, the speed of NRA decision-making is
seen as a potential barrier to entry in some jurisdictions. Inexperience, insuffi-
cient powers, and appeal procedures often resulted in significant delays, which
usually disadvantaged the market entrant. The Framework Directive therefore im-
poses an obligation upon NRAs to reach a binding decision within four months.196
The centrality of Member State NRAs in the regulation of the electronic com-
munications sector continues to be a defining feature of EU law and regulation.
National divergences in NRAs as institutions and personalities would seem an in-
evitable outcome of the unique historical, political, and juridical characteristics of
the various Member States; as much as they are a result of market differences in each
national market. However, the expression of these differences impacts on the real-
ization of a single market for the electronic communications sector and, as such,
is the concern of the Commission. Striking a balance between independent NRAs
and a harmonized EU regulatory approach remains an ongoing challenge.
196
Framework Directive, at Art 20.
197
Report by NERA and Denton Hall, ‘Issues Associated with the Creation of a European Regulatory
Authority for Telecommunications’ (March 1997); also ‘Report on the value added of an independent
European Regulatory Authority for telecommunications’ (September 1999).
198
Speech of Viviane Reding, ‘From Service Competition to Infrastructure Competition: the Policy Options
Now on the Table’ at ETCA Conference, Brussels, 16 November 2006.
815
199
Proposal for a Regulation of the European Parliament and of the Council establishing the European
Electronic Communications Market Authority, COM(2007)699 rev 2.
200
Commission Decision 2002/ 627/
EC establishing the European Regulators Group for Electronic
Communications Networks and Services, OJ L 200/38, 30 July 2002.
201
See n 39, at Recital 6.
202
Ibid, at Art 3(1). BEREC was given additional tasks to draft guidelines for the implementation of the obli-
gations of NRAs on open internet access, under Regulation 2015/2120, at Art 5(3).
203
Proposal for a Regulation establishing the Body of European Regulators for Electronic Communications,
COM(2016) 591 final, 14 September 2016.
204
Established under Directive 90/387, Art 9, and Directive 97/13/EC on a common framework for general
authorizations and individual licences in the field of telecommunications services, OJ L 117, 7 May 1997, Art
14, respectively.
205
Established by the Commission under Council Resolution of 17 December 1992 on the assessment of the
situation in the Community telecommunications sector, OJ C 2/5, 6 January 93.
816
4. 8 UNIV ER S A L SERV IC E
One key area of ongoing concern of Member States towards the policy of market lib-
eralization has been the ability to preserve and pursue the potentially conflicting
public policy objective of ‘universal service’: the provision of access to telecommu-
nications services for all the state’s citizens. In many jurisdictions, the belief that
the telecommunications market was one of natural monopoly was closely allied
with this need to ensure ‘universal service’.
Article 106(2) of the TEC recognizes that undertakings may be entrusted ‘with
the operation of services of general economic interest’ and that the competition
rules may be not be applicable to such undertakings where they ‘obstruct the
206
Framework Directive at Art 22. See further <https://ec.europa.eu/d igital-single-market/en/
communications-committee>.
207
Decision No 676/2002/EC of the European Parliament and of the Council on a regulatory framework for
radio spectrum policy in the European Community, OJ L 108/1, 24 April 2002, at Art 3. See further <https://
ec.europa.eu/d igital-single-market/en/radio-spectrum-committee-rsc>.
208
Commission Decision 2002/622/EC establishing a Radio Spectrum Policy Group, OJ L 198/49, 27 July
2002, as amended by Commission Decision 2009/978/E U, OJ L 336/50, 18 December 2009. See further <http://
rspg-spectrum.eu>.
209
Directive 99/5/EC, at Art 13. 210
See n 172, at 3.1. 211
See <https://etno.eu>.
817
performance, in law or in fact, of the particular tasks assigned to them’: the so-
called ‘public service defence’.212
The initial liberalization process envisaged under the 1987 Green Paper was
not seen as greatly disturbing the policy of universal service, since the provision
of voice telephony (as a ‘reserved service’) and network infrastructure remained
with the national incumbent operator. However, the issue came to the forefront
of EU telecommunications policy with the Commission’s 1992 telecommunica-
tion review, which proposed extending the liberalization process from services to
network infrastructure.213 The endorsement of this policy by the Member States
was therefore qualified by the need to protect universal service, as noted by the
European Parliament:
. . . the process of liberalization has to be accompanied by maximum protection of
the universal service . . . especially that of weaker consumers and that of periph-
eral and disadvantaged countries and regions.214
212
See Taylor, SM, ‘Article 90 and telecommunications monopolies’, (1994) 15(6) European Competition Law
Review 332 et seq.
213
Commission Communication to the Council and European Parliament, ‘1992 Review of the situation in
the telecommunications services sector’, SEC(92) 1048, 21 October 1992.
214
European Parliament Resolution of 20 April 1993 on the Commission’s 1992 review of the situation in the
telecommunications services sector; OJ C 150/39, 31 May 1993.
215
Commission Communication to the Council and the Council and European Parliament, ‘Developing
universal service for telecommunications in a competitive environment’, COM(93) 543, 15 November 1993.
81
216
Universal Services Directive, Art 3(1).
217
The provision of an Integrated Services Digital Network (ISDN) connection is expressly excluded from
the concept of the universal service ‘connection’ obligation (Recital 8). Under the pre-2003 regime, Germany
included such connections within its USO regime.
218
2016 Proposal, at Part III, Title I. Member States would have the option to retain these legacy services, if
the need could be demonstrated.
819
The Universal Services Directive also provides for a process of periodic review of
the scope of ‘universal service’, to be carried out by the Commission. Reviews were
carried out in 2005,219 2008,220 2011,221 and in 2015 as part of the 2016 Proposal. The
2009 Reforms, however, contained no significant amendment to the definition.
The reviews consider a range of factors, such as whether the majority of consumers
use the specific service and whether ‘non-use by a minority of consumers result in
social exclusion’ (as provided for at Annex V). As noted already, the scope is likely
to expand soon, to reflect the status of the internet as the ubiquitous communica-
tions platform, although the entry of the twelve Accession States delayed some-
what the raising of the threshold.
In respect of the second element of USO, quality, Member States must ensure
that all designated operators publish information regarding their performance
against certain parameters (Article 11(1)), addressing such matters as the supply
time for initial connection, fault repair time, and complaints concerning the
correctness of bills (Annex III). NRAs may also set additional quality of service
parameters in respect of the provision of services to disabled end-users and con-
sumers (Article 11(2)). NRAs may set and monitor performance against certain tar-
gets, with the right to take measures where an operator persistently fails to meet
such targets (Article 11(4)–(6)). These measures are supplemented by the general
consumer-related measures in Chapter IV of the Universal Service Directive,
which impose transparency obligations on operators (Article 22) and, following
the 2009 Reforms, the ability for an NRA to set minimum requirements (Article
22(3)).222
In terms of ‘affordability’, the cost of access is as critical an element as the ac-
tual provision of a connection. Under the Universal Services Directive, NRAs may
require designated operators to offer tariff options or packages targeted specific-
ally at those on low incomes or with special social needs (Article 9(2)). In addition,
common tariffs, such as geographic averaging, may be imposed, or price caps
(Article 9(3)–(4)). In reality, geographic averaging was a traditional mechanism for
funding the USO, which has been retained in all Member States.
The NRAs have the right to designate which operators are required to ensure
provision of the ‘set’ of services (Article 8(1)). While in most Member States the
obligation will primarily lie with the incumbent operator, as markets become fully
competitive USO may be imposed on a number of operators, including the provi-
sion of different service elements by different operators in different geographical
219
Commission Communication ‘on the review of the scope of universal service in accordance with Article
15 of Directive 2002/22/EC’, COM(2005) 203, 24 May 2005.
220
COM(2008) 572 final, 25 September 2008. 221
COM(2011) 975 final, 23 November 2011.
222
See further Chapters 9 and 15.
910
223
See further Chapter 8 at Section 8.4.2.
224
Directive 96/19/EC, Art 6, inserting Art 4c into Directive 90/388/EC. 225
47 USC §254(d).
91
The term ‘universal service’ is supposed to have been originally coined by Theodore
Vail, Chairman of AT&T, in 1907;228 although the concept he was promoting was
that of universal interconnection, rather than universal access. However, there is
an important relationship between network interconnection and the promotion of
universal service. If an operator is providing elements of a universal service policy,
such as full national network coverage, and also has an obligation to interconnect
to any new entrant operator, then the former operator may be placed in a disad-
vantageous competitive position. In the absence of a regulatory obligation to pro-
vide such services, the operator would inevitably withdraw from the provision of
any uneconomic universal service elements. This connection was recognized by
the Council in its 1994 Resolution on universal service,229 and was given explicit
recognition in the Interconnection Directive.230
Under the Interconnection Directive, where a Member State determined that
meeting any universal service obligations represents an unfair burden upon an
operator, the Member State could establish a mechanism to share the net cost.
However, new entrants inevitably have concerns that any compensation mech-
anism may operate as a barrier to market entry, benefiting the incumbent. Calls
have therefore been made for the cost of universal service, as a social policy
226
Case C-320/91 Corbeau [1993] ECR I-2533, at para 17 et seq.
See Sixth Implementation Report, see n 103, at p 27. See further Chapter 2, at Section 2.11.
227
228
Stated by Garnham, N, ‘Universal Service’, Melody (ed), Telecom Reform (Technical University of
Denmark, 1997) at 207.
229
Council Resolution of 7 February 1994 on universal service principles in the telecommunications sector,
at ‘Recognises’ (e).
230
See n 144.
912
231
USD, at Art 13(1).
232
See, for example, Ofcom Statement, Review of the Universal Service Obligation, March 2006, at p 3.
233
See 15th Implementation Report, COM(2010) 253 final/3, 25 August 2010, at p 13.
234
Universal Services Directive, at Art 12 and Annex IV, Part A.
235
See further Chapter 2, at 2.12.2.
913
To date, Member State experience would not appear to reflect the historic con-
cern shown towards the threat posed by a competitive market to the provision
of universal service. Instead, the perception of universal service provision is in
the process of being transformed from a burden into an opportunity for market
players.
This chapter has attempted to examine the development of European Union com-
munications law over the past thirty years. As the third distinct phase of develop-
ment, the 2003 Regime inevitably raises the question whether it will be the final
phase of regulatory evolution or whether a fourth, fifth, or even sixth phase can
be envisaged.
The NRF embodies a range of different regulatory initiatives. Perhaps the most
significant and revolutionary of which is the idea that a single regulatory regime
or framework should govern all forms of communications infrastructure and
services, irrelevant of the content being communicated. Such an idea is based on
current technological and market developments, generally referred to as conver-
gence, which, although reflecting reality to an extent, also anticipates a process
that has a long and unpredictable way to go. A truly converged environment may
enable the removal of certain legacy regulatory concepts, such as the ‘must-carry’
obligation in relation to broadcasting, and yet it may require others to be extended,
such as the scope and nature of universal service.236 In addition, as the provision of
network becomes a commodity, bundled into the cost of the content being trans-
mitted, the bright line between carriage and content may become either more
problematic or an irrelevant or meaningless regulatory distinction.
A second objective of the NRF was to move from ex ante regulatory intervention
towards ex post reactive regulation. The rationale being that with the successful
introduction of competition, traditional market mechanisms will control anti-
competitive practices, with traditional competition law rules operating as a back-
stop against abusive practices and situations. This is the model that operates in
the information technology sector, whether successfully or not, and was viewed
as an inevitable consequence of both liberalization and convergence. However,
during the consultation on the 1999 Review, new entrants made it very clear that
the current market was not yet sufficiently competitive and was unlikely to be in
certain market segments for some time to come, if ever. Hence the NRF continues
The 2016 proposal retains ‘must-c arry’, but does amend the USO.
236
914
US TELECOMMUNIC ATIONS L AW
Karen Lee and Jamison Prime
5.1 INTRODUC TION
Evolution of the Bell network Until its patents expired in 1893–4, the Bell
Company (Bell)— t he company founded by Alexander Graham Bell, the in-
ventor of the telephone, and later known as AT&T—monopolized the telephony
market. Thereafter competition flourished for about a decade. At one stage, inde-
pendent companies provided up to half of the telephone stations in local areas.
However, absent an obligation for Bell to interconnect with the independents,
Pub L No 104–104, 110 Stat 56 (8 February 1996) (codified at 47 USC §§251–261, 271–276, 336, 363, 571–573,
2
549, 613, 160–161, 660–561, 230, 232, 614, and at 15 USC §79z–5c).
917
5 US Telecommunications Law 197
the competition which existed was inefficient. It was not unusual for businesses
to subscribe to two or more local telephone networks utilizing separate lines and
equipment. Bell, meanwhile, successfully reorganized into a vertically integrated
company, making use of its patented technology that significantly improved the
sound quality of long-d istance calls, and began to acquire independent phone
companies at a rapid pace.
Initial federal regulation In response to calls to halt the rise of AT&T and cease
the unnecessary duplication of infrastructure, Congress enacted the Mann-Elkins
Act 3 in 1910, which marked the start of telephony regulation at the federal level.
Prior to the Mann-Elkins Act, the states were largely responsible for it. Modelled
on the Interstate Commerce Commission Act of 1877, which governed railroads,
the Act gave the Interstate Commerce Commission (ICC) the power to regulate
telecommunications providers as ‘common carriers’,4 which were required to ‘pro-
vide service on request at just and reasonable rates without unjust discrimination
or undue preference’. The ICC had the power to set aside ‘unjust and unreasonable’
rates of common carriers providing communications services but lacked the au-
thority to require them to file tariffs or interconnect their networks, thus greatly
hindering its ability to regulate them effectively.
Long-distance competition For decades after passage of the 1934 Act, the FCC
allowed AT&T to retain its monopoly on telecommunications in order to secure the
goal of securing a ‘rapid, efficient, Nation-w ide and worldwide . . . communication
3
Mann-E lkins Act (18 June 1910, ch 309, 36 Stat 539).
4
See Section 5.8.1 for the meaning of ‘common carrier’.
918
Divestiture of AT&T and the modification of final judgment In 1974, the
Department of Justice brought a suit against AT&T, Western Electric, and Bell
Telephone Laboratories, Inc alleging that the monopolies held by the defendants
in several telecommunications service areas and equipment manufacturing vio-
lated the Sherman Anti-t rust Act.5 The case was pending for eight years until Judge
Harold Greene of the DC Circuit Court entered the Modification of Final Judgment
(MFJ), which slightly modified divestiture provisions voluntarily agreed to by the
parties (US v AT&T Corp, 552 F Supp 131 (DC Cir 1982), aff’d sub nom Maryland v
US, 460 US 1001 (1983)). The MFJ ordered AT&T to divest itself of its twenty-t wo
Regional Bell Operating Companies (RBOCs), which resulted in the separation of
local and interexchange (long-d istance) markets, and established procedures for
the implementation of divestiture.
Under the provisions of the MFJ, the twenty-two RBOCs, which by 2011 had
been consolidated into three main holding companies—AT&T, CenturyLink, and
Verizon—would provide communication in ‘exchange areas’ (also known as local
access and transport areas (LATAs)). Exchange areas referred to a geographic area
5
Sherman Anti-t rust Act, ch 647, 26 Stat 209, 209–10 (1892) (codified as amended at 15 USC §§1–7).
91
5 US Telecommunications Law 199
that encompassed one or more contiguous local exchange areas serving common
social, economic, and other purposes. Within these exchange areas, RBOCs could
originate and terminate calls but were prohibited from providing interexchange
telecommunication services and information services (see Section 5.2.5). The ori-
ginal settlement provisions sought to restrict the manufacture and provision of
customer premises equipment but the MFJ allowed RBOCs to market equipment
once they divested from AT&T.
The MFJ also sought to ensure that all interexchange service providers (eg MCI and
Sprint) obtained equal access to RBOC services. The judgment imposed a duty on
local exchange carriers to provide service on an ‘unbundled, tariffed basis’ that was
equal in quality, type, and cost to that provided to AT&T and its affiliates. In addition,
RBOCs were prohibited from discriminating against other service providers in fa-
vour of AT&T in the following areas: procurement, establishment and dissemination
of technical information, interconnection standards, interconnection, and provision
of new services and facilities.
overturned the MFJ provisions6 that allowed RBOCs to retain monopolies in the
lucrative local market. It also removed the MFJ’s restrictions on the provision of
interstate telephony services by RBOCs, provided they comply with a 15-point
‘competitive’ checklist to the satisfaction of the FCC.7 This checklist included a
number of resale, access, and interconnection obligations, including the pro-
vision of non-d iscriminatory access to unbundled network elements, which are
discussed in Section 5.10.1. In December 1999, Bell Atlantic was the first RBOC
permitted by the FCC to offer interstate telephony services, provided it met certain
conditions in the State of New York and complied with other specified safeguards.8
By 2003, all RBOCs had received approval to provide long-d istance services in all
of their regional areas. Since then, there has been significant consolidation in the
sector and a return to vertical integration brought about by declining revenue for
long-d istance services and high prices for local access. Despite the mergers, RBOCs
owned by Verizon, AT&T, and CenturyLink providing interstate telephony services
must continue to comply with the competitive checklist and other requirements.9
5.2.2 Cable
Since its inception in the 1950s, cable television has evolved from a simple video
transmission service to an important provider of broadband services. The FCC
has applied varying degrees of regulation to this medium throughout this evolu-
tion. This section focuses on cable’s historical development and legacy position
as the original multichannel video programming distributor (MVPD). The regula-
tion of cable’s broadband offerings is addressed in greater depth elsewhere in this
chapter.
Individual cable systems have traditionally been local in nature due to their
design, although industry consolidation and regulatory changes have made this
characteristic increasingly less important. Typically, a cable ‘head end’ facility re-
ceives terrestrial and satellite broadcast signals via a series of antennae and dishes.
These signals are then transmitted via wire throughout the community the cable
provider serves, usually on telephone poles or along streets. An individual cable
line runs to each subscriber. While cable companies are subject to competition
for the delivery of video programming from fibre and satellite providers and are
themselves robust competitors in the fixed broadband market, cable companies
6
The MFJ was officially terminated on 11 April 1996 following the enactment of the 1996 Act.
7
See 47 USC §271. 8
See 47 USC §272.
9
For further discussion of the obligations on RBOCs, see Section 272(f)(1) Sunset of the BOC Separate
Affiliate and Related Requirements, Report and Order and Memorandum Opinion and Order, WC Docket No
02–122, CC Docket No 00–175, WC Docket No 06–120, FCC No 07–159, 22 FCC Rcd 16440 (2007).
201
5 US Telecommunications Law 201
traditionally have respected each other’s exclusive franchise areas with respect to
the delivery of video programming.
Cable Communications Policy Act of 1984, Pub L No 98–5 49, 98 Stat 2779 (1984).
10
services. To that end, the 1996 Act removed restrictions that had limited telephone
companies from providing cable services, while concurrently, over a three-year
period, phased out many of the cable rate regulations adopted in 1992.
Cable entities traditionally have been required to obtain and periodically renew
a local franchise licence in order to operate. A franchise benefits the cable provider
by permitting access to public rights of way, as well as the local franchise authority,
which can set renewal standards and charge the cable operator a fee for the right
to operate its system in the designated area. However, the scope of what local au-
thorities can regulate has generally diminished over time. The 1996 Act imposed
limitations on local and state regulation of cable, such as the prohibition on grants
of exclusive franchises or unreasonably withholding consent for a new service.
In addition, some states have allowed cable and telecommunications entities to
obtain a single franchise from the state, bypassing the need to negotiate indi-
vidual local agreements. In 2005, the FCC released a then-controversial Report
and Order,12 that found evidence that local authorities had acted unreasonably to
delay the entry of new competitors. In the Report and Order, it established rules
regulating how local franchise authorities can act by, among other things, setting
strict time limits for local governments to act on new applications to provide video
services. In 2015, the FCC found that cable companies face ‘effective competition’
nationwide.13 As a result, local cable franchising authorities can no longer regulate
the rates of basic cable services and equipment unless they overcome a rebuttable
presumption that the particular local market is competitive. Previously, there was
a rebuttable presumption that cable operators were not subject to effective com-
petition, which resulted in numerous petitions to the FCC from cable operators
(or other interested parties) seeking case-by-case determinations that a particular
franchise area contained a sufficient quantity of consumer options to be deemed
competitive.14
Cable’s once dominant role in the bundling and delivery of video content con-
tinues to be diminished by new entrants and challenged by new technologies. In
response, cable providers have attempted to maintain their role by, for example,
deploying a nationwide network of WiFi hotspots and promoting access to sub-
scription video anywhere from any device. In addition, Comcast, the largest cable
12
Implementation of Section 621(a)(1) of the Cable Communications Policy Act of 1984 as amended by the
Cable Television Consumer Protection and Competition Act of 1992, Report and Order and Further Notice of
Proposed Rulemaking, MB Docket No 05–311, FCC No 06–180, 22 FCC Rcd 5101 (2007) (upheld in Alliance for
Community Media v FCC, 529 F 3d 763 (6th Cir 2008)).
13
Concerning Effective Competition; Implementation of Section 111 of the STELA Reauthorization Act, 80
Fed Reg 38001 (2015) (upheld in Nat’l Ass’n of Telecom. Officers and Advisors v FCC (DC Cir 2017)).
14
Implementation of Sections of the Cable Television Consumer Protection and Competition Act of 1992
Rate Regulation, Report and Order, 8 FCC Rcd 5631, 5669–5670 (1993).
230
5 US Telecommunications Law 203
operator, entered the video content market directly through its purchase of NBC
Universal from General Electric, which was approved in 2011. The industry con-
tinues to see consolidation including, most recently, the merger of Time Warner
Cable and Bright House Networks with Charter into a single entity that markets
itself under the ‘Spectrum’ brand. Today, MVPDs are a mix of traditional cable op-
erators; Direct Broadcast Satellite (DBS) systems that allow consumers to receive
video via pizza box-size dish antennae; fibre-to-the-home services provided by
traditional telephone companies; and to a lesser extent, competitive ‘overbuilders’
such as RCN and new technology companies such as Google. Premium subscrip-
tion internet video services such as those provided by Hulu, Roku, and Apple TV,
as well as original content from entities such as Amazon and Netflix, are also chal-
lenging the traditional cable model. As the cost of cable programming continues
to rise due to increased fees for retransmission, particularly for sports program-
ming, consumers appear to be more and more willing to embrace substitutes and
drop their traditional cable television subscriptions. This evolving landscape can
be illustrated by the experiences of AT&T. It had previously operated a traditional
cable franchise model through AT&T Broadband, but sold that division to Comcast
in 2002. However, by 2017 it had once again become a major MVPD through a com-
bination of fibre-to-t he-home, satellite television (it purchased DirecTV in 2015),
and internet television subscription services. It counted 30 million content sub-
scribers in a competitive marketplace, despite the fact that it no longer held any
traditional cable assets.
Throughout the evolution of cable, the courts have generally upheld efforts to
regulate the medium. In 1968, the Supreme Court acknowledged the FCC’s right to
regulate cable, concluding that it was ‘interstate commerce by wire or radio’ sub-
ject to the FCC’s authority under the broad provisions of the 1934 Act (United States
v Southwestern Cable Co, 392 US 157 (1968)). Although cable providers are akin to
broadcasters and newspapers, in that they select programming for distribution,
they are also similar to common carriers in that they mostly transmit, unaltered,
content originated by third parties. Courts have been deferential to cable regula-
tion, but have been unwilling to afford the types of First Amendment protection for
regulation offered to newspapers and, to a lesser extent, broadcasters.
The FCC continues to have statutory obligations that relate to cable’s traditional
role as a video provider. Recent regulatory decisions have involved the programme
access rules, which are designed to ensure that all MVPDs have access to cable-
owned programme networks, and retransmission consent, which relates to how
cable providers and television stations negotiate for the carriage of over-the-a ir
broadcast channels. Notwithstanding recent developments in broadband, these
and other video issues will likely remain a significant component of the US cable
regulatory practice.
024
5.2.3 Wireless
The regulation of radio communications has long been a part of the FCC’s mission,
and government interest in this area can be traced back to before the agency’s
founding. Similarly, the development of commercial wireless telephone systems
can trace its roots to the development of cellular networks in 1947. Under the cel-
lular concept, the use of geographically small service areas (cells) allows a limited
number of frequencies to be reused across a larger geographic area, which in turn
increases the capacity of a mobile network to process a large number of telephone
calls using relatively few frequencies. At the time the cellular system was envi-
sioned, however, the technology did not exist to deploy widespread wireless net-
works. From 1947 until 1968, the FCC sharply limited the number of frequencies
available for cellular-t ype telephone operations, and thus there was little research
or development in the area.
5 US Telecommunications Law 205
15
See generally Redevelopment of Spectrum to Encourage Innovation in the Use of New Telecom
munications Technologies, ET Docket No 92–9.
16
These reports are docketed under the caption ‘In the matter of Implementation of Section 6002(b) of the
Omnibus Budget Reconciliation Act of 1993; Annual Report and Analysis of Competitive Market Conditions
With Respect to Commercial Mobile Services’ and are maintained on the FCC’s website at <http://w ireless.
fcc.gov/i ndex.htm?job=cmrs_ reports>.
026
rural areas who are only just completing initial roll-outs or are transitioning from
early-generation technology best suited for voice-based communications.
5 US Telecommunications Law 207
incumbent wireless carriers due to their short propagation distances and greater
susceptibly to disruption. Use of such spectrum is expected to result in the deploy-
ment of dense networks consisting of clusters of low-power small cells with ad-
vanced beam-forming technologies, especially in urban areas, where such small
cells can be readily deployed within buildings and on light poles and other public
infrastructure.
Mobile telephone services do not have the long history associated with fixed-
line services, and have operated under a relatively simpler regulatory structure.
US policy has focused on fostering robust competition in this space, and regu-
lators have relied on vigorous competition among licensees to deliver service im-
provements and ensure reasonable prices. As a result, there has been relatively
limited involvement in setting pricing or service quality standards. Instead, the
FCC has focused much of its attention on the amount of spectrum a particular en-
tity controls, and has employed various means to ensure that no particular entity
becomes so dominant as to threaten the competitive environment. One notable
exception to this rule involves net neutrality. When the FCC adopted net neutrality
rules in 2015, it applied the rules equally to fixed and mobile broadband networks
despite strong objections from the wireless industry.18
As wireless telephony transitions to wireless broadband, services that have not
been traditionally considered part of the mobile telephone service are more easily
integrated into a broadband network. Spectrum previously set aside for educa-
tional broadcast purposes, recovered from television broadcasting, and repur-
posed from satellite use have all been used to expand mobile broadband networks.
Because many of these services have been licensed under different schemes, regu-
lators have revisited existing allocations and service rules to remove impediments
to flexible spectrum use. While the mechanisms for reallocating or repurposing
spectrum continue to evolve, it has become increasingly difficult to identify users
that can be easily relocated and suitable spectrum in which to relocate their
services. Accordingly, spectrum policy increasingly focuses on how to promote
successful band sharing among different and traditionally incompatible services.
5.2.4 Satellite
The provision of domestic and international communications services by satel-
lite in the US has increased dramatically since the 1960s. Historically, domestic
satellite services in the US were provided by private entities; however, inter-
national satellite communications services in the US were offered exclusively by
18
See further Section 5.2.5.3 below.
028
19
Pub L No 87–624, 76 Stat 419 (1962). 20
Presidential Determination No 85–2 .
21
Pub L No 106–180, 114 Stat 48 (2000).
22
Inmarsat was privatized on 15 April 1999, prior to the enactment of ORBIT; Intelsat was privatized on 18
July 2001. See further Chapter 16, at Section 16.2.1.2.
029
5 US Telecommunications Law 209
the provision of public safety services during times of emergency when terrestrial
networks may be unavailable; and the provision of telephony services where trad-
itional communications networks are not present, such as in wilderness areas and
aboard yachts and other vessels. Most recently, there has been growing interest
in developing and deploying networks that consist of constellations of thousands
of satellites that would operate in low-earth orbits. Because such satellites would
reside much closer to the surface of the earth than many traditional satellites,
these new networks would allow for high bandwidth transmissions with minimal
latency. With the advent of small satellite forms that can be produced quickly, rela-
tively inexpensively, and in large quantities; increased competitive commercial
launch options offered by companies such as SpaceX; and technological improve-
ments in satellite communications, these once audacious plans now appear more
practical.
5.2.5 Broadband and IP
The deployment of broadband services and the IP technology used to provide
them over the last decade has created a number of regulatory difficulties for the
FCC. One of the most contentious issues has been whether broadband services
should be classified as ‘telecommunications services’ or ‘information services’. If
these services are ‘telecommunications services’ they are, absent a decision of the
FCC to forbear from regulation, regulated in accordance with the obligations of
Title II of the 1934 Act including tariff notification, access, and interconnection.23
If they are properly classified as ‘information services’, they are subject to the rules
(if any) adopted by the FCC exercising its ‘ancillary jurisdiction’ set out in Title
I—t he power to ‘perform any and all acts, make such rules and regulations, and
issue such orders, not inconsistent with [the 1934] Act, as may be necessary in the
execution of its functions’. The 1996 Act defines the terms ‘telecommunications
services’—the ‘offering of telecommunications24 for a fee directly to the public,
or to such classes of users as to be effectively available directly to the public, re-
gardless of the facilities used’25—a nd ‘information services’—‘the offering of a
capability for generating, acquiring, storing, transforming, processing, retrieving,
utilizing, or making available information via telecommunications, and includes
electronic publishing, but does not include any use of any such capability for the
23
See Sections 5.8 and 5.10.
24
‘Telecommunications’ is defined as ‘the transmission, between or among points specified by the user,
of information of the user’s choosing, without change in the form or content of the information as sent and
received’. 47 USC §153(43).
25
47 USC §153(46).
201
47 USC §153(20).
26
21
5 US Telecommunications Law 211
enhanced services, provided they comply with specified cost-a llocation methods
and targeted regulations designed to prevent RBOCs from abusing their market
power in basic services. Initially, RBOCs were expected to comply with ‘compar-
ably efficient interconnection’ (CEI) requirements. In the longer term, RBOCs had
to comply with certain Open Network Access (ONA) obligations that required
them to unbundle their basic services into ‘basic service elements’ for purchase by
enhanced service providers. In addition, quality, installation, and maintenance
reporting requirements were imposed. Like other carriers, RBOCs had to offer the
basic services used in their enhanced service offerings pursuant to tariff and on a
non-d iscriminatory basis. Because of procedural errors some aspects of the FCC’s
decision in Computer Inquiry III were overturned on appeal (People of the State
of California v FCC, 905 F 2d 1217 (9th Cir 1990)), but CEI requirements and some
ONA obligations were eventually imposed.
27
Inquiry Concerning High-Speed Access to the Internet Over Cable and Other Facilities, Declaratory
Ruling and Notice of Proposed Rulemaking, GN Docket No 00–185, CS Docket No 02–52, FCC No 02–77, 17
FCC Rcd 4798 (2002).
21
The Supreme Court’s decision gave the FCC a legal basis on which to implement
a deregulatory approach to broadband services. Shortly after it was made, the FCC
determined that ‘wireline broadband Internet access services’ were ‘information
services’.28 The reasons given for the FCC’s decision, which was upheld by the US
Court of Appeals for the Third Circuit in Time Warner Telecom v FCC, 507 F 3d
205 (3rd Cir 2007), were similar to those articulated in its cable modem decision.
Significantly, the FCC also decided that common carriers offering these services,
including BOCs, no longer had to comply with its Computer Inquiry rules. In 2006,
the FCC determined that Broadband over Power Line-enabled internet access
services were ‘information services’. In 2007, wireless broadband internet ac-
cess services were classified as ‘information services’.29 In its wireless declaratory
ruling, the FCC found that wireless broadband internet access services were not
‘commercial mobile services’30 on the basis that they do not involve the provision
of an ‘interconnected service’.31 Hence they were not subject to the application of
Title II of the 1934 Act.
Importantly, in each regulatory classification decision, the FCC argued that if
it needed to regulate broadband services it could rely on its ancillary jurisdiction
under Title I of the 1934 Act. However, the April 2010 decision of the US Court of
Appeals in Comcast v FCC, 600 F 3d 642 (DC Cir 2010) called into question the
FCC’s ability to regulate on this basis. The case, discussed further in Section 5.2.5.3,
overturned the FCC’s 2008 decision that Comcast had breached the Commission’s
policy on network neutrality, holding that absent a ‘statutorily mandated respon-
sibility’, such as Title II (common carrier), Title III (spectrum management), and
Title VI (cable), the FCC had no authority to regulate Comcast’s internet man-
agement practices. In its Comcast order, the FCC asserted jurisdiction by relying
28
Appropriate Framework for Broadband Access to the Internet over Wireline Facilities, Report and Order
and Notice of Proposed Rulemaking, CC Docket Nos 02–33, 01–337, 95–20, 98–10, WC Docket Nos 04–2 42, 05–
271, FCC No 05–150, 20 FCC Rcd 14853 (2005). The FCC’s decision was in sharp contrast to its earlier policy.
Previously, the FCC had sought to require ILECs providing wireline broadband internet access services using
xDSL services to provide access to the high frequency portion of the local loop (or line share) in order to in-
crease the roll-out of broadband services on the basis that the conveyance element was a ‘telecommunica-
tions service’. See Deployment of Wireline Services Offering Advanced Telecommunications Capability, Third
Report and Order in CC Docket No 98–147 and Fourth Report and Order in CC Docket No 96–98, FCC No 99–
355, 14 FCC Rcd 20912 (1999).
29
Appropriate Regulatory Treatment for Broadband Access to the Internet Over Wireless Networks,
Declaratory Ruling, WT Docket No 07–53, FCC No 07–30, 32 FCC Rcd 5901 (2007).
30
These are defined in the 1934 Act, §332(d)(1) (as amended) as ‘any mobile service . . . that is provided for
profit and makes interconnected services available (A) to the public or (B) to such classes of eligible users as
to be effectively available to a substantial portion of the public, as specified by regulation by the Commission’.
31
47 USC §332(d)(2) defines the term ‘interconnected service’ as a ‘service that is interconnected with the
public switched network . . . or service for which a request for interconnection is pending pursuant to subsec-
tion (c)(1)(B)’.
231
5 US Telecommunications Law 213
primarily on two policy statements contained in §§1 and 230(b) of the 1934 Act.
Section 1 specifies the purpose for which the FCC was created: ‘regulating inter-
state and foreign commerce in communication by wire and radio so as to make
available, so far as possible, to all of the people of the United States . . . a rapid, ef-
ficient, Nation-w ide, and world-w ide wire and radio communication service . . . at
reasonable charges’. Section 230(b) states it is the policy of the United States to
‘preserve the vibrant and competitive free market that presently exists for the
Internet and other interactive computer services, unfettered by Federal or State
regulation’. In its judgment, the Court of Appeals left open the possibility that the
FCC could regulate internet management practices if it could sustain an argument
that such regulation was necessary in order to regulate matters over which it did
have express statutory authority.
Shortly after the decision of the Court of Appeals, the FCC launched a notice of
inquiry32 on the adequacy of the legal framework for broadband internet access
services soliciting comments on three options:
The 1996 Act gives the FCC the power not to apply Title II in whole or part to pro-
viders of telecommunications services provided specified criteria are met.
Despite quickly launching an inquiry on the matter, the FCC did not reclassify
broadband services until the decision of the US Court of Appeals in Verizon v FCC,
740 F 3d 623 (DC Cir 2014). In that decision, the Court of Appeals overturned por-
tions of the FCC’s 2010 Open Internet Order that sought to codify the Commission’s
network neutrality policy.33 The FCC’s order was grounded primarily in §706 of
the 1996—a provision that directs the FCC to encourage the deployment of ‘ad-
vanced telecommunications capability’. The Court of Appeals held that §706 em-
powered the FCC to adopt rules regulating broadband services, including network
neutrality provisions. However, the proposed network neutrality provisions regu-
lated broadband providers as if they were common carriers subject to Title II of the
1934 Act, and the FCC was prohibited from relying on its §706 power to regulate
broadband providers in this way unless the FCC found that broadband services
32
Framework for Internet Broadband Service, Notice of Inquiry, GN Docket No 10–127, FCC No 10–114, 25
FCC Rcd 7866 (2010).
33
Preserving the Open Internet, Report and Order, GN Docket No 09-191, WC Docket No 07-52, FCC No 10-
201, 25 FCC Rcd 17905 (2010).
241
34
Protecting and Promoting the Open Internet, Report and Order on Remand, Declaratory Ruling, and
Order, GN Docket No 14-2 8, FCC No 15-2 4, 30 FCC Rcd 5601 (2015).
35
USTA v FCC, 825 F 3d 674 (DC Cir 2016). En banc review by the DC Circuit Court of Appeals of its 2016 de-
cision was denied on 1 May 2017. See USTA v FCC and USA, No 15-1063 (DC Cir 1 May 2017).
36
See further Section 5.13.5. 37
See further Section 5.11.3.
38
Restoring Internet Freedom, Notice of Proposed Rulemaking, WC Docket No 17-108, FCC No 17-6 0, 32
FCC Rcd 4434 (2017).
39
Restoring Internet Freedom, Declaratory Ruling, Report and Order, and Order, WC Docket No 17-108,
FCC No 17-166, 33 FCC Rcd 311 (2018).
251
5 US Telecommunications Law 215
a reasonable and timely fashion,40 the FCC observed that Americans are increas-
ingly using mobile broadband services to achieve advanced telecommunications
capability. Notably, it asked whether it should evaluate deployment based on the
presence of both fixed and mobile services. Whether and when consumers or
policy makers might begin viewing fixed and mobile broadband as substitutes for
each other was unclear at the time this chapter was written.
40
Inquiry Concerning Deployment of Advanced Telecommunications Capability to All Americans in a
Reasonable and Timely Fashion, Notice of Inquiry, GN Docket No 17-199, FCC No 17-109, 32 FCC Rcd 7029
(2017).
41
See further Chapter 15, at Section 15.8.
42
Appropriate Framework for Broadband Access to the Internet over Wireline Facilities, Policy Statement,
CC Docket Nos 02–33, 01–337, 95–20, 98–10, GN Docket No 00–185, CS Docket 02–52, FCC No 05–151, 20 FCC
Rcd 14986 (2005).
43
Madison River Communications, Order, File No EB-05–1H-0110, 20 FCC Rcd 4295 (2005).
271
5 US Telecommunications Law 217
customers to use BitTorrent and other peer-to-peer applications which allow the
sharing of video and other large data files.44
As discussed in Section 5.2.5.2, the FCC’s Comcast decision was overturned by
the Court of Appeals. However, prior to the decision of the court, the FCC signalled
its intention to codify the net neutrality policy. In December 2010, the FCC adopted
three rules implementing and expanding upon the principles of the policy. The
‘transparency rule’ applied to all providers of ‘mass market’ broadband internet
access services (other than dial-up), regardless of the type of network technology
used. It required relevant providers to publicly disclose accurate information
about their network management practices and the performance and commercial
terms of their broadband internet access services. The ‘no blocking rule’ stipulated
that fixed and mobile broadband providers could not prevent customers from ac-
cessing, amongst other things, lawful content, applications and services, subject
to ‘reasonable network management’. The ‘no unreasonable discrimination’ rule
prohibited fixed broadband providers from unreasonably discriminating in the
transmission of lawful network traffic over a consumer’s broadband internet ac-
cess service. These provisions were overturned by the Court of Appeal in Verizon v
FCC, 740 F 3d 623, in 2014, although the FCC adopted similar provisions in 2015.45
As discussed in Section 5.2.5.2, the FCC concluded in December 2017 that the
network neutrality framework adopted in 2015 was too onerous because it hin-
dered ISP investment in broadband infrastructure. It therefore decided to abolish
the no blocking rule and the ‘no-u nreasonable interference/d isadvantage rule’
(the Commission’s revised ‘no unreasonable discrimination’ rule adopted in
2015). It decided that a revised transparency rule that stipulates broadband pro-
viders must also disclose information about their blocking, throttling and paid
and affiliated prioritization practices to their customers, in conjunction with anti-
trust and general consumer protection legislation, was better tailored to address
identified regulatory harms.
Determining the appropriate regulatory framework for broadband services has
been and remains one of the most controversial areas of US telecommunications
regulation. The issue has generated a level of public awareness and interest that
is uncharacteristic of most FCC regulatory matters. Moreover, participants in the
debate continue to be split along party political lines. The FCC commissioners
appointed by President Obama who were members of the Democratic Party
supported more robust network neutrality rules. The current chair of the FCC
44
Formal Complaint of Free Press and Public Knowledge Against Comcast Corporation for Secretly
Degrading Peer-to-Peer Applications, Memorandum Opinion and Order, File No EB-0 8–1H-1518, WC Docket
No 07–52, FCC No 08–183, 23 FCC Rcd 13028 (2008).
45
See n 34.
281
appointed by President Trump, Ajit Pai, and Commissioners Michael O’Rielly and
Brendan Carr, also Republicans, staunchly oppose them, preferring less interven-
tion in the market to address regulatory harms.
46
Vonage Holdings Corporation Petition for Declaratory Ruling Concerning an Order of the Minnesota
Public Utilities Commission, Memorandum Opinion and Order, WC Docket No 03–211, FCC No 04–267, 19
FCC Rcd 22404 (2004).
47
These are services that satisfy four criteria: first, they enable ‘real-t ime, two-w ay voice communications’;
second, they require ‘a broadband connection from the user’s location’; third, they require ‘Internet protocol-
compatible customer premises equipment’; fourth, they allow ‘users generally to receive calls that originate
on the public switched telephone network and to terminate calls to the public switched telephone network’.
48
Implementation of the Telecommunications Act of 1996, Report and Order and Further Notice of
Proposed Rulemaking, CC Docket No 96–115, WC Docket No 04–36, FCC No 07–22, 22 FCC Rcd 6927 (2007).
49
Telephone Number Requirements for IP-E nabled Service Providers, Report and Order, Declaratory
Ruling, Order on Remand, and Notice of Proposed Rulemaking, WC Docket No 07–2 43, FCC No 07–188, 22
FCC Rcd 19531 (2007).
50
IP-E nabled Services, First Report and Order and Notice of Proposed Rulemaking, WC Docket Nos 04–36,
05–196, FCC No 05–116, 20 FCC Rcd 10245 (2005).
51
IP-E nabled Services, Report and Order, WC Docket No 04–36, FCC No 07–110, 22 FCC Rcd 11275 (2007).
52
Universal Service Contribution Fund, Report and Order and Notice of Proposed Rulemaking, WC Docket
Nos 06–122, 04–36, CC Docket Nos 96–45, 98–171, 90–571, 92–2 37, 99–200, 95–116, 98–170, NSD File No L–0 0–
72, FCC No 06–94, 21 FCC Rcd 7518 (2006).
53
The Proposed Extension of Part 4 of the Commission’s Rules Regarding Outage Reporting to
Interconnected Voice Over Internet Protocol Service Providers and Broadband Internet Service Providers, PS
Docket No 11-82, FCC No 12-22, 27 FCC Rcd 2650 (2012).
291
5 US Telecommunications Law 219
slow to emerge. In recent years, however, the FCC has adopted a number of im-
portant regulatory measures to support the deployment of and transition to IP-
enabled networks and services. In 2011, the Commission developed transitional
intercarrier compensation arrangements for VoIP-PSTN traffic, which it has since
clarified on several occasions (see Section 5.10.4.1). In 2015, it allowed intercon-
nected VoIP providers to directly obtain telephone numbers from American
numbering administrators (see Section 5.10.3). In addition, there is an ongoing in-
quiry into whether telephone numbers should no longer be associated with the
geographic location of a user.54
54
Numbering Policies for Modern Communications, Notice of Proposed Rulemaking, Order and Notice of
Inquiry, WC Docket Nos 13-97, 04-36, 07-2 43, 10-9 0; CC Docket Nos 95-116, 01-92, 99-200, FCC No 13-51, 28
FCC Rcd 5842 (2013).
20
federal pre-emption doctrine (see Section 5.7). The work of the state regulators is
also coordinated by the National Association of Regulatory Utility Commissioners
(commonly known by its acronym, NARUC). The FCC is significant in that it wields
a significant amount of policy-making authority. It has expansive jurisdiction over
telecommunications issues, despite certain statutory limitations contained in the
Communications Act of 1934. It exercises its authority via its rule-making and
order functions, but like all federal government agencies, remains subject to cer-
tain restrictions and key procedural principles contained in the Administrative
Procedure Act.55
5.4.1.2 Commissioners
The FCC currently consists of five Commissioners, each of whom is appointed by the
President and confirmed by the US Senate. Commissioners serve five-year terms,
and Commissioners may be reappointed. Commissioners who are appointed to fill
vacant positions must serve the remaining term (as opposed to starting a five-year
term), and Commissioners who are not reappointed are limited in how long they
may remain in office even if a replacement has yet to be confirmed. Accordingly,
the Commission can and often will operate with fewer than five Commissioners if
55
Administrative Procedure Act, ch 324, 60 Stat 237 (1946) (codified as amended at 5 USC).
21
5 US Telecommunications Law 221
there are delays in the nomination and confirmation process. All Commissioners
must be US citizens, and a maximum of three Commissioners may have the same
political party affiliation. On a practical level, the Commissioners are responsible
for formulating key policy initiatives, implementing new legislation, and adopting
agency rules and regulations. However, the Commissioners delegate the day-to-
day running of the FCC to its bureaux and offices.
One of the five Commissioners is designated by the President to serve as its
chair, whose general duty is to coordinate the ‘prompt and efficient disposition of
all matters within the jurisdiction of the Commission’. In practice, the chair wields
considerable power by setting the agency’s agenda and directing the work of the
Commission’s bureaux. While the chair serves as the public face of the agency,
all Commissioners are entitled to present their own non-binding views on any
particular issue.
5.4.1.4 Enforcement powers
Under the Communications Act of 1934, the FCC enjoys broad and powerful en-
forcement mechanisms. The FCC may enforce the provisions of the Act directly,
or request the US federal district courts to initiate enforcement proceedings.
Breach of the 1934 Act’s provisions may result in monetary fines, revocation of the
Establishment of the Office of Economics and Analytics, Order, MD Docket No 18-3, FCC No 18-7 (2018).
56
23
5 US Telecommunications Law 223
57
Federal Trade Commission Act, ch 311, 38 Stat 711 (26 September 1914) (codified as amended at 15 USC
§§41–58).
24
Act, 58 and the Sherman Anti-t rust Act. All of the legislation is enforced by the
Federal Trade Commission’s Bureau of Competition and the Antitrust Division
of the Department of Justice (DoJ). Technically, the jurisdiction of these two
bodies overlaps, but the agencies have agreed that the DoJ has primary respon-
sibility for the enforcement of US anti-trust law in the telecommunications
sector. Both the DoJ and the Federal Trade Commission (FTC) have also been
heavily involved in assessing the competitive effects of key industry mergers.
The FTC has also traditionally worked to promote consumer rights. For example,
it manages the Do-Not-Call Registry, which places limitations on telemarketing
phone calls, and investigates complaints against businesses that violate the rules,
even though such jurisdiction is shared with the FCC and state and local author-
ities. It has also used its authority under §5 of the Federal Trade Commission Act
to become the most active US agency in matters of consumer and online privacy.
It is worth noting that the FTC lacks the jurisdiction to take action against unfair
or deceptive acts or practices and unfair methods of competition by common car-
riers. This particular power, which is reserved to the FCC, took on added relevance
within the context of net neutrality when the FCC’s reclassification of fixed and
mobile internet providers as common carriers had the side effect of reducing the
scope of the FTC’s jurisdiction for the period of time when the 2015 network neu-
trality decision was in effect. Moreover, the line between the FTC’s and the FCC’s
authority is not always clear. Recent litigation challenged the assumption that the
FTC could regulate non-communication services that are provided by telecom-
munication companies. The US Court of Appeals for the Ninth Circuit initially
found otherwise in 2016, but the full court subsequently overturned that decision
after rehearing the case.59
The Department of Homeland Security (DHS) was created in March 2003 in the
wake of the September 2001 terrorist attacks against the US, and integrated all or
part of twenty-t wo different federal departments and agencies into a single cab-
inet-level agency. DHS works with the FCC on matters relating to public safety and
security and, in conjunction with the DoJ and Federal Bureau of Investigation, has
taken an active role in matters relating to cybersecurity.
Other federal entities can become involved in telecommunications matters.
The US Department of State coordinates treaty negotiations and preparation for
international radiocommunication conferences, in coordination with the FCC’s
International Bureau. Both the US Patent and Trademark Office and the Copyright
58
Clayton Act, ch 323, 38 Stat 730 (15 October 1914) (codified as amended at 15 USC §§12–278 and 29 USC
§§52, 53).
59
FTC v AT&T Mobility, 835 F 3d 993 (9th Cir 2016); 883 F 3d 848 (9th Cir 2018) (en banc).
25
5 US Telecommunications Law 225
Office, part of the Library of Congress, are involved in issues relating to intellec-
tual property and content rights. New technologies increasingly require the FCC
to work with different specialized federal agencies. For example, the Department
of Transportation has been involved in the development of policies related to un-
manned aeronautical vehicles (‘UAVs’ or ‘drones’) and autonomous automobiles,
while the Food and Drug Administration has an important role in the regulation
of medical devices that incorporate radio transmitters.
5.4.4 Courts
The passage and implementation of the Telecommunications Act of 1996 trig-
gered a flurry of legal challenges to the federal courts (and the DC Circuit Court in
particular), resulting in several key decisions regarding federal pre-emption and
unbundled network elements. Judicial intervention is not new in the telecommu-
nications area, however. Where judicial authority has been exercised in the past,
the courts have tended to adopt a more pro-competitive approach than the FCC.
5 US Telecommunications Law 227
New York, and California, embraced competition within the local exchange mar-
kets; others thwarted competitive efforts by the FCC.
Although the Telecommunications Act of 1996 extends the FCC’s authority to
cover local competition, state regulators retain some jurisdiction over telephony
issues. Their jurisdiction is limited, however, in many cases to ensuring compli-
ance with federal regulations rather than developing policy. State regulators have
the right to prohibit market entry of service providers if necessary to advance
or preserve universal service, public safety, and telecommunications services.
However, any regulation imposed by the states must be done so on a ‘competitively
neutral basis’ and be consistent with the universal service obligations set forth in
§254 of the 1934 Act.
Because the FCC is neither a judicial nor a legislative body, it operates under the
general principles of administrative law. Administrative agencies such as the FCC
are considered to have regulatory expertise in discrete subject areas. Under a
theory of delegation, agencies exercise broad discretion to interpret and apply the
laws passed by Congress and use their authority to enact their own specific rules
and regulations that are legally enforceable. For example, §303(a) of the 1934 Act
gives the FCC authority to classify radio stations, prescribe the nature of service to
be provided in each class, and to determine the location and frequency bands of
such stations. The FCC has used this broad authority to establish different types
of radio services, such as the Microwave Radio Service and Television Broadcast
Stations, and to establish rules and regulations regarding their operation. Each
agency’s rules are compiled in the Code of Federal Regulations (CFR). The FCC’s
28
5 US Telecommunications Law 229
the record’ of the proceeding. In addition, while the agency does not have to adopt
proposals submitted by commenting parties, it cannot ignore them altogether. It
must acknowledge those comments and explain why it has not adopted the par-
ties’ proposals.
Rules adopted in a Report and Order typically do not take effect immediately upon
release of the Commission’s decision. After a Report and Order is adopted, a sum-
mary of the rules it contains and the date, designated by Commission, on which the
rules will become effective have to be published in the Federal Register. Because it
can take weeks (and sometimes even months) for this information to be published,
a significant amount of time can elapse between the date the Commission makes its
decision and when the rules it adopts come into force. In especially complex pro-
ceedings, the FCC may issue a document adopting rules while simultaneously pro-
posing additional rules. Thus, a docket in a proceeding may remain ‘open’ for years
and the FCC may issue documents with complex titles such as ‘Second Report and
Order and Third Notice of Proposed Rulemaking’. The majority of Commissioners’
votes is needed to adopt an item, and it is typical for individual Commissioners to
attach statements explaining their decisions. Items require an affirmative vote of the
majority of the sitting Commissioners to be adopted. Such decisions are made at the
regular monthly meeting of the Commission that is open to the public or through an
internal circulation process. The FCC publishes announcements of upcoming meet-
ings and associated agenda items, as well as a list of draft rulemaking documents
that are ‘on circulation’ awaiting votes.
a chief of a bureau or office), by an ‘Application for Review’. A party must file its
application or petition within a set time period (generally thirty days) to preserve
its right of review. In addition, the FCC may set aside an action on its own motion
within thirty days. The Commission will respond to an application or petition by
issuing a ‘Memorandum Opinion and Order’. If the Commission makes new deter-
minations and also modifies, clarifies, or upholds prior decisions in a particular
proceeding, it may combine a further ‘Report and Order’ and a ‘Memorandum
Opinion and Order’ into one document. The document will, however, always list
the relevant docket and include both titles.
Both the Administrative Procedure Act and the FCC’s rules contain provisions
for formal hearings. However, the use of these procedures has become uncommon,
most likely because the decision of an administrative law judge is still subject to
review by the full Commission. Parties typically choose to bring matters to the
Commissioners directly by filing a petition for reconsideration or they will appeal
directly to federal courts, including the DC Circuit Court.
A party may seek court review of final FCC actions. In its review, a court will
consider whether the FCC acted within its powers, both within the broad powers
of the Communications Act of 1934 and under the specific legislation upon which
the FCC based its rule or action. In addition, a court may, under the Administrative
Procedure Act, set aside the FCC’s decision if it is arbitrary, capricious, an abuse
of discretion, or unsupported by evidence in the record. Courts often invoke the
Administrative Procedure Act when the FCC has not explained the basis for its de-
cision in the written order it adopted. In many cases, the court will send the matter
back to the FCC (remand), with instructions to adequately explain all or a portion
of the decision.
The 1934 Act created a two-t iered system of regulators: (i) the FCC, which is respon-
sible for regulating interstate and foreign commerce in wire and radio communi-
cations; and (ii) state PUCs, with implicitly reserved powers to regulate intrastate
communications. Under the Tenth Amendment of the US Constitution, all powers
not expressly given to the federal government are reserved to the states. The cre-
ation of dual regulators reflected the need to balance the interests of state and fed-
eral governments in the US federal system and, in theory, states retain complete
control over common carriers providing telecommunication services within their
borders.
The actual power states have to regulate intrastate commerce, however, has
been reduced as a result of expansive interpretations of the Commerce Clause
213
5 US Telecommunications Law 231
by the Supreme Court and the use of the pre-emption doctrine based on the
Supremacy Clause. The Commerce Clause of the US Constitution gives the federal
government the power to regulate commerce ‘among the several states’ and with
foreign nations. The Supremacy Clause of the US Constitution enables Congress to
pass law overriding state legislation. It also enables federal agencies, acting within
the scope of their statutory authority, to pre-empt state law when, for example,
state law frustrates or is in conflict with the federal purpose of legislation. In the
telecommunications sector, the Supreme Court has found that many seemingly
intrastate activities directly and/or indirectly affect interstate commerce and thus
fall within the ambit of the FCC.
The FCC began to rely on the pre-emption doctrine in the 1960s, when it sought
to stimulate competition in the intrastate telephony market and tension be-
tween state regulators arose over the funding of universal service. North Carolina
Utilities Commission v FCC, 537 F 2d 787 (4th Cir 1976), cert denied, 429 US 1027
(1976) (NCUC I) was the first in a series of cases that enlarged the jurisdiction of
the FCC to include some power over intrastate communications via reliance on
the pre-emption doctrine. NCUC I arose because several state regulators imposed
conditions on the interconnection of non-AT&T telephone hardware to the local
system in an effort to limit the scope of the FCC’s Carterfone decision (The Use of
the Carterfone Device in Message Toll Service v AT&T, 13 FCC 2d 420 (1968)), which
permitted apparatus conforming to AT&T’s system specifications to be connected
to the phone network. The Fourth Circuit reasoned that because the same hand-
sets were used by customers to place interstate and intrastate calls, the state and
federal regulations were incompatible with each other and that state regulation
had to give way to federal law.
The case is significant as it attempted to define the ambiguous terms ‘inter-
state’ and ‘intrastate’ found in the 1934 Act. The court held that §2(b) of the 1934
Act only limits the FCC from regulating matters that ‘in their nature and effect
are separable from and do not substantially affect the conduct or development of
interstate communications’ (NCUC I at 793). Under this two-prong test, state regu-
lators retain jurisdiction over issues that are separable from interstate communi-
cations and that have no impact on interstate telecommunications. If separation
of interstate and intrastate communications is impossible, the FCC has or acquires
jurisdiction.
The Supreme Court modified the NCUC I test in Louisiana Pub Serv Commission
v FCC, 476 US 355, in 1986. In Louisiana, the court held that the FCC has jurisdic-
tion only if the FCC can demonstrate that interstate and intrastate issues are in-
separable and that the exercise of jurisdiction by the state frustrates the statutory
authority of the FCC.
23
The scope of the FCC’s jurisdiction over intrastate telephony matters was
formally augmented in the Telecommunications Act of 1996. The 1996 Act re-
quired the FCC to introduce competition into the local loop. To that end, the
Act expressly enables the FCC to pre-empt any state legislation that contra-
venes the purposes of local competition.60 In addition, the FCC may pre-empt
any state regulation that may ‘prohibit or have the effect of prohibiting the
ability of any entity to provide interstate and intrastate telecommunications
service’.
Attempts by the FCC to implement measures to introduce local competition
were challenged by incumbent local exchange carriers (ILECs) and PUCs on the
grounds that the FCC lacked the requisite authority to promulgate rules on such
issues as pricing of local services and dialling parity. The Supreme Court in AT&T
Corporation v Iowa Utilities Board, 525 US 366 (1999), however, affirmed that the
FCC had general jurisdiction to implement the provisions of the 1996 Act, notwith-
standing the provisions of the 1934 Act which reserve jurisdiction over intrastate
matters to the states.
More recent examples of the FCC’s reliance on the pre-emption doctrine in-
clude its decisions concerning VoIP,61 the deployment of wireless facilities,62 and
municipal broadband providers.63 In the municipal broadband providers deci-
sion, taken in 2015, the FCC attempted to pre-empt certain state law provisions
restricting the ability of municipal providers to offer cable, video, and internet
services outside of their service areas by arguing that the provisions conflicted
with the federal policy, set out in §706 of the 1996 Act, of ensuring ‘reasonable
and timely’ deployment of broadband services. The Court of Appeals for the Sixth
Circuit eventually overturned the FCC’s decision,64 holding that §706 did not con-
tain a sufficiently clear statement of pre-emption. However, the decision (along
with those concerning VoIP and the deployment of wireless facilities) serves to il-
lustrate the continuing importance of the pre-emption doctrine and the role of the
courts in delineating the boundaries of federal and state jurisdiction over commu-
nications matters.
60
47 USC §251(d)(3)(A)–(C).
61
See n 46. The FCC’s decision was upheld in Minnesota Public Utilities Commission v FCC, 483 F 3d 570
(8th Cir 2007).
62
Acceleration of Broadband Deployment by Improving Wireless Facilities Siting Policies, Report and
Order, WT Docket Nos 13-2 38, 13-32, WC Docket No 11-59, FCC No 14-153, 29 FCC Rcd 12865 (upheld in
Montgomery County v FCC, 811 F 3d 121 (4th Cir 2015)). See also Section 5.10.9.
63
City of Wilson, North Carolina Petition for Preemption of North Carolina General Statutes Sections 160A-
340 et seq, Memorandum Opinion and Order, WC Docket Nos 14-115, 14-116, FCC No 15-25, 30 FCC Rcd 2408
(2015).
64
Tennessee v FCC, 832 F 3d 597 (6th Cir 2016).
23
5 US Telecommunications Law 233
5. 8 L IC ENSING 6 5
Subject to exemptions adopted by the FCC and limited statutory exceptions, Title
II of the Communications Act of 1934 (1934 Act) requires operators and providers
of interstate and overseas communications services who meet the definition of
‘common carrier’ to obtain the permission of the FCC before network operation
and service provision. Common carriers must therefore ensure they hold and
comply with the relevant authorization(s). Carriers that wish to use radio broad-
casting, such as microwave links, must also obtain permission from the FCC to
use the radio spectrum (see Section 5.8.4). If carriers wish to provide intrastate
services, they must obtain the requisite authorizations from the PUC in each
relevant state.
5 US Telecommunications Law 235
The FCC’s auction authority dates to 1993, when Congress added §309(j) to the
1934 Act. Spectrum auctions are available for situations where there would be ‘mu-
tually exclusive’ applications for the same licence, such as licensing an exclusive
right to operate in a particular frequency band within a set geographic area. When
multiple applications can be accommodated without conflict, such as narrow
point-to-point microwave links, there is no mutual exclusivity and auctions are
not appropriate. Although the billions of dollars in public revenue raised by spec-
trum auctions have attracted considerable attention, the 1934 Act (as amended)
requires the FCC to consider efficient spectrum use and not the expectation of rev-
enues as the dominant factor in designing and implementing auctions. The FCC is
also mandated to ensure that licences are disseminated among a ‘wide variety of
licensees’, including small businesses, rural telephone companies, and women-
and minority-owned businesses. The FCC has addressed this requirement by
establishing bidding credits for ‘designated entities’. Other methods which have
been employed (and generally without widespread success) include the offering of
FCC-sponsored financing for winning designated entities and the setting aside of
specific spectrum blocks that only designated entities may bid on.
The current auction process promotes both efficiency and participation by ser-
ious applicants. Whereas the very first auctions were chaotic affairs with a live auc-
tioneer conducting proceedings in a large public function space, modern auctions
are more akin to routine business transactions. Simultaneous bidding for multiple
licences is conducted remotely by computer, and bidding rounds can last weeks if
not months. In general, each interested bidder must file a ‘short form’ application
prior to an auction that discloses its qualifications, and must submit an upfront
deposit in relation to the licences it wishes to bid on. In each auction round, the
bids may be increased only by a set increment. Once bidding activity drops below a
set level, the auction closes and the FCC announces tentative winners for licences
that have satisfied the auction’s conditions (such as the ‘reserve’—a minimum bid
amount). Shortly thereafter, winning bidders must file a ‘long form’ application
and submit payments. Bid withdrawal and default penalties are designed to en-
sure that only serious bidders participate. In addition, the FCC has adopted rules
to prevent bidding collusion.
The spectrum auction policy is considered to be a success. Auctions can quickly
allocate licences and promote the rapid deployment of service. In addition, the FCC
has developed considerable expertise in designing and conducting auctions that
have served as models for other countries considering their own spectrum auc-
tions. The auction policy has not been without problems, however. Traditionally,
the FCC faced the greatest difficulties in implementing its designated entity pro-
cedures. Constitutional challenges undermined the FCC’s women and minority
bidding preference programmes, and some designated entities either defaulted on
273
5 US Telecommunications Law 237
made unlicensed operations an attractive proposition for entities that may not
have the resources or business need to invest in exclusive spectrum licences, and
the unlicensed model has fostered innovations in wireless spectrum technologies
and use. Originally characterized by cordless telephones, garage door openers,
and baby monitors, unlicensed devices have now become a vital part of the com-
munications landscape. WiFi and Bluetooth protocols are designed for unlicensed
use, and the future ‘Internet of Things’ is expected to make extensive use of the
unlicensed model. Wireless internet service providers (mostly small and rural in
nature) have long operated on an unlicensed basis. Commercial mobile operators
have increasingly come to rely on WiFi to provide backhaul support for their li-
censed networks and, in 2016, began deploying equipment designed to provide
4G LTE services under the unlicensed rules. This development has been contro-
versial; while the technology promises to enhance mobile carriers’ service and re-
liability by increasing data speeds over short distances, WiFi and cable providers
expressed concerns over its compatibility with existing unlicensed devices and
protocols that already crowd the spectrum available for unlicensed use.
The future of spectrum licensing is likely to be increasingly complex, and will in-
clude combinations of the licensing and authorization models discussed above.
For example, the Commercial Spectrum Enhancement Act of 2004,67 which es-
tablished a mechanism for reimbursing federal agencies out of spectrum auction
proceeds for the cost of relocating existing operations and was used in the auction
of Advanced Wireless Services in 2006, represents an important milestone for the
repurposing of federal spectrum for commercial use. The television incentive auc-
tion in 2016 provided evidence that incumbent licensees will voluntarily agree to
give up some or all spectrum rights in exchange for a portion of the auction pro-
ceeds. The adoption of service rules for a new radio service, the Citizens Broadband
Radio Service, in 2015 is particularly noteworthy because it introduced a mech-
anism by which different users holding different spectrum rights and authorized
under different models will be permitted to operate in the same 3.5 GHz band spec-
trum. Incumbent Access users, consisting of federal government users and grand-
fathered fixed satellite service operations, will receive protection from harmful
interference from all other band users. Priority Access Licenses, awarded by auc-
tion, will be authorized to use their licensed channel(s) in a specific geographic area
for a multiple-year period. General Authorized Access users will be permitted to
use any portion of the band not assigned to Incumbent Access users and Priority
Access users and will be licensed-by-rule. In addition, General Authorized Access
67
Commercial Spectrum Enhancement Act, Pub L No 108–494, 118 Stat 3986, Title II (2004) (codified in
various sections of 47 USC).
293
5 US Telecommunications Law 239
users may operate in unused Priority Access channels by using advanced radios that
consult with online databases to determine where and when such vacant channels
exist. This idea—that users will be authorized to access spectrum opportunistically
on an ad hoc basis as opposed to a pre-arranged assignment—draws on the work of
the ‘TV White Spaces’ proceeding. There, the FCC permitted the operation of con-
sumer devices that make use of location-sensing technologies to avoid interference
with the signals of incumbent broadcasters, on an unlicensed basis in spectrum be-
tween licensed television channels.68
68
ET Docket 04–186. At the time of writing, the Commission was re-evaluating its rules for the 3.5 GHz ser-
vice, but was not expected to alter the fundamental three-t iered licensing approach.
69
See further Chapter 16, at Section 16.4.3.1.
420
70
Review of Foreign Ownership Policies for Common Carrier and Aeronautical Radio Licensees under
Section 310(b)(4) of the Communications Act of 1934, as Amended, Second Report and Order, IB Docket No
11-133, FCC No 13-50, 28 FCC Rcd 5741 (2013).
71
Review of Foreign Ownership Policies for Common Carrier and Aeronautical Radio Licensees under
Section 310(b)(4) of the Communications Act of 1934, as Amended, First Report and Order, IB Docket No 11-
133, FCC No 12-93, 27 FCC Rcd TBA 9832 (2012).
72
See n 70.
214
5 US Telecommunications Law 241
for approval to own 100 per cent of shares in US entities that control broadcast radio
licensees.73
73
Review of Foreign Ownership Policies for Broadcast, Common Carrier and Aeronautical Radio Licensees
under Section 310(b)(4) of the Communications Act of 1934, as Amended, Report and Order, GN Docket No
15-2 36, FCC No 16-128, 31 FCC Rcd 11272 (2016).
74
See further Chapter 7. 75
See further Chapter 16, at Section 16.3.2.
24
systems located near international borders, avoid harmful interference from the
use of incompatible types of services. The US uses the same terms to designate
categories of services and allocations as does the ITU in the international Radio
Regulations.76
The FCC publishes a table of frequency allocations at 47 CFR §2.106 that lists the
international allocation for each region, the US table of frequency allocations for
both federal government and non-federal government use, and a list of the rele-
vant FCC rules for each band. As discussed in Section 5.4.2, a separate body, NTIA,
administers spectrum used by federal government entities. Accordingly, while the
FCC lists federal government allocations in the US table of frequency allocations, it
does not control that use. A spectrum band may have both primary and secondary
allocations. Within a band, secondary services must not cause harmful interfer-
ence to stations of primary services, nor may secondary services claim protection
from stations of a primary service. Generally, however, a station operating on a
secondary basis may claim protection from a secondary station that begins oper-
ation at a later date.
This allocation model serves as the basic framework for organizing the electro-
magnetic spectrum. Once spectrum has been allocated for a particular purpose
(eg fixed or mobile services, broadcasting, earth-to-space satellite operations), the
FCC may then designate a particular type of radio service to use that spectrum
band and set forth appropriate licensing and operational rules for that service.
Spectrum licensing is discussed in greater detail in Section 5.8.4. It is important
to keep in mind that spectrum use under an unlicensed authorization model, as
discussed in that section, falls outside of the allocation framework. There are no
allocations for unlicensed services. Instead, unlicensed devices operate on a suf-
ferance basis and must accept any interference from and not cause interference to
any and all licensed services.
This section reviews some of the important access, interconnection, and related
measures imposed by Congress, the FCC, and other bodies. Where possible, it
highlights the tensions which have arisen between established operators and
their competitors, both of whose revenue streams are significantly affected by the
underlying policy.
76
See further Chapter 16, at Section 16.3.4. See further Chapter 8.
77
243
5 US Telecommunications Law 243
FCC rules mandate that ILECs provide local loops, subloops, network interface de-
vices, dedicated transport, 911 and Enhanced 911 databases which enable calls
to emergency services and operations support systems (OSS), although certain
restrictions apply.85 Access to these network elements is designed to facilitate
competition in narrowband and broadband services. When carriers are unable
to agree the prices of UNEs, PUCs determine them in accordance with a forward
long-r un incremental methodology adopted by the FCC or a series of proxy ceil-
ings and ranges, also set by the FCC.86
It is an understatement to say that, for the FCC, formulating the necessary and
impair tests and determining the specific network elements to be unbundled was
a fraught process. Implementation of the unbundling provisions finally ended in
2006, a decade after the 1996 Act was adopted. The FCC’s difficulty was in part due
to the lack of guidance given in the Act about the factors it should take into account
when determining which network elements should be unbundled. The Act stated
only that the FCC had to consider ‘at a minimum’ if access to proprietary network
elements was ‘necessary’ and if denial of access to non-proprietary network elem-
ents would ‘impair’ the ability of a carrier seeking UNEs to provide telecommuni-
cations services.87 In addition, the FCC had significant difficulty developing rules
that passed judicial scrutiny. The courts overturned the FCC’s unbundling rules
in whole or in part on three occasions. In 1999, the Supreme Court rejected the
FCC’s first formulations88 of the necessary and impair standards.89 In 2002, the
US Court of Appeals’ DC Circuit 90 upheld the FCC’s revised necessary standard91
but overturned its new impairment test. In 2004, the court92 was again critical of
the FCC’s third formulation93 of the impairment standard. However, in 2006, the
court 94 upheld the FCC’s new definition of impair developed in light of the 2004
85
47 CFR §51.319.
86
47 USC §252(d)(1); 47 CFR §§51.501–51.515. See also Review of the Commission’s Rules Regarding the
Pricing of Unbundled Network Elements, Notice of Proposed Rulemaking, WC Docket No 03–173, FCC No
03–224, 18 FCC Rcd 18945 (2003).
87
See 47 USC 251(d)(2).
88
Implementation of the Local Competition Provisions in the Telecommunications Act 1996, First Report
and Order, CC Docket Nos 96–98, 95–185, FCC No 9–325, 11 FCC Rcd 15499 (1996). This report and order is also
known as the Local Competition Order.
89
FCC v Iowa Utilities Bd, 525 US 1133 (1999).
90
United States Telecom Association v FCC, 290 F 3d 415 (DC Cir 2001) (USTA 1).
91
Implementation of the Local Competition Provisions of the Telecommunications Act of 1996, Third
Report and Order and Fourth Further Notice of Proposed Rulemaking, CC Docket No 96–98, FCC No 99–2 38,
15 FCC Rcd 3696 (1999). This report and order is also known as the UNE Remand Order.
92
United States Telecom Association v FCC, 359 F 3d 554 (DC Cir 2004) (USTA II).
93
Review of the Section 251 Unbundling Obligations of Incumbent Local Exchange Carriers, Report and
Order and Order on Remand and Further Notice of Proposed Rulemaking, CC Docket Nos 01–338, 96–98, 98–
147, FCC No 03–0 6, 18 FCC Rcd 16978 (2003). This report and order is also known as the Triennial Review Order.
94
Covad Communications Co and DIECA Communications, Inc v FCC, 450 F 3d 528 (DC Cir 2006).
245
5 US Telecommunications Law 245
ruling.95 As the court stated, ‘the Commission’s fourth try [was] a charm’.96 One of
the consequences of the extensive litigation has been a shift to a higher threshold
for unbundling than the FCC had originally envisaged and competitive local ex-
change carriers (CLECs) had wished. The revised standards implicitly adopt an
‘essential-facilities’ type focus which favours ILECs.97
95
Unbundled Access to Network Elements, Order on Remand, WC Docket No 04–313, CC Docket No 01–338,
FCC No 04–290, 20 FCC Rcd 2533 (2005).
96
Covad, 450 F 3d at 531.
97
For further information on the saga of the FCC’s unbundling rules, see Lee, K and Prime, J, ‘Overview
of US Telecommunications Law’, in Walden, I (ed), Telecommunications Law and Regulation (2nd edn,
Oxford: OUP, 2005), at 531–538.
98
See 47 USC §251(c)(6).
99
Deployment of Wireline Services Offering Advanced Telecommunications Capability, Third Report and
Order in CC Docket No 98–147 and Fourth Report and Order in CC Docket No 96–98, FCC No 99–355, 14 FCC
Rcd 20912 (1999).
426
equipment co-located at different ILEC premises. The right of CLECs to select the
physical co-location space at an ILEC’s premises and the prohibitions on ILECs re-
quiring CLECs to use separate rooms for co-location were also overturned.
The FCC published its revised rules in 2001.100 It determined that ILECs had to
permit CLECs to co-locate switching and routing equipment. Other multifunction
equipment could be co-located only if its primary purpose was for interconnection
and/or access. ILECs also had to provide cross-connection to CLECs co-located
within the same premises as ILECs.
The FCC’s current co-location rules are set out in 47 CFR §51.323.
5.10.3 Interconnection
The 1996 Act, at 47 USC §251(a), imposes a duty on all telecommunications carriers
to interconnect ‘directly or indirectly’ with the facilities and equipment of other
carriers. ILECs must permit other carriers to interconnect to the PSTN at any tech-
nically feasible point in their networks to enable the transmission and routing of
telephone exchange and exchange access services.101 The rates ILECs charge must
be ‘just, reasonable and non-d iscriminatory’,102 and any service provided by an
ILEC must be equal in quality to the service it supplies to itself or any affiliate.
The responsibility for day-to-day interconnection issues, including determin-
ations of whether or not rates are ‘just and reasonable’, falls to PUCs. However, in
the Local Competition First Report and Order, the FCC mandated that the states
had to apply the same long-run incremental cost methodology (LRIC) pricing
standard it adopted for UNEs. PUCs challenged the FCC’s decision but in FCC v
Iowa Utilities Bd, 525 US 1133 (1999), the Supreme Court held that the FCC had
the authority to direct states to adopt a uniform pricing methodology. The adop-
tion of a LRIC standard was also attacked by ILECs but was ultimately upheld by
the Supreme Court in Verizon v FCC, 535 US 467 (2002). The FCC may also directly
intervene in interconnection disputes where a PUC fails to carry out its responsi-
bilities under the Telecommunications Act of 1996 codified at 47 USC §252.
In an important ruling for VoIP providers in March 2007,103 the FCC affirmed
that CLECs providing wholesale telecommunications services to VoIP service pro-
viders are entitled to interconnect with ILECs under §251(a) and (b) of the Act.
100
Deployment of Wireline Services Offering Advanced Telecommunications Capacity, Fourth Report and
Order, CC Docket No 98–147, FCC No 01–204, 16 FCC Rcd 15435 (2001).
101
47 USC §252(c)(2)(A)–(B). 102
47 USC §252(c)(2)(D).
103
Time Warner Cable Request for Declaratory Ruling that Competitive Local Exchange Carriers May
Obtain Interconnection Under Section 251 of the Communications Act of 1934, as Amended, to Provide
Wholesale Telecommunications Services to VoIP Providers, Memorandum Opinion and Order, WC Docket
No 06–55, DA 07–709, 22 FCC Rcd 3513 (2007).
247
5 US Telecommunications Law 247
Certain PUCs had denied CLECs rights of access to ILEC interconnection services
in 2005 on the ground that CLECs were not telecommunications carriers when
providing wholesale services to VoIP service providers—they were not offering
‘telecommunications for a fee directly to the public’. The FCC’s ruling enables the
exchange of voice calls between broadband networks and the PSTN when VoIP
service providers procure interconnection services through CLECs. The FCC has
yet to determine if VoIP providers are ‘telecommunications carriers’ and conse-
quently permitted in their own right to rely on §251 of the Act to interconnect dir-
ectly with ILECs.
There remain no FCC rules specific to the interconnection of IP networks,
even though the Commission has recognized for many years that the rights and
obligations of providers need to be clarified. However, it has adopted a number
of measures designed to facilitate IP-to-IP interconnection. One example is its
2015 decision to permit interconnected VoIP providers to directly access tele-
phone numbers from American numbering administrators. Previously, they had
to obtain them from a telecommunications carrier.104 The FCC also continues to
reiterate its expectation that carriers will negotiate IP-to-IP interconnection re-
quests in good faith.
104
Numbering Policies for Modern Communications, Report and Order, WC Docket Nos 13-97, 04-36, 07-
243, 10-9 0; CC Docket Nos 95-116, 01-92, 99-200, FCC No 15-70, 30 FCC Rcd 6839 (2015).
105
47 CFR §§51.701–51.717.
428
106
Implementation of the Local Competition Provisions in the Telecommunications Act of 1996,
Declaratory Ruling and Notice of Proposed Rulemaking, CC Docket Nos 96–98, 99–6 8, FCC No 99–38, 14 FCC
Rcd 3689 (1999). The jurisdiction of the FCC to regulate ISP-bound traffic was upheld in Bell Atlantic Telephone
Companies v FCC, 206 F 3d 1 (2000), although other aspects of the Declaratory Ruling were overturned.
107
Developing a Unified Intercarrier Compensation Regime, Notice of Proposed Rulemaking, CC Docket
No 01–92, FCC No 01–132, 16 FCC Rcd 9610 (2001).
108
See ibid paras 77–8 8 for a description of the interim regime. The FCC argued that ISP-bound traffic was
not subject to the reciprocal compensation requirements of §251(b)(5). Nevertheless, it had authority to regu-
late ISP-bound traffic pursuant to §251(g) of the 1934 Act.
109
Developing a Unified Intercarrier Compensation Regime, Notice of Proposed Rulemaking, CC Docket
No 01–92, FCC No 05–33, 20 FCC Rcd 4685 (2005).
110
Federal Communications Commission, Public Notice, Comment Sought on Missoula Intercarrier
Compensation Reform Plan, DA 06–1510 (25 July 2006).
111
A copy of the plan is available at <http://w ww.fcc.gov/wcb/ppd/I ntercarrierCompensation/h istory.
html>.
112
In re Core Communications, Inc, 531 F 3d 849 (2008).
249
5 US Telecommunications Law 249
identify a legal basis on which the FCC could exclude ISP-bound traffic from its
reciprocal compensation rules. The decision was prompted by the FCC’s failure to
articulate a legal basis for the interim arrangement following the decision of the
Court of Appeals in 2002113 not to vacate the FCC’s decision adopting the interim
rules even though it held the FCC could not base the interim regime on 47 USC
§251(g) added by the 1996 Act. The unusual action of the Court of Appeals was
a result of the court’s belief that there was a strong likelihood that the FCC had
statutory power to regulate ISP-bound traffic. The FCC’s response to the 2008 deci-
sion of the Court of Appeals disappointed many, however. Rather than announce
sweeping reforms, the FCC adopted a narrow order114 to address the specific con-
cerns of the court,115 enabling the FCC to keep the interim arrangements in place,
and solicited comment on yet another three proposals for reform. One proposed
option involved doing nothing.
Unsurprisingly, the FCC task force responsible for development of the National
Broadband Plan identified reform of the intercarrier compensation regime as a pri-
ority. It called for a three-stage reform process which included the rebalancing of
intrastate and interstate termination rates and the abolition of per-m inute charges
phased in over time with corresponding incremental increases in subscriber
line charges to offset revenue losses. Implementation of these broad objectives
began in 2011 with the publication of a Report and Order and Further Notice of
Proposed Rulemaking.116 In the Report and Order, the FCC adopted amendments
to its interstate access rules to address the problems of ‘access stimulation’117 and
made changes to its call-signalling rules to stop the problem of ‘phantom traffic’.118
Further, it stated the current per-m inute system would eventually be replaced
with a bill and keep framework as the default methodology for all intercarrier
compensation traffic. Carriers remain free to negotiate alternative arrangements.
In order to reach the goal of bill and keep, the FCC adopted a series of transitional
High-Cost Universal Service Support, Order on Remand and Report and Order and Further Notice of
114
Proposed Rulemaking, WC Docket Nos 05–337, 03–109, 06–122, 04–36, CC Docket Nos 96–45, 06–122, 99–200,
96–98, 01–92, 99–6 8, FCC No 08–262, 24 FCC Rcd 6475 (2008).
115
The FCC determined that ISP-bound traffic was subject to the statutory requirements of reciprocal com-
pensation. Nevertheless, it asserted it could regulate ISP-bound traffic in accordance with the interim ar-
rangements as a result of §201 of the 1934 Act.
116
Connect America Fund, Report and Order and Further Notice of Proposed Rulemaking, WC Docket Nos
10–9 0, 07–135, 05–337, 03–109, GN Docket No 09–51, CC Docket Nos 01–92, 03–109, WT Docket No 10–208, FCC
No 11–161, 26 FCC Rcd 17663 (2011).
117
Access stimulation occurs when LECs seek to generate revenue by increasing the volume of traffic ter-
minated to their networks by entering into access revenue sharing agreements with high volume customers
such as chat line and adult entertainment providers.
118
Phantom traffic is the practice of carriers and providers disguising the origin of calls in an effort to avoid
or minimize termination payments.
520
arrangements which included capping rates for all forms of interstate access and
intrastate termination. Carriers had to bring interstate and intrastate termin-
ation rates to parity by July 2013 and to gradually reduce their rates to bill and
keep over a period of six to nine years, depending on the type of carrier involved.
To help offset termination revenue lost during the period of transition, ILECs and
other carriers may recoup some costs from their customers in accordance with
FCC rules developed for this purpose and, in some instances, from the Connect
America Fund (see Section 5.11). The FCC also adopted interim arrangements for
VoIP–PSTN traffic but has yet to determine the appropriate transitional measures
for origination and transportation rates. The FCC has had to clarify the complex
rules implementing its 2011 Report and Order on numerous occasions,119 but it re-
mains committed to the adoption of a bill and keep framework.
5.10.5 Re-sale
47 USC §251(b)(1), added by the 1996 Act, requires all LECs to make their tele-
communications services available for resale on reasonable and non- d is-
criminatory terms. All resale services must be of equal quality and provided
119
See, eg, Connect America Fund, WC Docket No 10-9 0, CC Docket No 01-92, FCC No 15-14, 30 FCC Rcd
1587 (2015).
120
Calling Party Pays Service Offering in the Commercial Mobile Radio Services, Memorandum Opinion
and Order on Reconsideration and Order Terminating Proceeding, WT Docket No 97–207, FCC No 01–125, 16
FCC Rcd 8297 (2001).
121
The Effect of Foreign Mobile Termination Rates on US Customers, Notice of Inquiry, IB Docket No 04–
398, FCC No 04–2 47, 19 FCC Rcd 21395 (2004).
215
5 US Telecommunications Law 251
within the same period of time as LECs provide to others.122 The wholesale rates
that incumbent LECs may charge are determined by PUCs, subject to direc-
tions from the FCC, and are calculated by reference to the retail rate for the
relevant telecommunications service less ‘the portion thereof attributable to
any marketing, billing, collection and other costs’123 avoided by the ILEC by not
providing its services to retail customers. In its Local Competition First Report
and Order, the FCC adopted a ‘reasonably avoidable standard’ to determine
avoidable costs. However, the Commission’s ILEC re-sale pricing rules were
overturned by the Eighth Circuit Court of Appeals in Iowa Utilities Bd v FCC, 219
F 3d 744 (2000). The court held that the FCC’s interpretation was inconsistent
with the plain meaning of the statute. The appropriate standard for determining
avoided costs was those costs that the ILEC actually avoid incurring in the fu-
ture, because of its wholesale efforts, not costs that ‘can be avoided’. The court
also stated that PUCs must assume that ILECs are acting as both a wholesale
and retail provider when they determine costs whereas previously the FCC’s
rules had treated ILECs as only wholesalers. No modifications to the resale rules
were made following the court’s decision, although the FCC sought comment
on the need to adopt new rules implementing the relevant statutory provision
in 2003.124
47 CFR §51.603.
122 123
47 USC §252(d)(3).
Review of the Commission’s Rules Regarding the Pricing of Unbundled Network Elements and the
124
Resale of Service by Incumbent Local Exchange Carriers, Notice of Proposed Rulemaking, WC Docket No
03–173, FCC No 03–224, 18 FCC Rcd 20265 (2003).
125
The FCC’s interpretation received the backing of the Supreme Court in National Cable &
Telecommunications, Inc v Gulf Power Co, 534 US 327 (2002).
25
The National Broadband Plan, which the FCC was required to submit to
Congress in 2010,126 highlighted a number of weaknesses in the FCC’s rules dealing
with pole attachments, including significant disparities between the rates cable
and telecommunications carriers paid for access to poles (cable rates were lower)
and inefficiencies in access procedures, all of which it was argued increased the
cost of broadband deployment for both wireless and wired carriers. The FCC sub-
sequently amended its rules in April 2011127 to establish a four-stage process for
pole attachments, each with prescribed deadlines. Among other measures, it
modified its cost formulae to ensure that the rates paid by telecommunications
carriers were on par with those of cable operators and imposed obligations on
parties to escalate disputes to their respective corporate executives prior to filing
complaints with the FCC. In an Order on Reconsideration issued more than four
years later,128 the FCC addressed concerns that its 2011 Order, as written, allowed
pole owners to apply the cost formulae in a way that still allowed for disparities
between the rates charged to cable and telecommunications carriers. The deci-
sion was intended to foster a more harmonized cost model to bring the cost of
pole attachments for telecommunications carriers in line with the traditionally
lower cable rate. This FCC action was especially important in light of its decision
to reclassify broadband services as telecommunications services.129 That decision
led to the worry that pole owners would begin imposing higher telecommunica-
tions carrier rates on cable companies. Although a consortium of electric utilities
challenged the FCC’s decision, it was upheld in 2017 by the US Court of Appeals
for the Eighth Circuit.130
The FCC continues to recognize that access to poles, conduits, ducts, and rights
of way are keys to improved infrastructure use and broadband deployment, es-
pecially with the advent of 5G and the consequential need for additional mobile
network infrastructure. Popular ideas being discussed by policy makers include
mandating ‘one touch’ access in which a single construction crew would be
permitted to complete all of the work necessary to make a pole ready for a new
attachment, including making necessary modifications to existing equipment in-
stalled by attachers, and ‘dig once’ requirements in which in-street conduits must
be built to accommodate future users so the streetscape is not re-d isturbed. In
2017, the Commission formed a new federal advisory committee, the Broadband
126
See further Section 5.11.1.
127
Implementation of Section 224 of the Act and a National Broadband Plan for Our Future, Report and
Order and Order on Reconsideration, WC Docket No 07-2 45, GN Docket No 09-51, FCC No 11–50, 26 FCC Rcd
5240 (2011).
128
Implementation of Section 224 of the Act and a National Broadband Plan for Our Future, Order on
Reconsideration, WC Docket No 07-2 45, GN Docket No 09-51, FCC No 15–151, 30 FCC Rcd 13731 (2015).
129
See further Section 5.2.5.2. 130
Ameren Corporation v FCC, 865 F 3d 1009 (8th Cir 2017).
253
5 US Telecommunications Law 253
131
Rules and Regulations Implementing Minimum Customer Account Record Exchange Obligations on All
Local and Interexchange Carriers, Report and Order and Further Notice of Proposed Rulemaking, CG Docket
No 02–386, FCC No 05–29, 20 FCC Rcd 4560 (2005).
132
Telephone Number Portability, Memorandum Opinion and Order and Further Notice of Proposed
Rulemaking, CC Docket No 95–116, FCC No 03–2 84, 18 FCC Rcd 23697 (2003).
133
Telephone Number Requirements for IP-E nabled Service Providers, Report and Order, Declaratory
Ruling, Order on Remand and Notice of Proposed Rulemaking, WC Docket Nos 07–2 43, 07–2 44, 04–36, CC
Docket Nos 95–116, 99–200, FCC No 07–188, 22 FCC Rcd 19531 (2007).
524
134
See n 104.
25
5 US Telecommunications Law 255
on a church’s roof despite the fact that the denial effectively precluded the provi-
sion of service to a part of the community. Subsequently, courts distinguished or
explicitly rejected the Virginia Beach decision as not adequately recognizing the
extent of pre-emption provided in the Telecommunications Act of 1996.
The FCC has frequently taken actions to try to reduce the regulatory barriers
faced by providers who need to deploy additional towers and antenna to fulfil con-
sumer demand. For example, in 2009 it defined the meaning of ‘reasonable period
of time’ for the purposes of §332(c)(7) as ninety days for co-location applications
and 150 days for all other siting applications.135 Prior to its decision, there were re-
ports of applications pending for up to three years. The National Broadband Plan
the FCC submitted to Congress in 2010 emphasized the need for federal, state,
and local governments to further minimize barriers to infrastructure roll-out,
calling for, among other things, the development of a standard form master con-
tract for the placement of wireless towers on all federal land. The agency has also
regularly turned to advisory committees to investigate this issue. The Technology
Advisory Council has periodically examined wireless antenna requirements, re-
commending such measures as the adoption of an Executive Order to expedite
the approval process for antenna sitings on federal land and the elimination of
redundant requirements in state and local planning practices. The Broadband
Deployment Advisory Committee, discussed in Section 5.10.6, began identifying
barriers to wireless siting in late 2017, information which the FCC has incorporated
into its 5G-related rulemaking proceedings.
When making decisions in this area, the FCC has to consider the requirements
imposed on it by the National Historic Preservation Act (NHPA),136 the Endangered
Species Act (ESA),137 and the National Environmental Policy Act (NEPA).138 NEPA
is a ‘cross-cutting law’ in that it applies broadly to federal undertakings and re-
quires agencies to implement procedures for considering potential environ-
mental effects during the agency’s decision-making process. FCC licensees and
applicants are required to review whether or not their proposed actions will have
environmental consequences and, if applicable, prepare an environmental as-
sessment that leads to a series of steps designed to evaluate the environmental
135
Petition for Declaratory Ruling to Clarify Provisions of Section 332(c)(7)(B) to Ensure Timely Siting
Review and to Preempt Under Section 253 State and Local Ordinances that Classify All Wireless Siting
Proposals as Requiring a Variance, Declaratory Ruling, WT Docket No 08–165, FCC 09–99, 24 FCC Rcd 13994
(2009).
136
National Historic Preservation Act of 1966, Pub L No 89–6 65, 80 Stat 915 (1966) (codified as amended at
16 USC §470 et seq).
137
Endangered Species Act of 1973, Pub L No 93–205, 87 Stat 884 (1973) (codified as amended at 16 USC
§1531 et seq).
138
National Environmental Policy Act of 1969, Pub L No 91–190, 83 Stat 852 (1970) (codified as amended at
42 USC §§4321–4347).
526
effect of the proposed action and, if the impact is significant, to seek alternatives or
mitigations. The filing of an environmental assessment is required, among other
times, when a proposed facility may have a significant effect on historic proper-
ties, could threaten endangered species or critical habitats, or may affect Native
American religious sites.
One particular challenge has been the issue of migratory birds. Many parties,
including the US Fish and Wildlife Service (FWS), have estimated that commu-
nications towers kill between four and five million birds per year, and have a
potentially significant impact on migratory birds—including some 350 species
of night-m igrating birds that may be affected by tower lights. In February 2008,
the US Court of Appeals for the DC Circuit139 struck down a 2005 FCC decision
denying a petition of the American Bird Conservancy and the Forest Conservation
Council which requested, among other things, that the FCC prepare an environ-
mental impact assessment of communications towers on migratory birds in the
Gulf Coast region of the US. The 2008 decision of the Court of Appeals also trig-
gered an ongoing nationwide environmental assessment of the FCC’s antenna
structure registration procedures. Since that time, federal agencies have worked
to better harmonize their interests in this area, and to unify the guidance they give
regarding tower siting issues. For example, the Federal Aviation Administration
(FAA) has revised its procedures to make it easier to discontinue the use of steady
burning lights on towers (which are more likely to attract birds); the FCC provides
more detailed guidance for when tower construction might trigger the need to
conduct an environmental assessment; and FWS has issued a set of recommended
best practices for the design, siting, construction, operation, maintenance, and
decommissioning of communications towers.
Wireless siting issues have taken on a newfound importance with the advent of
5G networks. The deployment of dense networks of small cells threaten to over-
whelm existing site approval processes, and the Commission has indicated that it
intends to play an active role in this area. A Notice of Proposed Rulemaking and
Notice of Inquiry issued in April 2017140 critically reviewed, among other things,
the FCC’s existing rules and procedures for site evaluations under NHPA and
NEPA, as well as the effect of state, local, and tribal review. In announcing the
proceeding, the Commission cited ‘evidence that despite Commission action to
reduce delays and costs of infrastructure review, providers continue to face sig-
nificant costs and delays and reform may be needed’. The FCC adopted an initial
American Bird Conservancy, Inc and Forest Conservation Council v FCC, 516 F 3d 1027 (DC Cir 2008).
139
Notice of Proposed Rulemaking and Notice of Inquiry, WT Docket No 17-79, FCC No 17-38, 32 FCC Rcd 3330
(2017).
257
5 US Telecommunications Law 257
set of rules intended to remove barriers to the siting process in March 2018, while
continuing to examine other issues raised in the proceeding. The Commission will
still have to find ways to balance its obligation to promote the widespread avail-
ability of telecommunications with the interests of state, local, and tribal commu-
nities, historical preservationists and environmental groups, as well as the limits
imposed by NEPA and other laws. However, the extensive deregulatory activities
of the Trump administration in conjunction with pressure to ensure that the US
does not fall behind in 5G development may alter the balance in favour of telecom-
munications interests.
5.10.10 Roaming
Since 2007, all Commercial Mobile Radio Service (CMRS) carriers offering text
messaging, push-to-talk, and mobile voice and data services that interconnect
to the PSTN must permit subscribers of other CMRS networks to ‘roam’ onto
their networks on a non-d iscriminatory basis (with an exception for when the
networks are not technologically compatible).141 Following submission of the
National Broadband Plan to Congress, the FCC moved swiftly to impose an obliga-
tion to roam on all facilities-based providers of commercial mobile data services,
including those that do not interconnect with the PSTN.142 Subject to certain ex-
ceptions, facilities-based providers must enter into roaming arrangements with
other providers on commercially reasonable terms and conditions. It provided
further guidance on the commercially reasonable standard in a December 2014
declaratory ruling.143 Within the US, roaming is no longer a significant consumer
issue. Most carriers have implemented ‘nationwide’ plans and no longer charge
for domestic roaming, although off-network access often does not provide the full
breadth of services available on the user’s home network and some carriers will
limit or even cap excessive data use while roaming. Moreover, the prevalence of
WiFi access offers an acceptable substitute in many cases. The experience is mark-
edly different for travellers who leave the US, who remain subject to high inter-
national roaming charges for voice and data use.
141
Reexamination of Roaming Obligations of Commercial Mobile Radio Service Providers, Report and
Order and Further Notice of Proposed Rulemaking, WT Docket No 05–265, FCC No 07–143, 22 FCC Rcd 15817
(2007). The obligation arises from the FCC’s interpretation of §§201 and 202 of the 1934 Act.
142
Reexamination of Roaming Obligations of Commercial Mobile Radio Service Providers and Other
Providers of Mobile Data Services, Second Report and Order, WT Docket No 05–265, FCC No 11–52, 26 FCC
Rcd 5411 (2011).
143
Reexamination of Roaming Obligations of Commercial Mobile Radio Service Providers and Other
Providers of Mobile Data Services, Declaratory Ruling, WT Docket No 05-265, DA 14-1865, 29 FCC Rcd 15483
(2014).
528
144
Stated by Garnham, N, ‘Universal Service’, in Telecom Reform (ed Melody) (Technical University of
Denmark, 1997), at 207. See also Chapter 4, at Section 4.8.
259
5 US Telecommunications Law 259
a telephone company would provide all who wanted service in an area in return
for continued regulation as the sole service provider in a given area. Later, as uni-
versal service came to represent the policy that all Americans should have basic
telephone access at a reasonable rate, the primary issues related to the subsidy of
high-cost users (such as rural and residential customers) by low-cost users (such
as urban and business customers).
Some eighty years after the concept was first adopted, Congress finally wrote the
principle of universal service into law by enacting the Telecommunications Act
1996 which added §254 to the 1934 Act. However, Congress declined specifically
to define universal service, instead recognizing it as an ‘evolving level of telecom-
munications services’. When determining the services that should be provided
universally under §254, the FCC is required to take into account ‘advances in tele-
communications and information technologies and services’ and consider four
factors, including, for example, whether a particular service is essential to edu-
cation, public health, or public safety and a ‘substantial majority’ of residential
customers have subscribed to the service. Universal service policy must also be in-
formed by seven broad principles, such as that quality service should be available
at just, reasonable, and affordable rates; that access to advanced telecommunica-
tions and information services should be provided in all regions of the US; and that
low-i ncome consumers and consumers in rural and other high-cost areas should
have access to telecommunications services of the same quality and at the same
rates as those provided to consumers in urban areas. In addition, the 1996 Act ex-
panded the concept of universal service to include the principle that health care
providers in rural areas, schools, and libraries should have access to advanced
telecommunications services, such as the internet and other broadband services.
Following the adoption of the Act, the FCC developed four universal service
programmes to implement §254: (1) a programme for low-income users; (2) the
high-cost fund for rural communities; (3) the schools and libraries programme;
and (4) a programme for public and non-profit rural health care providers. All four
schemes were developed with the assistance of the Federal-State Joint Board, a
body comprised of FCC commissioners, PUCs, and a state-appointed utility con-
sumer advocate. The function of the Joint Board is to keep universal service policy
and related mechanisms under review and make recommendations from time to
time to the FCC.
Until the adoption of the American Recovery and Reinvestment Act of 2009 (the
Recovery Act), the principal focus of the FCC’s four universal service programmes
was to ensure access to voice services. The schools and libraries programme (also
known as the ‘e-rate’ programme) and the rural health care programme sup-
ported access to the internet, but most universal service funding was spent on
supporting voice services. Following the adoption of the Recovery Act, however,
620
the focus of all four universal service programmes has shifted to ensuring access
to broadband services. The Recovery Act required the FCC to develop a National
Broadband Plan (NBP) that sought to ‘ensure that all people of the United States
have access to broadband capability’. The term broadband was not defined in the
Recovery Act, but the plan the FCC submitted to Congress in March 2010 set a
target of ensuring all households and business had affordable broadband access
with an actual download speed of at least 4 Mbps and actual upload speed of at
least 1 Mbps by 2020. The NBP also included a number of recommendations to
ensure that schools, libraries, and health care facilities had the high-speed broad-
band services and facilities needed for the twenty-fi rst century.
Since submission of the NBP, the FCC has adopted ‘support for advanced
services’ as a universal service principle which it must take into account when
formulating universal service policy.145 Moreover, the FCC continues to revise its
definition of advanced services. In 2016, for example, the Commission made uni-
versal service funding from the high-cost programme available to certain carriers
on the condition they offer broadband services with a minimum download speed
of 25 Mbps and a minimum upload speed of 3 Mbps.146 The Commission has also
adopted numerous measures to reform its universal service programmes, but
many of the programmes are still in a state of transition. For the time being, they
support voice; broadband services; and bundled voice and broadband services.
However, the FCC intends to eliminate support for voice services in the future.
The current programmes, each of which is administered by the Universal Service
Administrative Company (USAC), an independent, not-for-profit corporation ap-
pointed by the FCC, in accordance with Part 54 of the Commission’s rules, are dis-
cussed below.147
145
See Connect America Fund, Report and Order and Further Notice of Proposed Rulemaking, WC Docket
Nos 10-9 0, 07-135, 05-337, 03-109, GN Docket No 09-51, CC Docket No 01-92, 96-45, WT Docket No 10-208, FCC
No 11-161, 26 FCC Rcd 17663 (2011). Under §254(b)(7), the FCC may adopt other universal service principles to
protect ‘the public, convenience and necessity’.
146
See Connect America Fund, Report and Order, Order and Order on Reconsideration, and Further Notice
of Proposed Rulemaking, WC Docket Nos 10-9 0, 14-58; CC Docket No 01-92, FCC No 16-33, 31 FCC Rcd 3086
(2016).
147
For a summary of the programmes prior to the adoption of the NBP and the reforms needed to them as a
result of the NBP, see c hapter 5 of the previous edition of this book.
216
5 US Telecommunications Law 261
comprise all voice telephony services (fixed and mobile) and, since 2016, broad-
band services (fixed and mobile) and bundled voice and broadband services.
Current rules dictate that from 2021, Lifeline services will consist only of broad-
band services (fixed and mobile) and bundled voice and broadband services
that meet minimum service standards.148 It appears that these rules will be un-
affected by the FCC’s 2017 decision to reclassify broadband services as informa-
tion services. The FCC has stated that it will continue (and has the legal authority
under §254(e) of the 1934 Act149) to maintain support for broadband services in the
Lifeline programme.150
All Lifeline services are supplied by ‘eligible telecommunications carriers’
(ETCs). Voice providers are designated as ETCs by PUCs. At the time of writing,
broadband providers are designated as ETCs by the FCC, but it is intended that
they will be designated by PUCs in the future.151 ETCs must provide the speci-
fied services in accordance with standards set by the FCC. At the time of writing,
fixed broadband services must have a minimum download speed of 10 Mbps and
a minimum upload speed of 1 Mps; mobile broadband services must have speeds
of 3G mobile technology or better. The minimum data usage for fixed broadband
services is 150 GB per month. For mobile broadband services, it is 500 MB per
month, but it will rise to 2 GB per month by December 2018.
In addition to the Lifeline programme, there is the Link Up programme. It pro-
vides eligible low-income subscribers living on Native American land with dis-
counts on the installation costs associated with Lifeline services and allows them
to defer payment on any remaining charges.
Over the last few years, the FCC has been particularly concerned with stopping
waste, fraud, and abuse of the low-income scheme. To that end, it has required
USAC to develop a National Lifeline Accountability Database. The database con-
tains subscriber information provided by ETCs and is used to identify subscribers
receiving Lifeline services from more than one ETC.152 The FCC has also adopted
national eligibility criteria for subscribers. Until 2012, PUCs were permitted to
adopt eligibility criteria. In 2016, the FCC directed USAC to establish a National
Lifeline Eligibility Verifier, whose function is to determine if subscribers meet
148
Lifeline and Link Up Reform and Modernization, Third Report and Order, Further Report and Order, and
Order on Reconsideration, WC Docket Nos 11-42, 09-197, 10-9 0, FCC No 16-38, 31 FCC Rcd 3962 (2016).
149
Section 254(e) states, ‘[a]carrier that receives such [universal service] support shall use that support only
for the provision, maintenance, and upgrading of facilities and services for which the support is intended.’
150
See Restoring Internet Freedom, above n 39, para 193.
151
Bridging the Digital Divide for Low-I ncome Consumers, WC Docket Nos 17-2 87, 11-42, 09-197, FCC No
17-155, 32 FCC Rcd 10475 (2017).
152
Lifeline and Link Up Reform and Modernization, Report and Order and Further Notice of Proposed
Rulemaking, WC Docket Nos 11-42, 03-109, 12-2 3, CC Docket No 96-45, FCC No 12-11, 27 FCC Rcd 6656 (2012).
62
153
See n 145 above. The order was challenged but was eventually upheld by the Court of Appeals. See Direct
Communications Cedar Valley v FCC, 753 F 3d 1015 (2014).
154
Connect America Fund, Report and Order, WC Dockets Nos 10-9 0, 14-58, 14-192, FCC No 14-190, 29 FCC
Rcd 15644 (2014).
623
5 US Telecommunications Law 263
forward-looking cost model known as the Connect America Cost Model (CAM)155
and will be paid annually over six years. In 2016, the FCC announced a similar
scheme for rate-of-return carriers, although they will be funded for ten years and
their roll-out obligations will differ.156 In 2017, the Commission launched MF Phase
II during which the FCC will offer wireless carriers $453 million in financial sup-
port annually over a period of ten years. In exchange, wireless carriers must deploy
4G LTE services.157
The FCC continues to support price cap carriers who provide voice services in
high-cost areas, although the amount of money received by price cap carriers has
been frozen since 2011 and will continue to be reduced over time. Rate-of-return
carriers also remain eligible for CAF Broadband Loop Support (BLS) (formerly
known as Interstate Common Line Support), provided there is no other unsub-
sidized competitor in the area served, and High Cost Loop Support (HCLS), which
supports voice services. However, the FCC plans to abolish BLS and HCLS and de-
velop a single CAF programme for rate-of-return carriers.
A notable development across the high-cost programmes has been the FCC’s
use of reverse auctions to award available funding. In a reverse auction, carriers
bid for the subsidy they need to provide services in specified areas. The winning
bidder is the carrier that needs the smallest subsidy. Reverse auctions were first
used in MF Phase I and will be used in MF Phase II. Moreover, the FCC is already
planning the first CAF Phase II auction158 and intends that all available funds will
be awarded by a competitive bidding process.
5.11.2.3 E-rate
With a budget of $3.9 billion, the ‘e-r ate’ programme provides discounts of be-
tween 20 to 90 per cent on communications services specified by the FCC to
schools and libraries. The precise discount a school or library receives depends
on the household income levels of students in the community and whether the
school or library is in an urban or rural area. The services that currently attract
a discount include voice services, data transmission services, internet access,
internal connection services and equipment necessary for high-speed broad-
band connectivity, related maintenance and managed internal broadband
155
Connect America Fund, Report and Order, WC Dockets Nos 10-9 0, 05-337, DA No 14-534, 29 FCC Rcd
3964 (2014).
156
See n 148 above.
157
Connect America Fund, Report and Order and Further Notice of Proposed Rulemaking, WC Docket No
10-9 0, WT Docket No 10-208, FCC No 17-11, 32 FCC Rcd 2152 (2017).
158
Connect America Fund, Report and Order and Further Notice of Proposed Rulemaking, WC Dockets
Nos 10-9 0, 14-58, 14-259, FCC No 16-6 4, 31 FCC Rcd 5949 (2016); Connect America Fund, Report and Order and
Order on Reconsideration, WC Dockets Nos 10-9 0, 14-5, FCC No 17-12, 32 FCC Rcd 1624 (2017).
624
services, such as WiFi. However, the FCC plans to eliminate all support for voice
services in the near future, so that all funding can be directed to achieving the
programme’s principal goal: ensuring ‘affordable access to high-speed broad-
band sufficient to support digital learning in schools and robust connectivity
for libraries’.159
The FCC continues to work towards the high-speed internet access and WAN
connectivity targets it set for schools and libraries in July 2014. The Commission
has, for example, amended its rules to permit schools and libraries to build their
own broadband facilities in certain circumstances and has required recipients of
high-cost programme funding to offer high-speed broadband services to schools
and libraries.160 For schools, the high-speed internet access target is 1 Gbps per
1,000 students and staff. Libraries serving fewer than 50,000 people must have
internet access with speeds of at least 100 Mbps. For libraries serving more than
50,000 people, the internet access target is 1 Gbps. The WAN connectivity target for
schools is 10 Gbps per 1,000 students and staff.
159
Modernizing the E-rate Program for Schools and Libraries, Report and Order and Further Notice of
Proposed Rulemaking, WC Docket No 13-184, FCC No 14-99, 29 FCC Rcd 8870 (2014).
160
Modernizing the E-rate Program for Schools and Libraries, Second Report and Order and Order on
Reconsideration, WC Docket Nos 13-184, 10-9 0, FCC No 14-189, 29 FCC Rcd 15538 (2014).
161
Rural Health Care Support Mechanism, Report and Order, WC Docket No 02-6 0, FCC No 12-150, 27 FCC
Rcd 16678 (2012).
162
47 CFR §54, Subpart G.
265
5 US Telecommunications Law 265
5.11.3 Funding
Prior to the 1996 Act, universal service was paid for predominantly by AT&T and
other large long-d istance providers. Under the 1996 Act, all telecommunications
carriers providing interstate and international telecommunications services to
the public and other designated providers must pay towards the cost of universal
service unless their contribution is less than US$10,000.163 The meaning of inter-
state and international services is broad and encompasses satellite, mobile, and
payphone services. Telecommunications carriers and interconnected VoIP pro-
viders164 pay quarterly charges towards universal service provision to USAC, which
in turn makes payments in support of the universal service fund programmes.
Contributions are calculated by multiplying the projected revenues for inter-
state and international telecommunications services of a provider (less its pro-
jected universal service contribution) by a ‘contribution factor’.165 Quarterly, the
FCC determines the contribution factor, which is the ratio of the total projected
costs of the four universal service programmes for that period and the total pro-
jected revenue for all interstate and international telecommunications services
offered by all providers who must contribute to the scheme less their total esti-
mated universal service contributions.166 Providers are permitted to pass on some
of these charges to their customers.167 Revenue from interstate and international
telecommunications services has, however, been falling for years due to rigorous
competition for long-d istance services and the use of IP-enabled technologies. The
bundling of interstate and intrastate services and other products and the avail-
ability of wireless packages that enable users to make long-d istance calls at no
additional cost have also made it difficult to determine if all interstate revenue is
being fully counted.
As early as 2001 and 2002, the FCC considered alternatives to its current ap-
proach in an attempt to increase available funding and to ensure that telecommu-
nications carriers contribute to the fund equally.168 All involved the imposition of a
flat-rate fee on providers, but the criterion used to impose the flat-rate fee differed.
47 CFR §54.706(a).
163
Universal Service Contribution Methodology, Report and Order and Notice of Proposed Rulemaking,
164
WC Docket No 06–122, CC Docket Nos 96–45, 98–171, 90–571, 92–2 37, 99–200, 95–116, 98–170, WC Docket Nos
06–122, 04–36, FCC No 06–94, 21 FCC Rcd 7518 (2006) (upheld in Vonage Holdings Corp v FCC, 489 F 3d 1232
(DC Cir 2007)).
165
47 CFR §54.709(a) and (a)(1). 166
47 CFR §54.709(a)(2). 167
47 CFR §54.712(a).
168
Federal-State Joint Board on Universal Service, Notice of Proposed Rulemaking, CC Docket Nos 96–
45, 98–71, 90–571, 92–2 37, 99–200, 95–116, 98–170, FCC No 01–145, 16 FCC Rcd 9892 (2001); Further Notice of
Proposed Rulemaking and Report and Order, CC Docket Nos 96–45, 98–71, 90–571, 92–2 37, 99–200, 95–116,
98–170, FCC No 02–43, 17 FCC Rcd 3752 (2002); and Report and Order and Second Further Notice of Proposed
Rulemaking, CC Docket Nos 96–45, 98–71, 90–571, 92–2 37, 99–200, 95–116, 98–170, FCC No 02–329, 17 FCC Rcd
24952 (2002).
62
Under the first approach, the fee would be imposed for each residential, single-l ine
business, payphone, mobile wireless, and pager connection. Contributions in the
second approach were determined by the maximum capacity of a customer’s con-
nection. The third approach imposed a fee for each telephone number assigned
to a customer and the capacity of any connection for customers not allocated any
numbers. In 2008, the FCC sought comment on two variations of these methods.169
In the first option, subject to certain exceptions, any provider (other than a wire-
less prepaid provider) who assigned a telephone number to a residential customer
would pay US$1 per month for each number. Fees for wireless prepaid providers
would be determined by an alternative formula that took into account the number
of monthly minutes generated by the provider. All providers of business services
would be required to pay a fee based on the number of businesses connected to the
public network. In the second option, arrangements similar to those in the first op-
tion would apply for residential customers, although the amount contributed per
number would be US$.85 per month. The contribution for providers of business
services would be determined by the number and capacity of the dedicated access
connections used by their business customers. Each connection with a capacity of
up to 64 kbps would incur a charge of US$5; each connection in excess of 64 kbps
would incur a charge of US$35.
The NBP submitted to Congress in 2010 included a recommendation that the
contribution base for universal service be broadened. In response, the FCC again
sought comment on who should contribute to the universal service fund and
how contributions should be assessed.170 It asked if providers of certain specified
services, such as text messaging and broadband internet access services, should
contribute; or if all providers of interstate information and telecommunications
services, subject to some exclusions, should contribute. It also asked if contribu-
tions of providers should be assessed by reference to their revenue; the number of
connections to communications networks they provide to customers; the amount
of telephone numbers assigned to them; or by either of the two approaches on
which it consulted in 2008. However, the FCC made no changes to its rules.
In 2014, the FCC asked the Federal State Joint Board to make recommendations
as to how it should modify its universal service contribution methodology by 7
April 2015.171 Shortly before the Joint Board’s report was due, broadband services
were reclassified as telecommunications services. Consequently, pursuant to Title
169
High-Cost Universal Service Support, n 114, Appendix A paras 92–156; Appendix B paras 39–104;
Appendix C paras 88–151. Appendix A and Appendix C are in all material respects the same.
170
Universal Service Contribution Methodology, Further Notice of Proposed Rulemaking, WC Docket No
06-122, GN Docket 09-51, FCC No 12-4 6, 27 FCC Rcd 5357 (2012).
171
Federal State Joint Board on Universal Service, Order, WC Docket Nos 96-45, 06-122 and GN Docket No
09-51, FCC No 14-116, 29 FCC Rcd 9784 (2014).
627
5 US Telecommunications Law 267
II, the FCC could have required broadband providers to contribute to the universal
service fund. However, the Commission decided to forbear from enforcing the
contribution obligations and said it would revisit its decision following receipt of
the Joint Board’s report.
At the time of writing, the Joint Board has still not produced a report; broadband
services have been reclassified as information services; and the complex issues
surrounding how best to secure funding for the universal service programmes re-
main unresolved. In the absence of any decision to expand the contribution base,
the quarterly contribution factor used to calculate contributions paid by tele-
communications carriers continues to rise. In the first quarter of 2017, it was 16.7
per cent.
of conduct caught by §1: conduct that is ‘per se illegal’ and conduct which
violates the so-c alled ‘rule of reason’. An example of conduct which is ‘per
se illegal’ is price fixing. Conduct is contrary to the rule of reason when it is
otherwise lawful but unreasonably restrains trade. Such conduct is reviewed
on a case-b y-c ase basis and in light of its pro-and anti-c ompetitive effects on
relevant market(s).
In 2000, the DoJ and the FTC issued ‘Antitrust Guidelines for Collaborations
among Competitors’ which sets out the general principles that both agencies will
use when reviewing agreements between competitors and the potential com-
petition concerns likely to arise from collaboration. Section 1 of the Sherman
Anti-t rust Act does not expressly permit either the DoJ or the FTC to exempt anti-
competitive behaviour from the Act where it ‘contributes to improving the pro-
duction or distribution of goods or to promoting technical or economic progress,
while allowing consumers a fair share of the resulting benefit . . .’; nevertheless, in
practice, the pro-competitive effects of agreements will be taken into account by
the courts, the DoJ, and the FTC when assessing whether or not an agreement un-
reasonably restrains trade.
Section 2 of the Sherman Anti-t rust Act makes it unlawful for natural persons
and legal entities to monopolize any part of trade or commerce between the fifty
states and foreign countries. Monopolies are not per se illegal but where a party
has acquired or intends to acquire market power through anti-competitive means
then a violation of §2 of the Sherman Anti-t rust Act will occur. Where trade with
foreign countries is involved, §§1 and 2 of the Sherman Anti-t rust Act are not vio-
lated unless the conduct has a direct, substantial, and reasonably foreseeable
effect on domestic trade or commerce, or on export trade or commerce of a person
engaged in such trade or commerce in the US.
Of the two provisions, it is the application of §2 of the Sherman Act by the DoJ
which has had greater effect on the telecommunications industry due to AT&T’s
historical monopoly and the degree of market power ILECs hold in local access
markets.
5 US Telecommunications Law 269
5.12.4 Investigations
The DoJ and FTC may initiate investigations into alleged anti-competitive prac-
tices following internal reviews by in-house economists and lawyers and com-
plaints from industry participants, concerned citizens, informants, and other
174
Department of Justice, Corporate Leniency Policy, available at <http://w ww.usdoj.gov/atr/public/g uide-
lines/0 091.htm>.
175
Department of Justice, Leniency Policy for Individuals, available at <http://w ww.usdoj.gov/atr/public/
guidelines/0 092.htm>.
176
Clayton Act, §4(a), 15 USC §15.
270
government agencies. Both may compel legal and natural persons to produce in-
formation and other documentation relevant to a civil investigation by serving a
Civil Investigative Demand (CID). They may request in a CID for a witness to give
oral testimony or to answer questions in writing. Evidence in relation to criminal
investigations is gathered by the DoJ pursuant to subpoenas issued by a grand
jury. Assistance from the Federal Bureau of Investigation and other federal agen-
cies may also be requested during investigations.177
5.12.6 Mergers
Section 7 of the Clayton Act prohibits the acquisition of shares or capital in a nat-
ural or legal entity engaged in commerce or related activities affecting commerce
where such acquisition may be ‘substantially to lessen competition, or to tend to
create a monopoly’. This provision covers mergers, asset and share purchases, joint
ventures, and other acquisitions. The DoJ is responsible for ensuring that mergers
in the telecommunications sector comply with §7. Reviewing mergers of cable op-
erators and ISPs is the duty of the FTC.
177
For further information about the procedural matters surrounding a DoJ investigation, see the DoJ’s
Antitrust Division Manual available at <http://w ww.justice.gov/atr/public/d ivisionmanual>. For a copy of the
FTC’s manual, see <http://w ww.ftc.gov/foia/ch03investigations.pdf>.
271
5 US Telecommunications Law 271
The DoJ and the FTC have issued a number of guidelines relating to horizontal
and non-horizontal mergers. The Horizontal Merger Guidelines were first issued in
1992 and were revised in 1997 and 2010. The Guidelines focus on market definition
and concentration, the potential adverse competitive effects of mergers, the ability
of new participants to enter the relevant market(s), efficiencies resulting from the
merger, and the likelihood of either party to the merger failing if the merger does
not take place. The Non-Horizontal Merger Guidelines were issued in 1984. Non-
horizontal or vertical mergers are less likely to give rise to competition concerns
but the guidelines set out the principal theories under which either the DoJ or FTC
would challenge non-horizontal mergers. These include the elimination of actual
or potential competition, the creation of barriers to entry, and the ability of vertical
mergers to facilitate collusion at retail levels and in downstream markets.
Since 1978, unless exempted by §7A(c) of the Clayton Act, mergers which meet
specified criteria must be notified prior to the consummation of the transaction
to both the DoJ and the FTC.178 As a general rule, the purchaser and the target of
the acquisition must notify the DoJ and the FTC if the following conditions179 are
satisfied:
1. if one person has sales or assets of at least US$100 million; the other person has
sales or assets of at least US$10 million; and as a result of the transaction the ac-
quiring person will hold an aggregate amount of stock and assets valued at more
than US$50 million of the acquired person but less than US$200 million; or
2. as a result of the transaction, the acquiring person will hold an aggregate amount
of stock or assets of the acquired person valued at more than US$200 million re-
gardless of the sales or assets of the acquiring and acquired person.
Failure to notify may result in civil penalties or a court order requiring the parties
to divest any assets acquired in violation of the Clayton Act if the transaction is
already consummated.
Notifying parties must complete and have certified a Notification and Report
Form, which requires the notifying parties to provide details about the transaction
and information about acquisitions made in the last five years. The notifying par-
ties must inform the DoJ and FTC if the acquiring person and acquired entity earn
revenue from businesses that fall within any of the same industry and product
codes in accordance with the North American Industry Classification System and,
if so, the geographic areas in which they operate. They must also pay the appro-
priate filing fee which is determined by reference to the value of the transaction.
See the Clayton Act §7A(a) and the Premerger Notification Rules found at 16 CFR Parts 801, 802, and 803.
178
5 US Telecommunications Law 273
184
Promoting Efficient Use of Spectrum Through Elimination of Barriers to the Development of Secondary
Markets, Second Report and Order, Order on Reconsideration, and Second Further Notice of Proposed
Rulemaking, WT Docket No 00–2 30, FCC No 04–167, 19 FCC Rcd 17503 (2004).
247
public opposition that might jeopardize the likelihood of receiving a majority vote.
However, such practices have come under sharp criticism from some Republican
members of the FCC, including now Chair Ajit Pai. They claim that these condi-
tions do not relate to the merits of the transaction, but instead are designed to im-
pose regulations that further the majority members’ policy goals.
The FCC also has powers under §§7 and 11 of the Clayton Act to block acquisi-
tions of common carriers engaged in fixed and radio communications where the
effect of the acquisition would be to substantially lessen competition or create a
monopoly. In practice, these powers are rarely used and any action taken by the
FCC is based on its powers set out in the Communications Act of 1934.
185
See further Chapter 13.
275
5 US Telecommunications Law 275
costs and benefits associated with the regulation and the burden on commercial
speech it imposes.
Rules and Regulations Implementing the Telephone Consumer Protection Act of 1991, Report and Order,
187
proceeding maximizing consistency with the FTC’s rules, and called for annual
reports from both the FTC and FCC.
The FCC’s Report and Order of 26 June 2003 set out rules that supplemented
the FTC’s own telemarketing rules. Collectively, the FCC and FTC rules estab-
lished a single national database of fixed and mobile telephone numbers of sub-
scribers who object to receiving telephone solicitations. The FCC’s 26 June 2003
Report and Order also adopted more stringent measures to combat the increased
use of predictive diallers and required callers to display caller identification; and
concluded that the TCPA protects wireless subscribers from receiving unwanted
voice calls as well as text messages.190 It is worth noting that the FCC’s jurisdiction
over telemarketers is broader than the FTC’s jurisdiction. The FTC has no over-
sight of common carriers,191 banks, credit unions, savings and loans, insurance
providers, airlines, and intrastate telemarketing calls.192 The FCC rules apply to
all of these entities as well as interstate and intrastate telemarketing calls. The
FTC has effectively become the lead agency for administration and promotion of
the registry, as well as the key source of public outreach and consumer support
in this area. Because of the FCC’s more extensive jurisdiction in this area, how-
ever, the FCC, which has the power to impose fines on rule violators, has taken a
strong enforcement role.
As a result of the scheme, fixed and wireless subscribers now have three options
to preserve their privacy by:
190
On the issue of unwanted text messages, see also Satterfield v Simon & Schuster, 569 F 3d 946 (9th Cir
2009) and Rules and Regulations Implementing the Telephone Consumer Protection Act of 1991, Declaratory
Ruling and Order, CG Docket Nos 02-278, WC Docket No 07-135, FCC No 15-72, 30 FCC Rcd 7961 (2015).
191
See further Section 5.4.3. 192
15 USC §45(a)(2).
193
See, eg, Rules and Regulations Implementing the Telephone Consumer Protection Act of 1991,
above n 190.
72
5 US Telecommunications Law 277
had established business relationships with customers from respecting the do-not-
call list. However, this exemption was abolished in 2012 and replaced with a require-
ment that the express written consent of the called party has to be obtained prior
to making a call. Moreover, since 2012, the FCC requires callers to provide called
parties with an automated, interactive mechanism whereby they can opt out of re-
ceiving pre-recorded messages. Similarly, telemarketers cannot abandon more than
3 per cent of calls answered by customers for the duration of any calling campaign.194
Federal legislation has also resulted in modifications to the do-not-call list and
related TCPA provisions. When the Do-Not-Call Implementation Act was adopted,
numbers could be kept on the list for only a limited period of time. However, in
2008, Congress decided numbers could be kept on the list indefinitely.195 In 2009,
Congress passed the Truth in Caller ID Act. This legislation responded to concerns
that it had become increasingly easy for parties to alter the phone number dis-
played with a call (the ‘Caller ID’) to make it appear that the call was coming from
any number, and that such spoofing activity was being used to trick and defraud
consumers. In implementing this legislation, the FCC prohibited persons from
causing the display of inaccurate caller identification information for the pur-
poses of defrauding, harming, or otherwise obtaining anything of value, and ap-
plied the rule to calls made using any telecommunications or interconnected VoIP
service.196 Under these rules, callers may continue to choose to block their call
information, although telemarketers must transmit and display a valid number
in conjunction with their calls. In November 2017, the FCC adopted additional
rules to combat the continuing problems of spoofing and unwanted telemar-
keting calls.197 In particular, it now expressly permits providers of voice services
to block calls originating from any number at the request of a subscriber and calls
originating from certain numbers specified by the FCC. The new rules follow re-
quests for legal clarification from the ‘Robocall Strike Force’,198 a body established
by and comprised of industry representatives to address the problems of spoofing
and unwanted telemarketing calls.
194
See Rules and Regulations Implementing the Telephone Consumer Protection Act of 1991, Report and
Order, CG Docket No 02-278, FCC No 12-21, 27 FCC Rcd 1830 (2012).
195
Do-Not-Call Improvement Act of 2007, Pub L No 110-187, 122 Stat 633 (2008).
196
Rules and Regulations Implementing the Truth in Caller ID Act of 2009, Report and Order, WC Docket No
11–39, FCC No 11–100, 26 FCC Rcd 9114 (2011).
197
Advanced Methods to Target and Eliminate Unlawful Robocalls, Report and Order and Further Notice of
Proposed Rulemaking, CG Docket No 17-59, FCC No 17-151, 32 FCC Rcd 9706 (2017).
198
See, eg, Robocall Strike Force Report (26 October 2016) available at <https://t ransition.fcc.gov/cgb/
Robocall-Strike-Force-Final-Report.pdf>.
278
5.13.3 Unsolicited faxes
The TCPA, as amended by the Junk Fax Prevention Act199 enacted in 2005, makes it
unlawful for any person within the US, or any person outside the US if the recipient
is within the US, to use a fax machine, computer, or other device to send unsolicited
advertisements to a fax machine. The prohibition on sending unsolicited fax advert-
isements does not apply where recipients have given senders permission; or senders
have an established business relationship and recipients voluntarily provide them
with their fax numbers or have otherwise made their fax numbers publicly available
in a directory, advertisement, or website. The Act requires any person sending an
unsolicited fax advertisement under an exemption to include a notice that complies
with any applicable rules mandated by the FCC. Current rules require senders to in-
form recipients of their ability to and the means by which they can avoid unsolicited
faxes.200 In 2006, the FCC imposed similar opt-out requirements for persons sending
solicited fax advertisements, a decision it confirmed in 2014,201 but the requirements
were found to be unlawful in March 2017.202 As is the case for unsolicited calls and
texts, failure to comply with the relevant provisions of the TCPA may result in fines
imposed by the FCC. Violations of the fax rules also give subscribers a right to sue
the offending party in state court. Public dissatisfaction with junk faxes, additional
Congressional oversight, and continued FCC interest in this area have resulted in a
steady stream of FCC enforcement actions against violators. While forfeitures are
often in the thousands of dollars, large-scale violations have resulted in forfeitures
as high as US$5 million against a single company.
5.13.4 Unsolicited email
The provisions of the TCPA do not apply to unsolicited email that advertises or
promotes commercial products and services. To address the loopholes in the le-
gislation and consumer concerns about ‘spam’, Congress enacted the Controlling
the Assault of Non-Solicited Pornography and Marketing Act of 2003 or the
CAN-SPAM Act of 2003.203 The Act came into force on 1 January 2004 and makes
it unlawful to send to a computer commercial email that contains deceptive or
misleading subject headings or false information about the origin of an email. It
199
Pub L No 109-21, 119 Stat 359 (2005).
200
See Rules and Regulations Implementing the Telephone Consumer Protection Act of 1991, Report and
Order and Third Order on Reconsideration, CG Docket Nos 02-278, 05-338, FCC No 06-42, 21 FCC Rcd 3787
(2006); 47 CFR §64.1200(a)(4).
201
Rules and Regulations Implementing the Telephone Consumer Protection Act of 1991, Order, CG Docket
Nos 02-278, 05-338, FCC No 14-164, 29 FCC Rcd 13998 (2014).
202
See Bais Yaakov of Spring Valley v FCC, No. 14-1234 (DC Cir 31 March 2017).
203
Pub L No 108–187, 117 Stat 2699 (2003).
279
5 US Telecommunications Law 279
requires senders of commercial emails to notify recipients that such emails are
advertisements, to provide the sender’s physical postal address, and to give recipi-
ents an opportunity to ‘opt-out’ of receiving commercial emails in the future. The
opt-out notice must contain a return email address or other internet-based mech-
anism by which email recipients can indicate they do not wish to receive future
emails. The opt-out mechanism selected by a sender must remain active for thirty
days. In addition, the Act prohibits senders and anyone acting on their behalf from
sending commercial emails to a recipient who requests not to receive subsequent
emails. The prohibition begins ten business days from the date of the recipient’s
request. Violation of the civil provisions of the Act may lead to injunctive relief
and/or statutory fines. Breaches of criminal law may lead to imprisonment and
fines up to US$6 million. The FTC has primary responsibility for implementing
and enforcing the legislation.
In addition, the FCC adopted rules that took effect in 2005 that prohibit the
sending of unwanted commercial email messages to wireless devices without the
prior permission of subscribers.204 To facilitate compliance with the rules, it also
established a list of domain names typically used to send messages to wireless de-
vices. The CAN-SPAM Act of 2003 directed the FCC to issue regulations protecting
consumers from ‘unwanted mobile service commercial messages’.
204
Rules and Regulations Implementing the Controlling the Assault of Non-Solicited Pornography and
Marketing Act of 2002, Order, CG Docket Nos 04-53, C02-278, FCC 04-194, 19 FCC Rcd 15927 (2004).
205
Pub L No 106–81, 113 Stat 1286 (1999).
820
carrier may use, disclose, or permit access to ‘individually identifiable’ CPNI only
as required by law or with the approval of their customers. In addition, §222(c)(1)
of the 1934 Act permits a carrier to use, disclose, or permit access to individually
identifiable CPNI to provide the telecommunications service from which such in-
formation is derived; or to provide services necessary to, or used in, the provision
of that telecommunications service.
There are, however, a number of exceptions to the confidentiality obligations of
telecommunications carriers. Use, disclosure, and access to CPNI is permissible
by a carrier, either directly or indirectly through its agents, for such purposes as
billing and debt collection, the protection of a carrier’s rights and property, the
provision of telemarketing and referral services requested by subscribers, and
the provision of call location information concerning users of commercial mo-
bile services in specified emergency situations. Disclosure of CPNI upon the af-
firmative written request by a customer is permitted. CPNI that is ‘aggregated’ and
does not contain individually identifiable CPNI may be disclosed to any party.206
Moreover, telecommunications carriers that provide telephone exchange services
must disclose subscriber list information to any person upon request for the pur-
pose of publishing telephone directories.207
The FCC has adopted and revised rules implementing these substantive confi-
dentiality provisions on numerous occasions since 1996 to reflect changes in tech-
nology and market practices. In 2007, for example, the FCC extended its customer
proprietary information rules to providers of interconnected VoIP services.208
The FCC also revisited the rules in 2016 following its decision to reclassify broad-
band services as telecommunications services.209 When broadband services were
treated as information services, the FCC had no authority to regulate providers
of these services under §222 of the 1934 Act, because its jurisdiction is limited to
telecommunications carriers (as defined). The privacy practices of information
service providers and other entities providing broadband services were instead
regulated by the FTC. The FTC is responsible for enforcing §5 of the Federal Trade
Commission Act, a provision that prohibits unfair and deceptive trade prac-
tices, and one that the FTC has actively used to promote online privacy as part
of its broader mandate to protect consumer interests.210 When the FCC reclassi-
fied broadband services as telecommunications services, providers of broadband
services became telecommunications carriers and the FTC lost its authority to
206
47 USC §222(d). 207
47 USC §222(e).
208
Implementation of the Telecommunications Act of 1996, Report and Order and Further Notice of
Proposed Rulemaking, CC Docket No 96-115, WC Docket No 04-36, FCC No 07-22, 22 FCC Rcd 6927 (2007).
209
See Section 5.2.5.2.
210
For more information, see, eg, Solove, DJ and Hartzog, W, ‘The FTC and the New Common Law of
Privacy’, (2014) 114 Columbia Law Review 583.
281
5 US Telecommunications Law 281
regulate the privacy practices of these entities. The FTC has no jurisdiction over
common carriers or entities providing telecommunications services.211
The rules the Commission adopted in 2016, shortly before President Obama left
office, harmonized the privacy rules applicable to all telecommunications car-
riers. They imposed notice, customer approval, data security, and data breach no-
tification obligations on broadband providers. They also provided for protection of
precise geo-location information belonging to customers.212
However, the rules are no longer in effect. On 1 March 2017, following the in-
auguration of President Trump and the receipt of numerous petitions for recon-
sideration of the rules from industry, the Commission, operating with only three
Commissioners, issued a temporary stay of the data security obligation, which
required telecommunications carriers to take reasonable measures to protect
customer proprietary information from unauthorized use, disclosure, and ac-
cess.213 The stay was to remain in place until the Commission was able to recon-
sider the FCC’s rules in full, but on 3 April 2017, the Republican-controlled House
of Representatives and Senate adopted a joint resolution of disapproval under the
Congressional Review Act that President Trump signed the same day. The reso-
lution prevented the 2016 rules from taking effect.214
Adoption of the 2016 rules was so controversial because they were more onerous
than the approach applied by the FTC to information service providers such as
Google, Skype, and other so-called ‘edge providers’, providers of websites, web-
based email, applications, and search engines, even though it had been argued
that these entities pose a greater threat to consumer privacy than broadband pro-
viders. In addition, the FTC had argued that the FCC’s rules resulted in two dif-
ferent privacy frameworks, which was likely to generate confusion for consumers.
As a result of the Commission’s decision to reclassify broadband services as
information services in December 2017, broadband service providers are again
subject to the jurisdiction of the FTC and must therefore comply with §5 of the
Federal Trade Commission Act and related case law. Telecommunications carriers
providing telecommunications services remain subject to the jurisdiction of the
FCC and must comply with the customer proprietary information rules adopted
by the FCC prior to 2016 and codified in 47 CFR §§64.2001–6 4.2012. However, there
remains some concern that the FTC’s privacy framework is sufficiently robust to
211
See also Section 5.4.3.
212
Protecting the Privacy of Customers of Broadband and Other Telecommunications Services, Report and
Order, WC Docket No 16-106, FCC 16-148, 32 FCC Rcd 13911 (2016).
213
Protecting the Privacy of Customers of Broadband and Other Telecommunications Services, Order
Granting Stay Petition in Part, WC Docket No 16-106, FCC No 17-19, 32 FCC Rcd 1793 (2017).
214
The resolution also prevents the FCC from reissuing the rules in substantially the same form unless spe-
cifically authorized by a law enacted after the date of the joint resolution.
28
Part III
6 .1 INTRODUC TION
1
This is used here as shorthand for legal substantive and procedural requirements for fairness however
they arise whether pursuant to statutory obligation or otherwise. A contract can provide these, including
remedies such as early termination payments and what procedures will be used to resolve disputes or adju-
dicate breaches, eg arbitration procedures and rules as well as limitations on the reasons it can be abrogated
or breached.
286
might otherwise be reluctant to commit the capital required to roll out new tech-
nologies and/or networks to improve and update services. Without performance
obligations, countries might be unwilling to involve private parties in running
the state-owned incumbent.2 Licensing can also foster competitive markets by
imposing obligations on incumbents to level the playing field as well as ensure the
continuation of socially desirable services or outcomes such as disabled access or
universal service that competition will not.
Licensing is, therefore, an important regulatory tool in both developing mar-
kets and competitive markets although the same considerations may not equally
apply. For example, one of the most significant aspects of the current EU frame-
work is that, generally, electronic communications providers need not obtain in-
dividual licences requiring approvals to provide networks or services.3 Since the
2002 Authorisation Directive, a scheme of general authorizations applies to all
providers.4 EU Member States can subject these authorizations only to the defined
and limited set of general conditions it permits. Individual conditions can be ap-
plied only in certain circumstances, explored below. Individual grants of rights
may only be required for access to scarce resources, in the EU only radio spectrum
and numbers.
Licensing as a means to allocate, re-a llocate, and manage radio spectrum for
efficient telecommunications’ and other rapidly evolving, increasingly complex,
and potentially shared uses is an important issue in light of the unabated demand
and spectrum’s importance to an ever mobile, wireless, and digital society.5 Due
to these considerations and the distinct policy and licensing attributes that spec-
trum involves, it is considered in a separate chapter.6
This chapter examines the current EU framework for authorization of electronic
communications services and networks that, essentially, has been in place since
2002 albeit with some, more recent, harmonizing reforms intended to address on-
going Single Market and other concerns. It considers briefly the problems that the
2
For an examination of issues arising in developing markets, see Chapter 17.
3
This focuses primarily on the nature and scope of approvals needed to provide communications networks
and services. However, equipment intended to be attached to the network typically must also be ‘authorized’.
In many countries, it must meet type and safety requirements via an established approval process. Although
Chapter 4 (at Section 4.4.3) details the EU process and legislation governing this, the purpose of and legal jus-
tification for licences, inter alia, examined in this chapter, encompass public and consumer safety and extend,
of necessity, to the regulation of equipment used to provide communications networks and services. The dis-
cussion in this regard will make reference where applicable to this aspect of ‘authorization’.
4
Directive 2002/20/EC on the authorization of electronic communications networks and services, OJ L
108/21, 24 April 2002 (the Authorisation Directive).
5
See eg Report from Aspen Institute Roundtable on Spectrum Policy: ‘Revisiting Spectrum Policy: Seven
Years After the National Broadband Plan’ (Bollier D, Rapporteur, The Aspen Institute 2016), <http://c sreports.
aspeninstitute.org/Roundtable-on-Spectrum-Policy/2016/report/details/0227/Spectrum-2016>.
6
See Chapter 7.
287
2002 and the previous EU framework were designed to address, before discussing
some currently proposed reforms to address substantially the same concerns. The
chapter also explores the UK’s implementation of the EU regulatory scheme under
the Communications Act 2003 with the ensuing amendments and refining regula-
tion. The chapter will first, however, briefly examine the history and jurispruden-
tial underpinnings of licensing as it has evolved in England and the United States
and its use as a tool to both restrict markets and open them to competition.
7
The focus here primarily on law and principles derived from English common law does not suggest a lack
of comparable historical precedent within civil law systems.
8
See Bouvier’s Law Dictionary 711 (Baldwin WE, ed, Library Edition 1928). Accord, Black’s Law Dictionary
829–8 30 (6th edn 1979). The other secondary sense is that of the document that embodies these permissions
in writing. See, eg State ex rel Peterson v Martin, 180 Or 459, 474, 176 P 2d 636, 643 (1947); Mathias v Walling
Enterprises, 609 So 2d 1323, 1332 (Fla App 1992); 53 Corpus Juris Secundum, Licenses §2 (1983) (CJS). In tele-
communications licensing, the licensing document or certificate is often a complex writing with descriptions
of grants, rights, networks, approved equipment, and schedules of conditions and definitions. The two senses
are somewhat merged.
9
See eg Bouvier’s Law Dictionary, n 8. 10 Ibid.
11
Overlap exists with other contexts, eg in US communications licensing, some state and local franchise
and licence requirements stem from the use of public lands by communications companies. See Quirk, WJ, ‘A
Constitutional and Statutory History of the Telephone Business in South Carolina’, (2000) 51 South Carolina
L Rev 290, 293. Use of public land is among the privileges suggested as underlying the rationale for imposing
82
or carry on some trade or business, in its nature lawful but prohibited by statute
except with the permission of the civil authority or which otherwise would be un-
lawful’.12 It examines this in the context of England and its derivative common law
systems, primarily the US.
Public regulation of private parties providing goods or services, such as tele-
communications, stems partly from the common law’s imposition of greater duties
and legal obligations on so-called ‘public’ or ‘common’ callings.13 These selected
trades or undertakings that changed over time according to their economic neces-
sity14 and often scarcity, frequently comprising a monopoly,15 had a duty to serve
all members of the public on reasonable terms and with reasonable care.16 Inns
and carrier coaches that were essential17 to travel and travellers on the then few
roads were soon included within this group that is today referenced primarily by
the term ‘common carriers’, now far more limited.18 These are relevant here for sev-
eral reasons. Inns were the first post offices in Britain and, innkeepers, the first
postmasters in a system before that also ultimately governing telegraph and tele-
communications in the UK. Further, licensing of inns and common houses serves
to illustrate the nature and scope of licensing jurisdiction, generally, as exercised
over time in the UK and US. Finally, the ‘common carrier’ classification continues
regulation on common callings and undertakings granted privileges. See Burdick, C, ‘The Origin of the
Peculiar Duties of Public Service Companies’, (1911) 11 Colum L Rev 514, 616, 743.
12
Bouvier’s Law Dictionary, n 8.
13
See Wyman, B, ‘The Law of Public Callings as a Solution of the Trust Problem’ (1904) 17 Harv L Rev 156,
156–159 (suggesting that akin to this common law theory, enhanced duties be placed on monopolies in light of
their privileges and their economic necessity to society).
14
That the scarcity of inns and their importance to Britain’s emerging internal trade was a factor critical
to their regulation is suggested by the fact that prior to their inclusion as a common calling, they could be
indicted as a public nuisance ‘if it was set up where it was not needed’. Webb, S, and B, ‘The First Century of
Licensing’ in The History of Liquor Licensing in England, at 5, n 1 (Longmans, Green & Co, 1903).
15
‘The rule that one who pursued a common calling was obliged to serve all comers on reasonable terms
seems to have been based on the fact that innkeepers, carriers, farriers, and the like, were few, and each had
a virtual monopoly in his neighborhood.’ Wilson v Newspaper and Mail Deliverer’s Union of NY, 197 A 720, 722
(NJ Ch 1938) (citing Wyman, n 13). Another commentator notes that the doctrine emerged from the Statute of
Labourers in 1349 to prevent unjust wage demands due to labour shortages created by the Black Death and
eventually to those few tradesmen or professionals who worked outside the feudal domain. See Cherry, B,
‘Utilizing “Essentiality of Access” Analyses to Mitigate Risky, Costly and Untimely Government Interventions
in Converging Telecommunications Technologies and Markets’, (2003) 11 Common L Conspectus 251, at n 31.
16
See Speta, JB, ‘A Common Carrier Approach to Internet Connection’, (2002) 54 Fed Comm LJ 225, 251–256.
17
The concept of an ‘essential facility’ under US competition law whereby access is an economic necessity
was noted by the court to have its roots in common carrier doctrine. See Munn v Illinois, 94 US 113, 125–126
(1877) (citing extensively Hale, De Portibus Maris, 1 Harg, Tracts 78).
18
Since the nineteenth century, US courts have applied the ‘common carrier’ doctrine to undertakings re-
lated to infrastructure such as docks, roads, railroads, telegraph, and ultimately telephone, and constitutes a
narrower group. See eg Candeub, A, ‘Network Interconnection and Takings’, (2004) 54 Syracuse L Rev 369, 381.
289
19
In the US, telecommunication common carriers are not granted a licence certificate but rather, where re-
quired, approval to enter a market is made via an order by the FCC under the Communications Act 1934, s 214.
This individual approval process applies only to international service common carriers including facilities-
based carriers, resellers, prepaid calling card providers, and various wireless service providers offering calling
between the US and foreign points. Some of these are entitled to an expedited processing of 14 days. Here this
comprises licensing since the regulator makes the decision to allow market entry on an individual basis. For
US application and approval processes, see Crowe, TK, ‘FCC 214 Licensing’, <https://w ww.avvo.com/legal-
guides/u gc/fcc-s ection-214-l icensing-for-t elecommunications-providers-offering-i nternational-c alling>
(last updated 11 September 2012). Domestic, interstate services are subject to blanket approval pursuant to s
214 powers, with all individual scrutiny discontinued. This chapter will refer to this as a general authorization,
in keeping with the EU’s classifications. The FCC imposes conditions and requirements for the performance of
carriers’ service and network provision by further orders, regulations, and approval of filed tariffs.
20
47 USC 151 et seq. These businesses are also categorized as ‘public utilities’ both falling under a larger
category of businesses ‘affected with a public interest’, a doctrine expounded by Lord Chief Justice Hale and
utilized by US courts to justify economic regulation. See Munn v Illinois, 94 US 113, 125–126 (1877) n 17. US
courts in the early twentieth century struggled to pin the boundaries of this doctrine in various scenarios,
including minimum wage statutes, ticket brokers, and employment agencies. They failed to do so with any real
cohesive analysis and appear to have largely turned away from the doctrine as a source of power for economic
regulation. See Candeub, n 18.
21
Postmaster General, ‘The Posts Before 1711’ Monarchs of All They Surveyed: The Story of the Post Office
Surveyors (London: HMSO, 1952) , at 6–7. Those originally surveying distances between and establishing these
posts later oversaw enforcement of the GPO’s monopoly and proper collection of its postage rates. See gener-
ally, ibid. Innkeepers remained postmasters under new systems by bidding for the contract to provide such
services which then extended to private mail and for which these postmasters recouped their contract price
and earnings from postage charged.
22
Ibid.
23
However, the ‘pretence d’un publike bien’ served as the basis for the grant of many privileges by the King.
4 Holdsworth, A History of English Law 344 at n 6 (2nd edn, 1938) (quoting YB Ed III Pasch pl 8).
920
these restrictions arose as the result of a royal franchise or privilege.24 Other legal
justifications were few. ‘Custom’ or ‘prescription’25 could legitimize royal privil-
eges or franchises held since the time of memory that often had the result of re-
straining trade26 such as the grants of political and commercial self-governance27
to guilds and local authorities.28 A national or public interest justified more recent
grants. Beyond this, the principle of medieval common law, upheld by subsequent
courts,29 was that ‘prima facie trade must be free, and that freedom could only be
curtailed by definite restrictions known to and recognized by the common law’.30
The prerogative to restrict economic liberty via limitations on trade, grants of in-
tellectual property, etc, was subsequently vested in Parliament, although the royal
prerogative did not disappear quickly or readily.31 Hence, sources and kinds of
lawful restrictions on trade, while limited by these legal principles, were still nu-
merous and varied. Moreover, common law understanding of freedom to practise
a trade without restriction was limited to the concept of freedom from arbitrary
restriction not defensible by public policy.32
24
See 4 Holdsworth, n 23, at 344. This commentator notes, ibid at n 4, that grants of the King could be con-
trary to the common law and public policy as restraints on trade. Accord, Webb, S, and B, n 14 at 5, n 1 (quoting
the King’s 1604 circular letter to the Privy Council that ‘By the law and statutes of this our realm, the keeping
of alehouses and victualling houses is none of those trades which it is free and lawful for any subject to set up
and exercise, but inhibited to all save such as are thereto licensed’).
25
Usage beyond the time of memory, defined as that before Richard I. 3 Blackstone’s Commentaries with
Notes of Reference to the Constitution and Laws of the Federal Government of the United States and of the
Commonwealth of Virginia, 36 at n 7 (St George Tucker, 1803) (reprinted Rothman Reprints, 1969).
26
Freedom from restraint according to common law standards was freedom from arbitrary restraints only,
ie those not recognized by law. See Maitland, WH, The Domesday Book and Beyond: Three Essays on the Early
History of England (CJ Holt, 1897), at 261–264.
27
These numerous and varied grants often included not only the right to revenues collected in whatever
undertaking it applied to but also jurisdiction over the persons and activities involved, or ‘soke’ and ‘sake’, re-
spectively. See Maitland, WH, n 26, at 261–264. Hence trade and governance were often intertwined.
28
4 Holdsworth, n 23, at 346. 29
See eg Darcy v Allin, (1602) Moore KB 671–675.
30
4 Holdsworth, n 23, at 350.
31
See generally, ibid, at 344–362. The English Statute of Monopolies of 1623, 21 Jac 1, ch 3, made void all priv-
ileges, commissions, and grants of monopoly not confirmed by statute.
32
4 Holdsworth, n 23, at 350–352.
219
the local justices of the peace who under ‘the statute of 5 and 6 Edward VI c 25
(1552) . . . were authorized to select from time to time, at their discretion, certain
persons who were alone to exercise the trade of keeping a common alehouse’33
and inns when they became subject to the legal regimen of common callings.34
This regulation arose from a balancing of public interests: making available a then
perceived necessity of life, the beer consumed at every meal, and control over the
disorder and problems produced by excessive drinking.35 This strengthened their
earlier powers to eliminate any alehouses that went beyond that number required
to serve the needs of the market,36 perhaps an early example of natural monopoly
theory that served to justify the virtually exclusive licensing of monopoly pro-
viders of telecommunications in the twentieth century.37
Parliament’s delegation to justices of the peace included ‘three distinct forms
of control: the power of selection, the power of withdrawal, and the power of
imposing conditions’38 (controls which also describe accurately telecommunica-
tions licensing authority, until recently, in the EU). Purposes, broadly, for which
licensing might be imposed, were described over 100 years ago as follows:
The device of licensing—that is, the requirement that any person desiring to
pursue a particular occupation shall first obtain specific permission from a
governing authority—may be used to attain many different ends. The license may
be merely an occasion for extracting a fee or levying a tax. It may be an instrument
for registering all those who are following a particular occupation, in order, for
some reason or another, to ensure their being brought under public notice. It may
be a device for limiting the numbers of those so engaged, or for selecting them
according to their possession of certain qualifications. Finally, the act of licensing
may be the means of imposing special rules upon the occupation, or of more easily
enforcing the fulfilment either of these special rules or of the general law of the
land.39
tion carriers as businesses affected with a public interest under a theory akin to that of common carriers and
limiting their numbers within certain regions for reasons that seem premised on natural monopoly and public
interest. See Fulda, CH, ‘The Regulation of Surface Transportation in the European Economic Community’,
(1963) 12 Am J Comp L 303, 308–313 (commenting that while Germany, Belgium, Italy, and others limited
numbers of competitors in trucking and railroads operations to ensure continued availability and safe op-
eration without threat of unlimited competition that here would cause more harm than in other economic
sectors, the limiting of numbers of business units by the US seems generally to have been confined to the
liquor industry).
38
Webb, S, and B, n 14, at 4–5. 39
Ibid, at 4.
29
the regulatory or the ‘police’ power of the state ordinarily comprises the primary
foundation of licensing requirements considered necessary for the public interest
or general welfare such as public health, morals, or safety.40 In the US, a significant
number of states view licensing of legitimate businesses or occupations without
such interests as outside the limits of their police power.41
The source of power underlying licence fees is less clear. While the sovereign’s
power to tax has sometimes been considered to underlie the imposition of licence
fees, it has also been held that a true ‘licence fee’ is imposed under the police
power with application only to a type of business that is supervised or subject to
regulation that does in fact occur, the expense of which is intended to be defrayed
by the fees equated to the ‘probable cost’ of supervision.42 Fees unrelated to the
cost of regulation have been considered a ‘tax’ on occupations in the nature of
an excise under the power to raise revenue, and subject to review under different
standards from licensing fees.43 It has been suggested that perhaps both powers
can apply within the same fee and be valid. Licence fees for telecommunications
providers appear to vary greatly within and across jurisdictions. However, since
2002, fees under the EU’s framework, other than those associated with scarce
public resources, must not only be based on the costs of licence issuance and
enforcement, but must also be demonstrably so or otherwise adjusted. They are
then truly ‘licence fees’. The CJEU has, however, upheld, in contrast, a regional tax
based on ‘establishment’ as applied to communications providers in light of the
presence of their poles, pylons, and masts installed on private or public property
in a province as distinct from any fee to install or operate that equipment under
the Authorisation Directive.44
40
See Sharp v Wakefield [1891] AC 173, (Bramwell, LJ) (noting that the licensing of public houses was largely
police rather than economic regulation). Accord, 53 CJS, n 8 at §5.
41
See 53 CJS, Licenses § 5.
42
National Biscuit Co v City of Philadelphia, 98 A 2d 182, 187–188 (Pa 1953). Accord, Hunt v Cooper, 110 SW
2d 896, 899–9 00 (Tex 1937).
43
See eg National Biscuit, n 42 at 187–189; Hunt, 110 SW 2d at 899–9 01. Where the licence requirement is
seen as revenue-raising rather than regulatory, it has been held that the licence is merely a receipt rather than
a permission. See Royall v State of Virginia, 6 Sup Ct 510 (US 1886) (noting that a municipal occupation licence
could not prevent an attorney licensed by the State to practise law in any part of the State from practising
within the city limits, although a valid tax). Failure to obtain or comply with a licence that is seen as admin-
istrative or merely revenue producing may have lesser consequences at law than one which seeks to regulate
skills or proficiency or is otherwise imposed for public safety or health. See eg Dubray v Horshaw, 884 P 2d
23, 28 (Wyo 2000) (where statutory requirement for licence transfer was merely administrative rather than
intended to protect a class of persons from a particular harm, citing s 286 of the Restatement 2d of Torts, it
created no duty of care on the plaintiff).
44
Joined Cases C-256/13, C-264/13, Provincie Antwerpen v Belgacom NV, Mobistar (2014).
239
45
See Shively v Bowlby, 152 US 1 (1984) (noting that public lands were traditionally held by the king for the
benefit of the nation under jus publicum, with such vesting in the federal and state governments of the US upon
the American Revolution).
46
Corbett, K, Note ‘The Rise of Private Property Rights in Broadcast Spectrum’, (1996) 46 Duke LJ 611,
616–619.
47
See Institutes of Justinian 2.1 (T Cooper trans. 2d edn 1841) positing that ‘Things common to mankind by
the law of nature, are the air, running water, the sea, and consequently the shores of the sea . . .’
48
See eg Corbett, K, n 46.
49
Events in Telecommunications History, BT Group Archives, <http://w ww.btplc.com/Thegroup/
BTsHistory/1851to1880/1869.htm>.
50
Attorney-G eneral v The Edison Telephone Co of London, Ltd [1880–81] LR 6 QBD 244. In upholding the
Postmaster General’s monopoly, the court relied on the public interest of the Act’s grant of special powers to
build networks/i nstall equipment on public and private lands and related duties and its obligation not to di-
vulge the content of a communication which would not apply to an unlicensed entity acting outside the scope
of the Act. See ibid, 254–255.
924
by a private company without the granted licence, ensuing revenue sharing and
liability to takeover.51
Historically in the US, except for some small, rural carriers, American
Telephone & Telegraph (AT&T) was the exclusive ‘common carrier’ licensed under
the Communications Act of 1934 pursuant to a natural monopoly theory and after
AT&T had acquired most of the other carriers in the country.52
Licensing to limit the numbers of market entrants was neither exclusive to these
two countries nor a mere historical curiosity. In 1999, the EU reported that the
individual licences required by a majority of Member States with their ensuing
complexities and, in some states, delay and expense, lack of transparency, and
excessive regulatory discretion were barriers to market entry.53 EU licensing then
had the effect, if not also the object, of limiting the number of market entrants and,
therefore, of protecting the national incumbent and maintaining near-monopoly
market structures. The present EU framework has largely resolved concern about
the authorization grant itself as a barrier to entry with its mandate for general au-
thorizations without individual regulatory decisions or permissions, although as-
sociated market entry concerns remain.54
Licensing to control against market entry must be contrasted with licensing as a
tool to introduce competition. This is done by requiring licensing of the incumbent
that may or may not be still government-owned and/or granting licences to new
entrant(s) that will offer services in competition with the incumbent. To level the
playing field, asymmetric conditions are often imposed on the incumbent, usu-
ally via the licence, such as requiring it to interconnect on a non-d iscriminatory
basis and at regulated, sometimes cost- based rates and maintaining sep-
arate accounting to ascertain and verify such cost. Specific examples of a li-
cence to open markets include the UK’s licensing of Mercury under the British
Telecommunications Act 1981 to provide a second fixed network in competition
with British Telecommunications (BT), therein granted an ‘exclusive privilege’55
and the 1969 licensing of MCI by the US Federal Communications Commission to
51
See Events in Telecommunications History, n 49.
52
See Chapter 5 for a discussion of the early history of US telecommunications. For a further discussion of
natural monopoly, see Chapter 2.
53
Commission Communication, ‘Toward a new framework for electronic communications infrastructure
and associated services: The 1999 Communications Review’ (1999 Communications Review), COM(1999)
539, 10 November 1999. Accord, Commission Communication, ‘5th Report on the implementation of the
Telecommunications Regulatory Package’ (1999).
54
These include such issues as rights of way, fees, and cost transparency, see ECTA Regulatory Scorecard
Report 2009, that continue in the Next Generation Access context. See also Allen, J and Arnell, A, Report for
ECTA: ‘The digital single market and telecoms regulation going forward’ (18 September 2015). Both at: <https://
www.ectaportal.com/policy-publications/reports>.
55
See Chapter 3.
259
construct a microwave network and provide long-d istance services over it to sub-
scribers for interoffice communications.56 Both were merely first steps, requiring
further intervention including additional legislation and court action to achieve
what could be called a competitive market.
Licensing can serve simultaneously to limit and open markets as evidenced with
wireless communications. Limited spectrum availability and the need to protect
revenue flows of a state-owned or recently privatized monopoly incumbent (pos-
sibly operating in both fixed and wireless markets) have caused countries to de-
limit the number of wireless providers licensed in a market essential for its ability
to penetrate geographically more broadly with lower infrastructure costs than was
possible by installing new or upgrading aged fixed-l ines infrastructure. Wireless
telephony, therefore, in developing countries, leapfrogged the old technology to
compete as an alternative infrastructure to fixed lines and enhanced greatly the
historically poor telephony penetration rates. The leapfrogging has continued
with developed countries now seeing a marked decline in fixed-l ine subscriptions
and mobile telephony penetration rates approaching 100 per cent in developing
countries.57
In a hybrid of these dual licensing objectives, regulators have limited the number
of 3G market entrants but then used licensing conditions to open further these
limited markets. Hong Kong, eg, seeking to increase both competition and in-
novation, set ‘open network access’ conditions on the four 3G licensees, requiring
them to make a minimum of 30 per cent of their capacity available to virtual mo-
bile network operators (MVNO), effectively at least doubling the number of market
entrants with access negotiated at market rates.58 In 2011, France, in awarding only
four 4G licences, credited spectrum auction participants agreeing to enhanced
MVNO access conditions with a multiplier on their bid price, increasing their
chance to win a licence.59
Limits on market entry can also be imposed via requirements for multiple li-
cences, ie different licences for different types of networks and services. Providers
might be required to have numerous licences based on the network operated or
the type of service provided. For example, before the 1999 EU reforms, the UK had
twenty-two licence categories based on the network operated. Some countries
56
See Chapter 5.
See ITU, ‘Key ICT indicators for developed and developing countries and the world’, 2017, <https://w ww.
57
itu.int/en/I TU-D/Statistics/Pages/stat/default.aspx>.
58
Special Condition 12, Hong Kong Mobile Carrier Licence for 3G Networks, <http://w ww.ofta.gov.hk/en/
3g-l icensing/publications_c12_ mcl.html>. (archived)
59
See Maxwell, W, ‘French 4G Auction Results Announced’, International Spectrum Rev (Hogan Lovells
23 December 2011), < https://w ww.hoganlovells.com/blogs/h lspectrumreview/f rench-4g-auction- results-
announced>.
296
may license providers of mobile voice telephony but limit licensing of fixed voice
or international voice to the national incumbent for various reasons. For example,
starting in 2006 and until recently, the UAE allowed a second provider in fixed
line services to compete only in different geographic markets from the incumbent.
The delayed duopoly in all fixed markets via mutual local loop unbundling, sched-
uled for the end of 2011, only occurred in late 2015 well after the competitors were
already competing in mobile markets.60 This delayed liberalization is typical of
Middle Eastern and African countries.61 Licensing with fixed network and services
markets’ exclusivity for a defined period has been a common strategy to attract
investors in newly privatized incumbents in order to ensure a return on their in-
vestment. Technologies using Voice over Internet Protocols (VoIP), such as Voice
on the Net or peer-to-peer applications, however, have challenged such exclusivity
with numerous countries limiting competing VoIP services. Wireless VoIP tel-
ephony poses licensing and other regulatory challenges with the growth of WiFi
hot spots providing broadband internet access using unlicensed (possibly illegal)
spectrum, and the growing availability of WiFi handsets and soft phone software
to convert mobile internet access devices into phones.
60
See Kapur, V, ‘UAE resident alert: You may now switch your Internet, fixed line provider’ (Emirates 24/7
20 October 2015), <http://w ww.emirates247.com/news/emirates/u ae-resident-a lert-y ou-m ay-now-switch-
your-i nternet-fi xed-l ine-provider-2015-10-20-1.607422>.
61
Algeria only recently required LLU by its incumbent. See ‘Algerian government adopts new telecoms bill,
report says’ (Telegeography 3 January 2017), <https://w ww.telegeography.com/products/commsupdate/a rt-
icles/2017/01/03/a lgerian-government-adopts-new-telecoms-bill-report-says/>.
62
See Carrier or Service Provider Licence, Form I, sec G, Jamaica Telecommunications Act 2000,
Telecommunications (Forms) Regulations 2000, <http://w ww.our.org.jm/ourweb/sites/default/fi les/docu-
ments/sector_documents/application_ for_c arrier_ l icense_or_ service_provider_ l icense.pdf>.
63
eg Japan requires the filing of a business plan for telecommunications licences and applies disqualifying
fitness criteria. See Ministry of Internal Affairs and Communications, Manual for Market Entry into Japanese
Telecommunications Business (2006).
279
64
See ITU 3G License Table (2001) (11 of 49 countries used beauty contest), <https://w ww.itu.int/osg/spu/
ni/3G/.../l icensing_policy/3G_l icense_t able_F INAL-3.xls>.
65
Ibid. This is discussed further in Chapter 7.
66
Enhanced, or information services, in contrast, may be unlicensed and not regulated, eg in the US. See
Chapter 5.
67
See Chapter 4, European Union Communications Law at Section 4.4.2.
298
imposing a USO obligation (or right to exploit this opportunity, as the EU now
views such service provision) or a duty of USO financial contribution may require
a licensing condition or right. These can be individual USO conditions imposed on
specific providers, still often the former monopolist with its state-revenue-built net-
works and large market share. Contribution can be via a general condition imposed
on all or a defined group of providers, such as providers of public telecommunica-
tions networks and services. As in the UK, these can exist but be untriggered, eg for
potential future contributions to any USO fund that might be established if the cost
to the former incumbent becomes unduly burdensome. This avoids the difficulty of
amending the licence after issuance.
68
Authorisation Directive, Art 6(3).
69
See generally, Ofcom, ‘Notice under s 155 (1) of the Enterprise Act 2002 Consultation on undertakings
offered by British Telecommunications plc in lieu of a reference under Part 4 of the Enterprise Act’, 2005, at
<http://w ww.ofcom.org.uk/static/telecoms_ review/june05.htm>.
29
referral for full market investigation by the Competition and Markets Authority
(and potential for the full structural separation of BT), to enter into binding
undertakings to remedy, mitigate, or prevent any adverse effect on competition,
or any detrimental effect on customers which has or may be expected to result
from the adverse effect on competition. After a market review, in 2017, Ofcom
varied the 2005 undertakings to provide for the legal separation of the unit run-
ning the network, Openreach, as the prior undertakings were found insufficient
to address its bias in favour of BT’s retail business.70 Openreach will now operate
as a limited private company wholly owned by BT with its own employees and a
board that pursuant to governance commitments will oversee its own operating
strategy and accountability.71
The ability to correct market abuses via sanctions under licence conditions
remains a valuable tool, however. Arguably, this is especially true in emer-
gent and developing markets where the need to control the former incumbents
may be more acute than in more evolved competitive markets. The serious and
repeated regulatory intervention required in the UK, one of the world’s most
evolved markets, belies this, suggesting that control over the essential facility of
the core telecommunications network always permits exclusionary behaviour.
Recognizing the limitations of its sectoral conditions, the EU now requires na-
tional regulatory authorities (NRAs) to have the extraordinary sectoral power
of functional separation for such persistent problems, the impetus for the UK’s
sectoral provision.72
One scenario evidenced in contemporary telecommunications, if not historically,
is licensing as a substitute for either and/or both special rules and general laws of the
land, eg, where the need exists to get outsiders to risk money and time either to invest
in the incumbent or competitors, but where the general legal framework might not
be developed sufficiently to protect that investment. It might also be helpful absent
a general competition law framework that places duties on all players in the market
or where authorities are without telecommunications expertise. Licences with clear
obligations and rights can protect new entrants where no adequate sector-specific
regulations exist, such as for cost-based interconnection which competition law is
unlikely to require.
70
Although sector-specific regulation now provides for this option, see Communications Act 2003, s 89A,
with requisite notification to the Commission under s 89B, the legal basis for the revised undertakings does
not appear to have changed, although Ofcom did notify the Commission as s 89B requires.
71
BT, ‘Proposals agree with Ofcom’, <http://w ww.btplc.com/U KDigitalFuture/A greed/i ndex.htm>, ac-
cessed 10/0 9/2017.
72
See Chapter 4. Also see n 70.
03
73
But see ‘Testimony of Barry M. Aarons et al before the US Senate Commerce, Science and Transportation
Committee’ (7 March 2006) (stating that US municipality licence and franchise fees for telecommunications
providers bear no relation to access to municipal rights of way (and therefore falling within the limited re-
source category) but are rather merely revenue raisers and ultimately a service tax on end users), <http://w ww.
ipi.org/ipi_ i ssues/detail/testimony-before-t he-s enate-c ommerce-s cience-a nd-t ransportation-c ommittee-
regarding-v ideo-f ranchise-reform-a nd-voip>. See eg Chapter 3.24, Midvale Municipal Telecommunications
License Tax Code, Codification of General Ordinances of Midvale, Utah (1988 and as amended 2004)(3 ½ % tax
rate based on gross receipts).
74
See ECTA Regulatory Scorecard Report 2009, n 54, at 10. Also see Pyddoke, R, ‘Transparency and account-
ability of telecommunications in Sweden’ (2013) (Koncurrensverket Projekt 2012/310), at 16, 20–22 (positing
that even a regulator as effective as the Swedish PTS could be more accountable and transparent, eg failing
to publish a 2013 annual work plan with budget), <http://w ww.konkurrensverket.se/g lobalassets/forskning/
projekt/2012-310-t ransparency-a nd-accountability-of-telecommunications-regulation-i n-s weden.pdf>.
75
See ‘Zambia lowers international gateway license fee’ (Lusakatimes.com 16 June 2010) (noting that
Zambia was seeking to attract investment comparable to Uganda, Tanzania, and Kenya which had much
lower license fees), <https://w ww.lusakatimes.com/2010/0 6/16/z ambia-lowers-i nternational-gateway-
license-fee/>.
76
See OECD Investment Policy Reviews: Zambia (2012), at 145 (noting that $12.5 million gateway fee in
Zambia limited competition to incumbent until it was lowered to $300,000).
301
are made. In the EU, this historical licensing function is met by the Authorisation
Directive’s permitted notification procedure or registry. Implementation is
varied. For example, while the UK regulator has eschewed this general register,
other more limited registers exist. The Electronic Communications Code registry
allows persons to check if an undertaking has powers to install networks on
public and private land and has completed its annual financial security filing.
Premium rate providers (services and networks) must register with the Phone-
paid Services Authority under a code of practice authorized by Ofcom under the
Communications Act 2003. Ofcom also created a registry to allow it and other pos-
sible spectrum users to know how spectrum is being traded and reused, although
seemingly unavailable on Ofcom’s site.77
77
Wireless Telegraphy (Register) Regulations 2004, SI 2004/3155 as amended. See also Ofcom TNR, <http://
spectruminfo.ofcom.org.uk/spectrumInfo/t rades>.
78
See Wellenius, B, and Neto, I, ‘The Radio Spectrum: Opportunities and Challenges for the Developing
World’ (World Bank, 2008).
79
More than one can be involved, for example, in a joint-venture type of public/private partnership (PPP).
80
The financing may come from a variety of sources including the host government and/or may, according
to its nature as debt or equity, require sovereign obligations.
302
the contract fulfils the purposes of licensing and may effectively serve as the li-
cence. However, there may be a separate licence included or referenced within the
BOT contract.
81
See CJS, n 8, at 3.
82
See eg Lap v Axelrod 95 A 2d 457 (NY App Div 3d Dept 1983), appeal denied, 460 NE 2d 1360. Also see Sharp
v Wakefield [1891] AC 173 (‘The hardship of stopping the trade of a man who is getting an honest living in an
honest trade, and has done so, perhaps, for years, with probably an expense at the outset, may well be taken
into consideration; but it must be done so in conjunction with considerations the other way, and must be left
to the discretion of the justices.’ Bramwell LJ, 182–183).
83
See Sharp v Wakefield, n 82 (noting that, absent a provision in the enabling statute to the contrary, local
justices had the same discretion to renew as to issue).
84
Directive 2002/21/02 EC on a common regulatory framework for electronic communications networks
and services, L 108/33, 24 April 2002, Art 4.
85
See n 1.
30
great value of these rights of use and that the ability to trade or ‘resell’ spectrum
rights may be important to their effective and efficient use, particularly with pro-
jected 5G characteristics suggesting a greater need for spectrum sharing.86
The nature of such greater rights seems a ‘propertization’, as one usually cannot
trade or sell something without having some legally recognized property interest
in it. This almost seems implicit in any creation of such secondary markets that
need to make spectrum legally transferable. The US FCC implemented a legal
framework for trading seemingly structured to avoid this legal effect. It origin-
ally only permitted non-prior approval for spectrum leasing where the original
licensee retained the full licence and remained responsible for compliance.87
For transfers or assignments, the licence had to be returned to the FCC and new
licence(s) issued with regulatory obligations running to the new licensee (or to
both licensees if the spectrum was merely partitioned).88 Clouding the prior legal
distinction somewhat, the FCC has, under its power of regulatory forbearance,
streamlined the process once described as ‘clumsy’,89 to allow certain transactions
in both categories of leased and transferred/assigned spectrum to receive an ‘in-
stantaneous’ (overnight), expedited processing based on parties’ certification of
compliance with various set criteria and has also created the concept of a ‘private
commons’ where licensees can allow for device-to-device type communications
not involving the full network infrastructure.90
Grant holder concerns about the nature of their interest and extent of their
controls over other intangible interests granted via regulatory permissions, eg
for numbers or rights of access, where a premium has been paid, are understand-
able. Efforts to make the status legally clearer can be seen, eg in Australia with
‘Smartnumbers’ (ie those arranged so as to be more memorable) that are auc-
tioned by the Australian Communications and Media Authority. ACMA advises
applicants that the auction creates only an enhanced ‘right of use’ in the number
(ROU) but that they don’t become an ‘owner’ although the number ‘remains’ theirs
unless inactive for three years and that they can sell or lease it.91
86
See eg Johnson, N, ‘Reality Check: The need for new spectrum sharing and small cell strat-
egies’ (RCR Wireless News 4 October 2016), <https://w ww.rcrwireless.com/20161004/opinion/
reality-check-need-new-spectrum-sharing-small-cell-strategies-t ag10>.
87
See generally, FCC 03-113, First Report and Order and Further Notice of Proposed Rulemaking that au-
thorizes spectrum leasing in a broad array of Wireless Radio Services (FCC Washington, DC, 15 June 2003).
88
Ibid.
89
Judge, P, ‘Ofcom to throw radio spectrum wide open’ (Tech-World, 23 November 2004).
90
See ‘Promoting Efficient Use of Spectrum Through Elimination of Barriers to the Development of
Secondary Markets’, Second Report and Order, Order on Reconsideration, and Second Further Notice of
Proposed Rulemaking, FCC 04-167, 17529-33, paras 53–6 6 (FCC Washington, DC, 8 July 2004). Also see Sayle
Carnell, W, ‘ “Private Commons” in Radio Spectrum: The FCC Avoids a Tragic Result’, (2004) 6(1) Engage, The
Journal of the Federalist Society Practice Groups 150.
91
ACMA, ‘About Smartnumbers’, <https://w ww.thenumberingsystem.com.au/#/about-smartnumbers>.
034
92
Republic of Trinidad and Tobago Telecommunications Act, 2001, §21 (Act 4 of 2001).
93
Directive 97/13/EC of the European Parliament and of the Council of 10 April 1997 on a common frame-
work for general authorizations and individual licences in the field of telecommunications services, OJ L 117,
7 May 1997.
94
See further Chapter 16, at Section 16.4.
305
95
NZ Ministry of Business, Innovation & Employment, ‘Telecommunications and Broadcasting Network
Operator’ (ICT Policy & Programmes, Wellington 2017), <http://w ww.mbie.govt.nz/i nfo-services/sectors
industries/t echnology- c ommunications/c ommunications/t elecommunications-b roadcasting-network-
operators/how-to-register>.
96
NZ Telecommunications Act 2001, ss 102–105, SI 2001/103.
97
Term used in New Zealand and Australia to define that area of land between the front boundary of pri-
vate property and the road. For purposes of rights of access, this entails the road plus this area, a boundary-
to-boundary concept.
98
This grant would be called ‘Code Powers’ in the UK. See Section 6.4.4.
99
See Sixth Report on the Implementation of the Telecommunications Regulatory Package, COM(2000)
814, 7 December 2000, at Annex 1, Licensing, n 3.
100
See ICT Regulation Toolkit, Module 3, Authorization of Telecommunications/ICT Services, Section3.2
General Authorization and Open Entry Policies (infoDev ITU 2011), <http://w ww.ictregulationtoolkit.org/
toolkit/3.2>.
036
101
See Telecommunications Act, 2001 as amended in 2004, at s 21. The Telecommunications Authority, in
contrast, grants all licences for radio spectrum equipment or services pursuant to s 36 of the Act.
102
See Eligibility and Evaluation Criteria for Concessions, p 22 (TATT 15 October 2007), <http://w ww.tatt.
org.tt/Portals/0/Documents/E ligibility%20and%20Evaluation%20Criteria%20for%20Concessions.pdf>.
103
See ibid, at 23. 104
Ibid, at 11.
105
Stated to be a service concession in one place, see ibid, at 6, a network concession in another, see ibid, at
10, and a network/service concession in another, see ibid, at 18, adding to the confusion.
106
Ibid, at 18.
307
licence. This may comprise a comparative evaluative process such as the ‘beauty
contests’ first used in the US. Of concern here, however, is that these may not be
based on objective criteria and fair to all parties (discussed further in Chapter 7).
A competitive auction process may also be used. However, to ensure that en-
trants have the expertise to utilize the spectrum effectively, a pre-qualification
procedure to enter the bidding has recently become a common approach. The
same concerns arise as with comparative procedures. Competitive auctions can
be used to award concessions with respect to any limited opportunity or resource,
eg the operation of Singapore’s NGN,107 spectrum and unique number grants.
Here, the monetary bid and additional criteria meeting economic or social object-
ives serve as the determinants. These can encompass such things as minority or
local ownership minimums, tariffs to be charged, service obligations to be met,
or geographical roll-out of network and services. Multiple-round auctions having
a series of bidding rounds with all licences remaining open to further bidding for
the entire process are those currently favoured by the US and other countries. The
process can vary from country to country but:
[t]y pical features of the rules governing such an auction include requirements
that bidders make upfront payments; minimum opening bid requirements and
increments for bid increases; activity rules to limit the ability of bidders to sit out
rounds; rules regarding auction stages (points at which introduction of tighter
activity rules may eliminate some bidders) and stage advancement designed to
move the process along; stopping rules to determine when the auction closes; and
rules and penalties for removal or withdrawal of bids.108
Many believe auctions to be the best way to allocate the use of public resources.
They raise revenues for the state and are considered to ensure efficient and best
use of the resource since only those most likely to do this will reflect this value
in their bid price, assuming sufficient bidders for there to be true competition.
The auction rules can also achieve social objectives as discussed. Since there are
winners and losers based on the auction criteria that must be fairly clear, auctions
are considered the most transparent means of allocation. Auctions are not, how-
ever, without their perceived flaws. A key concern is how highest-price auctions
can accommodate other public interests. This is especially the case where these
are represented by non-governmental organizations that individually could not
107
ITU, ‘Singapore: Towards a Next Generation Connected Nation’, 10 December 2010, <http://w ww.itu.int/
net/w sis/stocktaking/docs/activities/1291981845/Towards%20a%20Next%20Generation%20Connected%20
Nation_ Singapore.pdf>.
108
Goodman, E, McCoy, S, and Kumar, D, ‘An overview of problems and prospects in US spectrum manage-
ment’, in Telecommunications Convergence: Implications for the Industry & for the Practicing Lawyer, 698 PLI/
Pat 341 (Practising Law Institute, New York, 1 May 2002).
038
compete with for-profit undertakings. For example, where pristine public land use
is contemplated, consortia of environmental and other groups might wish to be
allowed to bid against communications or associated facilities providers. Also, the
effectiveness and efficiency of the use of the scarce resource is not typically further
examined once the auction has concluded.
109
Issues surrounding the complexities of opening emerging markets to competition are discussed in
Chapter 17.
110
TS 58(1996) Cm 3276; (1994) 33 ILM 44. See further Chapter 16.
111
There are limited circumstances for reservations to the MFN condition. See Chapter 16.
112
‘With respect to any measure covered by this Agreement, each Member shall accord immediately and
unconditionally to services and service suppliers of any other Member treatment no less favourable than it
accords to like services and suppliers of any other country.’ Art II (1), GATS.
039
• the ‘Transparency’ general condition (Article III) requires all Members to pub-
lish their laws and regulations that affect trade in all services, scheduled or not.
Licensing qualifications and conditions meet that criteria;
• the ‘Domestic Regulation’ condition (Article VI) requires where service or sector
commitments have been made, that licensing, qualifications, and standards be
based on transparent and objective criteria (4(a)) and not more burdensome
than is necessary to ensure quality (4(b)). Licensing procedures must not re-
strict service supply, (4(c)); Members must inform applicants of decisions within
a reasonable time after the submission of completed application and, upon re-
quest, advise of the status of the application, without delay (3);
• the ‘National Treatment’ condition (Article XVII) requires that there be no dis-
crimination against a foreign service or suppliers of that service as compared to
those domestic services or suppliers where the Member has included a sched-
uled commitment.
113
This Annex is intended to encompass the provision of communications networks/services only to the
extent of Members’ scheduled commitments. See Annex on Telecommunications at s 2(c).
310
The EU’s legal framework is one clearly compliant with WTO standards for li-
censing and authorizations as well as allocation of scarce resources. The following
examines it at length.
In 1999, the EU proposed a new framework to ensure equal regulation for conver-
ging markets and technologies. With respect to licensing, the new framework was
to sweep under one scheme all public electronic communications and services,
not merely those involving telecommunications networks.114 The ensuing 2002
Authorisation Directive115 does this but is not, however, radically different from the
prior Licensing Directive116 as it was intended to work. Rather, the Authorisation
Directive further refines the Licensing Directive with provisions to ensure imple-
mentation of its intended light-touch regulatory scheme with individual grants
of rights and conditions permitted only where justified. Examining the former
scheme is helpful, therefore. A new framework with a consolidated, recast single
Directive is now being considered that further emphasizes general authorizations
over individual licence grants and obligations particularly for spectrum. It intends
a lighter, more harmonized touch. Its key changes to the Authorisation Directive,
are noted in the discussions below.
See Chapter 4.
114 115
Directive 2002/20/EC as amended by Directive 2009/140/EC.
Directive 97/13/EC, n 93.
116
31
Member States could impose any of a limited set of conditions on all licences
where justified and proportionate (but that comprised the ‘least onerous system
possible’ for general authorizations) in the areas of: essential requirements, in-
formation required to verify compliance, prevention of anti-competitive behav-
iour and discriminatory tariffs, and efficient use of numbering capacity (Annex,
2). General authorization conditions encompassed a range of consumer protec-
tions, including billing and contract format (Annex, 3.1), provision of: emergency
services, customer information needed for directory services and services for dis-
abled people, compliance with interconnection, and contribution to universal
services (Annex, 3.2–3.6). Beyond these, Member States could further impose on
individual licences only those conditions related to the circumstances justifying
the requirement for an individual licence in the first place (Article 8). The permitted
additional conditions, therefore, included those linked to allocation of numbering
rights, efficient use of radio spectrum, specific environmental or local planning
requirements, maximum duration to promote certainty and planning ability, pro-
vision of universal service, quality and permanence of service/networks, manda-
tory provision of publicly available networks and services, and interconnection
and leased lines obligations pursuant to other directives (Annex, 4).
Individual licences were to be granted pursuant to objective, transparent, and
time-limited procedures that applied to all unless objectively justified, and that
were published in an appropriate manner (Article 9). Licences for any service or
infrastructure category could be limited in number only to the extent necessary
to ensure efficient use of spectrum, or for the time needed to make available suf-
ficient numbers, and only via a published decision detailing the reasons for the
limitation (Article 10). Detailed, objective, transparent, non-d iscriminatory, pro-
portionate, and pre-published criteria for awarding the limited licences were re-
quired, according due weight to promoting competition and maximizing user
benefits (Article 10(3)). Procedures to permit interested parties to comment on the
limitation were required, as were invitations to parties to apply. Appeal to an inde-
pendent body from any licence denial was required as was Member State review of
the limitations on licensing, at reasonable intervals (Article 10(2)).
An undertaking complying with the conditions imposed on general authoriza-
tions could not be prevented from providing the relevant network or service, al-
though there could be a requirement to notify the NRA of intent to provide these
and to supply that information concerning the service needed to ensure compli-
ance with applicable conditions (Article 5). A waiting period of four weeks could
be imposed from receipt of this information prior to commencing service/network
provision (Article 5(2)).
General authorization fees could be based solely on administrative costs of con-
trolling, managing, and enforcing the general authorization scheme and were to
321
be sufficiently detailed and published in an accessible way (Article 6). Fees for in-
dividual licences were to encompass only those administrative costs incurred in
the issuance, management, control, and enforcement of the applicable individual
licences. With spectrum or numbers, charges could reflect a value to ensure their
optimal use (Article 11), permitting Member States to auction spectrum. The
Directive suggested further that all imposed charges be based on objective, non-
discriminatory, and transparent criteria (Recital 12).
Where a general authorization holder failed to comply with any of its conditions,
the NRA could notify it that it could no longer avail itself of the authorization ‘and/
or’ impose specific measures to ensure compliance (Article 5(3)). Individual licen-
sees failing to comply with their licence conditions could have the licences with-
drawn, amended, or suspended, or have imposed measures to ensure compliance.
The undertaking had the opportunity to state its views on the condition’s applica-
tion and to remedy its breach within one month. The Directive imposed time limits
for making and communicating that the decision had been confirmed, modified,
or annulled, and procedures for appeal to an independent body (Articles 5(3), 9(4)).
The Directive, to facilitate Community-w ide services, established a ‘one-stop
shop’ for obtaining licences from the Member States via a single application point
(Article 13) and authorized the Commission to charge various European tele-
communications regulatory groups, such as CEPT, with developing a harmon-
ized regime for general authorizations (Article 12). Neither was very successful.
The current EU regulatory framework including the Authorisation Directive pro-
posed by the Commission soon pre-empted this only two years after the Licensing
Directive’s adoption.
As the Authorisation Directive notes, there was a documented need for more
harmonized and ‘less onerous’ regulation of market access (Recital 1, 2002/20/
EC). The problem was not with the legal framework as promulgated but rather as
implemented and enforced in many Member States. Rather than limiting indi-
vidual licences to those circumscribed circumstances necessary to impose condi-
tions, the Licensing Directive’s flexibility was exploited. Individual licences were
mandated for many situations and general authorizations were the exception ra-
ther than the rule as intended. Marked divergence among the Member States as to
types of licences, time for decisions, costs, and information required, especially
in connection with individual licences, meant that the EU licensing and author-
ization regime was a barrier not only to national market entry but to the Single
Market.117 This was a significant problem. The EU premises its continued evolution
117
See Communication from the Commission to the European Parliament, the Council, the Economic
and Social Committee, and the Committee of the Regions, Towards a new framework for Electronic
Communications infrastructure and associated services, COM(1990) 539, at vii.
31
118
Ibid.
119
See further Chapter 4. The 2002 regulatory framework was amended largely in 2007 and December 2009.
The 2007 reforms generally comprised a regulation mandating a glide path of price caps for mobile roaming,
Regulation (EC) No 717/2007 (Roaming Regulation). The 2009 amendments were via two Directives, the so-called
‘Better Regulation’ Directive (Directive 2009/140/EC) and the ‘Citizens’ rights’ Directives (Directive 2009/136/
EC), amending the various 2002 Directives. The Regulation establishing a body of European regulators for elec-
tronic communications, Regulation (EC) No 1211/2009 (BEREC Regulation) also amended this. To understand
the reform’s scope it’s helpful to read consolidated versions of the five Directives that can be found at <https://
ec.europa.eu/digital-single-market/en/telecoms-rules > and <http://www.culture.gov.uk/images/consultations/
10-1134-implementing-revised-electronic-communications-framework-revisions-directives.pdf>.
120
Executive Summary, ‘Support for Impact Assessment, Review of the regulatory framework for elec-
tronic communications’ (SMART 2015/0 005), <https://ec.europa.eu/d igital-single-market/en/news/support-
studies-i mpact-a ssessment-telecoms-review>.
121
See Commission Communication ‘Progress Report on the Single European Electronic Communications
Market 2007 (13th Report)’, COM(2008) 153, 19 March 2008. The 15th Report notes a slight drop in investment
by 1.5% from 2007, possibly attributable to the economic slowdown, but still significant at €46 billion.
122
See Fourneron, K, and Ciriani, S, ‘Investments in Telecommunications Services higher in the US than in
the EU: a robust, enduring and increasing gap observed whatever the source’, at Sec 4 (Orange 2015) (noting
other recent comparable study findings), <https://w ww.orange.com/f r/content/download/32216/955794/
version/2/fi le/telecom_ i nvestment_comparison_US_v s_ E U.pdf>.
341
OJ L 204/37 (21 June 1998) as amended by Directive 98/4 8/EC, OJ L 217/18 (5 August 1998).
123
eg email conveyance posited by the Framework Directive as a communications service rather than an
124
125
See Recital 2, Directive 98/4 8/EC, n 123. Accord, Europarl Ref E-0 969/0 9, Answer by M Barrot on behalf
of Commission (16 April 2009) (noting further the linkage between Article 50 and the Article 95 basis of the
Framework Directive), <http://w ww.europarl.europa.eu/sides/getAllAnswers.do?reference=E-2009-0 969&
language=EN>.
126
See eg Europarl Ref E-4364/0 9, Answer by Mrs Reding on behalf of the Commission (16 September 2009),
<http://w ww.europarl.europa.eu/sides/getAllAnswers.do?reference=E-2009-4364&language=EN>.
127
Added by Directive 2009/140/EC; UK implementation at Communications Act 2003, s 32(1) as amended
by para 9, Sch 1, The Electronic Communications And Wireless Telegraphy Regulations 2011. That this intends
a wide range of physical structures can be seen from the legislative history as this wording appears to derive
from the EU Parliament’s proposed additions to Article 12, Access Directive on first reading which provided in
part: ‘including entries to buildings, building wiring, masts, antennae, towers and other supporting construc-
tions, ducts, conduits, manholes, cabinets and all other network elements which are not active’. European
Parliament legislative resolution of 24 Sept 2008 on the proposal for a directive of the European Parliament
and of the Council amending Directive 2002/21/EC on a common regulatory framework for electronic com-
munications networks and services, Directive 2002/19/EC on access to, and interconnection of, electronic
communications networks and associated facilities, and Directive 2002/20/EC on the authorization of elec-
tronic communications networks and services (COM(2007)0697, C6-0 427/2007, 2007/0247(COD)).
128
COM(2016) 590 final, Proposal for a Directive of the European Parliament and of the Council establishing
the European Electronic Communications Code (Recast) (‘EECC’).
361
129
Defined as services usually provided for remuneration that enable ‘direct interpersonal and inter-
active exchange of information via electronic communications networks between a finite number of persons,
whereby the persons initiating or participating in the communication determine its recipient(s)’. See pro-
posed EECC, Art 2.
130
Ibid.
131
This freedom is also subject to restrictions pursuant to Member States’ power to legislate in defined areas
under EU Treaty, Art 46 (1), including public health, policy, and security. See Authorisation Directive, 2002/
20/EC, at Art 3(1); Recital 3.
132
See Recital 42, proposed EECC.
371
agency, a proposal encountering great resistance)133 that would in turn notify the
relevant Member State authority.
These Authorisation Directive mandates limit Member States’ discretion and
arguably remove the possibility of delaying the provider’s entry into the market.
This ‘least onerous’ system was viewed necessary to stimulate development of
new electronic communications services, including pan-European services, a sig-
nificant focus of the 2002 regulatory framework, and to permit persons to benefit
from EU Single Market economies of scale (Recital 7), not possible under the
Licensing Directive’s implementation.134 Cross-border focus also partly underlies
the Authorisation Directive’s requirement that rights of undertakings providing
networks and services be included within the authorization itself (see Recital 9).
Such harmonization creates greater certainty about the ability to enter a new na-
tional market and, as Recitals note, ensure a level playing field throughout the
Community. Specifically, the Directive requires Member States to ensure at least
the right to:
The Recitals to Directive 2009/140/EC indicate that the NRA should be able to
coordinate acquisition of rights of way and make information available on their
websites. Beyond the requirement for simple, efficient processes by the relevant
authority, this does not require NRA power to grant the rights of way or over local
or other authorities that may control them.
The general authorization must further grant those providing communications
networks and services to the public the right to:
133
See Teffer, P, ‘EU telecom watchdog plan dead on arrival’ (EU Observer, 27 April 2017), <https://
euobserver.com/d igital/137706>.
134
As to its continued relevance, see Recitals 26, 33, Art 9, Directive 2009/140/EC.
318
• negotiate and obtain interconnection with and access to other publicly available
networks covered by a general authorization anywhere in the EU under condi-
tions set by the Access Directive (Article 4 (2)(a));135
• be given the opportunity to provide elements of universal service (Article 4 (2)
(b)).136
With the same right to negotiate and obtain interconnection and access essen-
tially guaranteed in each Member State to those providing public networks and/
or services, cross-border entry and the possibility for pan-European services is fa-
cilitated. There is no longer the need for an individual designation as an operator
entitled to obtain such rights, as was the case under the early regime. Here again,
Member State discretion has been limited.
Those providing networks and services to other than the public are to negotiate
interconnection on commercial terms (Recital 10). Member States are to provide
declarations, either automatically upon entry notification or request, that con-
firm rights under the general authorization to interconnection and rights of way
in order to facilitate interconnection or negotiations with other authorities (Article
9). Such declarations, however, do not create or condition the exercise of rights.
As with the current framework, only the specified conditions (whether to the
general authorization or individually) can be imposed. These cannot duplicate
other obligations or national law. The proposed EECC reform leaves the above
rights unchanged but adds two new rights: to use spectrum as specified and to
have applications for numbers considered pursuant to provisions that are not gen-
erally dissimilar from the current regime,137 discussed below. BEREC, as the body
to be notified of market entry if required, would make the necessary facilitation
declarations.138
6.4.2.2 Individual rights
The current framework mandates that the only exceptions to the ‘general au-
thorization’ requirement concern individual grants of rights to use spectrum
and numbers. This discretion, however, is not unlimited.139 Grants of individual
rights for the use of spectrum and numbers must follow certain substantive and
procedural safeguards, addressed below in the context of numbers.140 Member
135
See further Chapter 8.
136
See further Chapter 9.
137
See proposed Art 2, EECC. That is once you get beyond the proposed enhanced roles of the Commission
and BEREC in numbering and spectrum conditions that are likely controversial.
138
See proposed Art 12 (3), EECC.
139
The requirements for making spectrum available are discussed in Chapter 7.
140
See also Recital 11. An amendment to Art 5, however, makes clear that these apply to both rights to use
numbers and spectrum. See Directive 2009/140/EC, Art 3.
391
States must make available such individual rights of use to any undertaking for
the provision of networks and services, irrespective of whether granted to the
network/service provider or their users (Article 5(2)).141 Addressing an inconsist-
ency142 the proposed EECC makes clear that numbers may be awarded to other
than providers as long as they are capable of managing them and there are suffi-
cient numbering resources available.143
142
Authorisation Directive, Recital 12 does not mention numbering in stating that it is irrelevant whether
the grantee is the provider or the user. Art 5(2) indicates that where an individual grant of rights to use num-
bers is made, that Member States ‘shall grant such rights, upon request, to any undertaking for the provision
of network or services under the general authorisation’, indicating that it is not a matter of discretion. In con-
trast, however, Recital 14 states that Member States are ‘neither obligated to grant nor prevented from granting
rights to use numbers from the national numbering plan . . . to undertakings other than providers of electronic
communications networks or services’, suggesting that it is totally a matter of discretion.
143
Although PECN/PECS are not subject to capability criteria, this might be justified for other parties not
subject to conditions.
144
Italics reflect amendments by Directive 2009/140/EC.
320
account of the need to allow for an appropriate period for investment amortization’
(Article 5(2)).145 The ‘objective pursued’ appears to refer to the regulatory objective
underlying the need for an individual grant and its time-limitation, although
clearly there could be more than one. Investment recovery is also a concern to en-
sure a balance of incentive with regulatory objectives. In the context of numbers,
it is possible that £ millions could be invested not only in procuring a particular
number but also in marketing and development costs. With an auction procedure,
however, the market’s valuation and shortened amortization of up-front time
limitations should be reflected in a lower price. If the period were too short, there
would be no takers or only those gambling on a future renewal right.
The Authorisation Directive itself does not specify an appeal right from the deci-
sion regarding the granting of individual rights of use.146 The Framework Directive,
however, requires Member States to ensure effective mechanisms for any provider
or user affected by any NRA decision to appeal its merits to an independent body
with appropriate expertise147 that must issue its decision in writing, if not a judi-
cial body (Article 4, 2002/21/EC). In this case, further review to a court or judicial
tribunal must exist.148 These requirements are unchanged in the proposed EECC.
6.4.2.4 Conditions
EU network and service providers may have only three types of conditions im-
posed on them: (i) those under the general authorization, (ii) individual obliga-
tions that attach to the granting of rights of use of numbers and spectrum, and (iii)
specific conditions to impose obligations under the Access and Universal Service
Directives (Article 6). Any condition imposed under any of these categories, how-
ever, must be proportionate, transparent, and non-d iscriminatory149 with further
procedures and limitations for the last two categories as per the Framework and
applicable specific Directive. The following examines each category in turn.
145
Italics reflect amendments by Directive 2009/140/EC.
146
This is in contrast to Art 10, which requires an appeal procedure from the imposition of conditions. See
Section 6.4.2.7.
147
A 2009 wording refinement makes clear that the appeal body must itself have appropriate expertise for
effective review rather than merely having such expertise available to it. See Directive 2009/140/EC, Art 1.
148
Ibid, at Art 4(2).
149
The 2009 reforms removed a requirement that NRAs objectively justify a condition under the general au-
thorization in light of the network/service concerned. Although theoretically lessening the NRA’s burden to
trigger or include a specific condition within Part A of the Annex’s permitted categories that may apply to all
providers or merely certain ones, after ten years in effect, such rationalizing on a per-service basis had likely
already taken place.
321
150
Authorisation Directive, Art 6. See also Part A, Annex’s maximum list of subject areas that a general
condition may govern.
151
Until 2016, the Universal Service Directive allowed Member States to determine whether to require ‘net
neutrality’, ie the ability of a provider to restrict access to content or provide unequal treatment to different
traffic for other than technical reasons, limiting itself to the above transparency. See Directive 2002/22/EC,
Art 1(3) as amended by Directive 2009/136/EC. Now, non-d iscriminatory access to content of choice is re-
quired by Regulation 2015/2120/E U laying down measures concerning open internet access and amending
Directive 2002/22/EC on universal service and users’ rights relating to electronic communications networks
and services and Regulation 531/2012/E U on roaming on public mobile communications networks within the
Union. The proposed EECC removes the related condition area.
Bold text indicates proposed EECC’s redactions.
152
153
Repealed and replaced by Directive 2014/53/E U on the harmonisation of the laws of the Member States
relating to the making available on the market of radio equipment, OJ L 153/62, 22 May 2014.
154
This condition would be unnecessary under the proposed EECC that would require payment for any un-
fairly burdensome USO from public revenues.
155
A further s (11a) clarifies that conditions regarding use for public communications can encompass elec-
tronic communications beyond broadcast and encompasses the warning of imminent threats and the mitiga-
tion of a major disaster. See Annex A.
23
156
See further Chapter 13.
157
Italics denote 2009 amendments; bold text, proposed EECC redactions. The referenced access as re-
quired by Art 6, is equivalent access and affordability of public telephony service (voice supporting local, na-
tional and international calling and data at functional internet levels), Universal Service Directive.
158
Bold indicates text redacted in proposed EECC. This condition duplicates national law that applies
otherwise.
159
Art 6(3).
23
The proposed EECC would further reduce flexibility. While it does not greatly
change the substance of conditions under the general authorization,160 it divides
them into three groupings with different potential applicability:
160
The bold text above indicates the proposed EECC’s redactions.
Administrative charges, information, general access, use in major disasters, privacy/data protection,
161
conformity to standards and specifications, transparency obligations. See proposed EECC, Part A, Annex I,
General conditions which may be attached to a general authorization.
162
Service interoperability/network interconnection, spectrum use, ‘must carry’, limiting electromag-
netic field exposure, network integrity and electromagnetic interference prevention, transparency. Proposed
EECC, Part B, Annex I, Specific conditions which may be attached to a general authorisation for the provision
of electronic communications networks.
163
Service interoperability/network interconnection, end-u ser access to numbers, sector-specific con-
sumer protection rules and illegal content restrictions. Proposed EECC, Part C, Annex I, Specific conditions
which may be attached to a general authorisation for the provision of electronic communications services.
164
The full import of this amendment is not clear. While it suggests merely a transparency obligation for
charging consumers, eg premium rate services, the only other use of the phrase is in connection with the jus-
tifications for the Commission’s authority to issue a decision or recommendation in the area. See Framework
Directive, Art 19.
165
Ofcom after consulting on charging for the use of geographic numbers due to their growing scarcity set a
pilot programme to charge CPs for geographic numbers. On review in 2016, Ofcom determined that the char-
ging resulted in a significant one-off return in number blocks, delaying scarcity and is proposing to charge
providers for numbers in areas where scarcity threatens.
166
See Sections 6.4.3 and 6.4.4.3 regarding number portability requirements.
167
Annex, Part C, s 6(1).
342
These individual obligations must be objectively justified with respect to the service,
proportionate, transparent, and non-discriminatory. They may not duplicate the
general conditions. The proposed EECC requires NRAs to ensure compliance with
other Member States’ consumer laws and rules on number use for numbers used
extraterritorially. A new number condition would permit obligations to ensure this,
effectively importing those requirements into the individual grant.168
168
See proposed Art 88(6), Annex II E (10), EECC.
169
Art 6(2). 170
Directive 2002/19/EC, at Art 8. See further Chapter 8.
171
Ibid, at Art 9 (revised to include technologically neutral wholesale reference offers under Art 12.
Additional transparency obligations regarding the impact on quality of services from traffic management
measures are imposed by Regulation (EU) 2015/2120 discussed at n 151).
172
Ibid, at Art 10. 173
Ibid, at Art 11. 174
Ibid, at Art 12. 175
Ibid, Art 13.
176
Ibid, at Art 15.
177
See ibid, at Art 8(3) (via the cross-reference to Art 14(2), this decision also requires adherence to
comitology procedures set forth in Arts 5 and 6 of Decision 1999/4 68/EC).
178
Framework Directive, Art 8 states the objectives of promoting competition and the interests of EU citi-
zens and contributing to the internal market’s development.
325
that are not effectively competitive, NRAs must impose appropriate regulatory obli-
gations on entities with SMP in related markets179 to prevent leveraging in the retail
markets.180 These may include bans on: excessive and predatory pricing, undue pref-
erential treatment, or unreasonable bundling of products/services.181 These must
meet the Framework Directive’s requirements for transparency, objectivity, pro-
portionality, and consultation as with other conditions. NRAs can also apply appro-
priate retail price caps to control individual tariffs or other measures to steer pricing
towards cost-based or that of other comparable markets.182
In the 2009 reforms, the Commission sought a killer solution for residual bottle-
necks in access markets despite on- going regulatory intervention. The Access
Directive requires NRAs to be empowered to impose the further extraordinary SMP
remedy of functional separation of wholesale access provision on vertically inte-
grated entities with SMP in relevant access markets, where persistent and important
market failures exist despite appropriate conditions.183 This approach was essen-
tially a page from Ofcom’s playbook in forcing BT to agree to restructure its oper-
ations with the core network in a separate operating unit under UK competition law
powers, as previously noted. Under this EU sectoral remedy, the separate unit would
have to supply such services to all undertakings including the parent, on equivalent
terms and over the same systems.184 The NRA must justify the need for and suitability
of this extraordinary remedy to the Commission with particulars of the transaction
proposed and its regulatory oversight;185 as with all SMP conditions under the Access
Directive, following the notification procedures under Article 6 of the Framework
Directive.186 The new business unit could also be subject to other Access Directive
conditions as above.187
The proposed EECC seeks in the access regime to address concerns that ac-
cess policies have promoted service-based over infrastructure-based com-
petition with an ensuing lag in the build-out of very high capacity networks
throughout the EU. Specific SMP access conditions at the wholesale level could
be imposed only where and when necessary to address retail market failures, in
179
Universal Service Directive, Art 17(1); also see Framework Directive, Art 14(3) as amended by Directive
2009/140/EC.
180
Framework Directive, Art 14(3).
181
See 2002/22/EC, Art 16 as amended by Directive 2009/136/EC (deleting specific retail price controls,
minimum leased lines, and carrier selection obligations).
182
Ibid, at (2).
183
See Access Directive, Art 13a. The Directive also provides for voluntary separation at Art 13b.
184
Ibid, at Art 13a(1). 185
Ibid, at Art 13a(2), (3).
186
See also Commission Recommendation 2008/850/EC on notifications, time limits, and consultations
provided for in Article 6 of Directive 2002/21/EC of the European Parliament and of the Council on a common
regulatory framework for electronic communications networks and services (15 October 2008).
187
Art 13a(5). See further Chapter 8.
326
light of end-u ser outcomes and via what might be considered a least-restrictive
alternative approach that considers eg whether access to civil engineering in-
frastructure alone might be more conducive to sustained competition than
other ex ante controls. This would be only after a modified market review pro-
cess evaluating, on a forward-looking basis, emerging commercial trends (eg
co-i nvestment or access agreements) and the potential impact of regulation on
these and applying previously only recommended criteria for evaluating mar-
kets amenable to continuing ex ante regulation (that would be codified under
the EECC).188 A lighter touch access regime based only on fair, reasonable, and
non-d iscriminatory (FRAND) terms and dispute resolution obligations would
apply to wholesale- only SMP network providers.189 Retail price regulation
would be eliminated.
The second category of specific conditions under the current Access Directive
concerns those NRAs may impose on non-SMP providers to ensure end-to-end con-
nectivity or to make services interoperable where they control access to end-users
(eg via unique numbering or addresses).190 These may include mandated intercon-
nection if commercial negotiations pursuant to general authorization conditions
fail191 as well as conditions imposing access to electric programme guides (EPGs)
and application programme interfaces (APIs) by providers of these associated facil-
ities on FRAND terms where needed to ensure end-user accessibility to digital TV
and radio services.192 The Directive also requires conditions on providers of condi-
tional access services necessary for end-user access to digital TV and effective com-
petition in such services.193
The proposed EECC would continue such non-SMP access conditions, adding the
additional requirement of being subject to a general authorization. In justified cases
where access to emergency services or end-to-end connectivity between end-users
is at risk from lack of interoperability, conditions necessary to address this, including
adherence to standards, could be imposed on number-independent ICS providers
but only after the Commission determines that national regulatory intervention is
needed following a report from BEREC under the rules for delegated acts.194 The pro-
posed framework would also allow, where no other viable alternative is offered on
fair terms, obligations for reasonable access to non-SMP owned network elements
not readily replicable (economically or physically) such as wiring/cables within a
188
Art 65 (1), proposed EECC. 189
Art 77, proposed EECC.
190
Non-SMP providers can also be required to share specific facilities where it will increase structural-
based competition and lower rollout costs for new networks. See Framework Directive, Art 12(1).
191
2002/19/EC, Art 5(1)(a). 192
Ibid, at Art 5(1)(b). 193
Ibid, at Art 6; see Chapter 8.
194
See Arts 59 (1), 110, proposed EECC.
327
building, to the first concentration/distribution point outside the building and fur-
ther where ‘strictly necessary’ or ‘insurmountable’ barriers exist.195 These may in-
clude rules governing access, transparency, non-discrimination, and cost allocation
in light of risk.
Universal Service Conditions The 2009 reforms via the ‘Citizens’ Rights
Directive’ made some changes to the Universal Service and Users’ Rights Directive
but did not really change the defined EU-w ide universal service level, itself. This
remains as access to a communications network at a fixed location and service
that supports voice, data, and ‘functional’ internet access defined as dial-up
modem, or ‘narrowband’ connection.196 The proposed EECC would upgrade the
level to functional internet access reflecting that used by most end-users but cap-
able of supporting a minimum list of services197 enabling civil society participation
and voice communications services at a nationally specified quality, at least at a
fixed location.198 Universal service would, however, no longer encompass access to
directory enquiry services or directories or provision of public pay phones unless
a national need for these is demonstrated.199 The requirement that Member States
ensure equivalence of access and choice for disabled end-users, a significant 2009
reform,200 continues in the proposed EECC, although seemingly only via a specific
designation as the proposed EECC deletes the relevant wording of the condition
under the General Authorisation.201
Currently the specific US conditions that can be imposed on designated US pro-
viders, including non-SMP, are:
195
See Art 59 (2), proposed EECC.
196
Member States can change this to reflect a level of function in keeping with the majority trend in a na-
tional market but pay for it with public funds. Recital 5, Citizens’ Rights Directive. See also, Case C-1/14, Base
Co. NV v Ministerraad (2015), paras 38–42. The Commission’s proposed reform would mandate both the ma-
jority measure of functionality and the public funding obligation. See Art 79, proposed EECC.
197
Annex V, proposed EECC details these as voice and video calls, email, search engines, online educa-
tion/t raining, news services, goods and services purchase, professional networking, online banking, use of
eGovernment services, social media and instant message, refinable at the national level.
198
Also at an affordable price in light of national conditions. See Art 79, proposed EECC. Member States can,
if needed, include mobile.
199
Art 82, proposed EECC.
200
Art 23a, Universal Service Directive (as amended by Directive 2009/136/EC).
201
Annex V (B)(3), proposed EECC.
202
The Universal Service Directive amended the definition of ‘PATS’ to remove the provision of ‘emergency
services’ from its defining criteria and a list of other possibly relevant specific services such as directory en-
quiry eliminating the possibility that service providers otherwise meeting the definition are not excluded from
238
2. provision of public pay phones and other voice telephony access points (Article 6,
2002/22/EC as amended);
3. provision of a printed or electronic directory, as required, comprising all PATS
subscribers and directory enquiry services accessible to all end-users (Article
5, 2002/22/EC);
4. measures for disabled persons to ensure equivalent access to PATS, emergency
services, directory enquiry (Article 7, 2002/22EC);203 and
5. affordable tariffing for such services where necessary to provide specified ac-
cess to persons with low income or having special needs.
obligations because they don’t provide emergency services. PATS is now defined under the Directive (and in the
UK General Conditions of entitlement) as ‘a service made available to the public for originating and receiving,
directly or indirectly, national or national and international calls through a number or numbers in a national or
international telephone numbering plan’ (Art 2(c); Definitions, Revised UK General Conditions of Entitlement).
203
See Art 7(1), suggesting that such USO obligations might be obviated where equivalence of access to
services and choice of providers enjoyed by the majority of end-u sers is provided for in consumer contracts.
204
The Access Directive imposes the requirement that such specific conditions comply with Arts 6 and 6a
of the Framework Directive (Arts 5(3), 6(3)), governing transparency, consistency and consultation, as above
described in connection with the granting of individual rights of use, see Section 6.4.2.3, and the reporting
requirements for certain NRA actions. The Universal Service Directive, in contrast, refers only to its own re-
quirements for consultation (Art 33) and Commission notification (Art 36), although it is likely, that Art 6 of
the Framework Directive governs too, with Art 33, USD, a refinement to include manufacturers and end-u ser
groups within interested parties.
205
Universal Service Directive as amended by Directive 2009/136/EC, at Art 8(3).
206
Art 81(5), proposed EECC.
239
1. payment obligations for USO contributions, administrative fees under the gen-
eral authorization, and usage fees;
2. those specific conditions permitted to be imposed under Article 6(2) of the
Authorisation Directive (see Section 6.4.2.4) (Article 11(a)).
These information requirements would remain, modified only to conform to the
revised conditions as above208 as would the ability of NRAs to require propor-
tionate, objectively justified information necessary to verify compliance with ap-
plicable conditions on a case-by-case basis where a complaint has been received,
or investigation or other reasons suggest problems with compliance as currently
under Article 11(b)).209
Such information can also be required for:
207
Arts 19 (4) and 23, proposed EECC. Art 21, proposed EECC.
208
Ibid.
209
30
The proposed EECC retains these, adding other competent national authorities as
possible overseers,210 and adds an ability for NRAs to require information on elec-
tronic communications networks and associated facilities disaggregated at a local
level so as to be able to conduct a geographical survey of the reach of broadband
networks for planning purposes and designating digital exclusion zones wherein the
NRA can make calls for interest in deploying networks.211 The EEEC would authorize
NRAs to sanction the deliberate provision of misleading, erroneous, or inaccurate
information, including the failure to respond to a call, the latter of which seems con-
troversial as possible future plans for network deployment would seem to be confi-
dential business data.212
• NRA obligation to monitor and supervise compliance with conditions of the gen-
eral authorization or rights of use and those non-SMP specific access or universal
service conditions as discussed above (Article 10 (1));
• Mandated power rather than a discretionary potential to require provision of in-
formation necessary to verify compliance with such obligations (Article 10(1));
• Mandated NRA power to impose dissuasive financial penalties (Article 10(3)(a));
• NRA power to require the cessation of a breach, including immediately (Article
10(3)); and
• Ability to order the delay or cessation of a service or service bundle likely to cause
significant harm to competition pending SMP compliance with a specified access
obligation (Article 10(3)(b)).213
210
Under the proposed EECC, Member States must ensure that a minimum list of tasks defined at Art 5(1)
are assigned to NRAs only, eg implementing ex ante regulation such as access and interconnection obliga-
tions, granting general authorizations, ensuring dispute resolution, etc. See Art 5. Beyond that, Member States
have flexibility to designate roles either to NRAs or other competent authorities, but must ensure the inde-
pendence of these other authorities.
211
Arts 20 (1), 22 (3), proposed EECC. 212
Art 22 (4).
213
Directive 2002/20/EC, as amended by Directive 2009/140/EC. 214
Art 30, proposed EECC.
31
215
Art 10(5), Directive 2002/20/EC as amended by Art 3(6)(c), Directive 2009/140/EC (substituting ‘or’ for
‘and’). Italics indicate the 2009 amendments.
216
This would remain unchanged. Art 30(2), proposed EECC.
217
Italics indicate 2009 amendments. This would remain unchanged in Art 30(2), proposed EECC.
218
The body must itself have the expertise rather than merely have it available to it. Directive 2002/21/EC,
Art 4(1) as amended by Directive 2009/140/EC. The revised Directive clarifies that interim measures may sub-
stitute for the NRA’s decision where granted in accordance with national law. Ibid.
219
Art 31, proposed EECC.
32
6.4.2.8 Fees
The 2002 Authorisation Directive, like its predecessor, mandates that administra-
tive charges imposed under the general authorization be only those incurred in its
‘management, control and enforcement’ (Article 12 (1)). This includes charges for
activities connected with rights of use and specific conditions imposed under Article
6(2) (see Section 6.4.2.4). It also details as permissible chargeable activities those
incurred for international cooperation (eg radio frequencies, numbering schemes),
harmonization and standardization, market analysis, monitoring compliance and
other market control, as well as regulatory work involving preparation and enforce-
ment of secondary legislation and administrative decisions, such as decisions on ac-
cess and interconnection (Article 12(1)). It requires administrative fees or charges to
be imposed in an objective, transparent, and non-discriminatory manner but, also,
one that minimizes additional costs (Article 12(b)).
The Authorisation Directive further provides not only that the charges be pub-
lished annually but that regulators provide an annual overview of their admin-
istrative costs for the permitted activities. This effectively requires accounting
separation. It also requires an appropriate adjustment to be made when there is a
difference between costs and charges (Article 12(2)). While accounting separation
and cost justification are tools previously used in EU telecommunications regula-
tion, they were controls imposed on former monopolist incumbents that enjoyed
special or exclusive rights and privileges and, subsequently, on SMP operators.
The proposed EECC would maintain these requirements, extending them to any
other competent authorities imposing administrative charges.221
Finally, the Authorisation Directive anticipates that non-cost related fees may
be imposed for ensuring optimal use of numbers, spectrum, and rights to install
facilities on public or private land (Article 13). In doing so these must be object-
ively justified, transparent, and non-discriminatory, as well as proportionate to
their intended use. This, with other Articles in the Directive that permit a compara-
tive/competitive procedure for granting individual rights of use, contemplates the
possibility of usage fees determined by auction. The proposed EECC does not sub-
stantively change this but deals with numbers separately from the other individual
rights,222 authorizing other ‘competent’ authorities to impose the charges in this
latter group.223
220
See eg Commission ‘15th Progress Report on the Single European Electronic Communications Market’,
COM(2010)253 final, 25 August 2010.
221
Art 16, proposed EECC. 222
Art 89, proposed EECC. 223
Art 42, proposed EECC.
3
224
Ofcom, ‘BT agrees to legal separation of Openreach’, 10 March 2017. It is to be noted that in 2015 O2 Czech
Republic chose to avail itself of a 2009 reform (Art 13b, Framework Directive) and spun off its infrastructure
into a separate company as a measure to enhance shareholder value, a measure it hailed as the ‘world’s first
voluntary’ structural separation. ‘O2 Czech Republic Investor Presentation’, September 2015.
225
Art 65(5)(a), proposed EECC.
226
See generally ‘Mobile Infrastructure Sharing’ (GSMA 2012) (although noting that it is technologically
challenging and involves different considerations for different market players according to their status).
227
Recital 10, Directive 2014/61/EC.
228
See Ofcom, ‘Penalty Guidelines: Section 392 Communications Act, (2003)’, 14 September 2017, at 1–2
(noting deterrence as the primary purpose and the need for sanctions to be appropriately high to have an
impact).
34
229
Explanatory Memorandum, proposed EECC, at 2.
230
Ibid, at 3 (noting only ‘modest’ Single Market results).
231
Report, ‘House of Commons Select Committee on European Scrutiny, Digital Single Market: Connectivity
(Telecoms) Package’ (UK Parliament, 25 April 2017), <https://publications.parliament.uk/pa/c m201617/
cmselect/c meuleg/71-x xxvii/7114.htm>.
232
These would encompass the range of end-u ser protections under the conditions to the general author-
ization such as transparency and minimum service quality requirements, minimum contract requirements,
and restrictions case.
35
about the impact on innovation and national/EU start-ups and are urging caution
in imposing equivalent regulation on OTT providers absent evidence of market
failure or consumer harm. As nothing is agreed until all is agreed, the proposed re-
forms remain uncertain. The Council has set a deadline of June 2018 for negotiation
agreement.
2003, Chapter 21.
233
Ofcom, ‘Original Notification setting general conditions under section 45 of the Communications Act
234
rights, obligations, and definitions, implementing sections 51–6 4 of the Act. These
specify the permissible content and scope of general conditions which the Act per-
mits to be applied ‘generally’ to every person providing an electronic communica-
tions network or service (s 46(2)(a)), or to every person providing those networks or
services of a particular description as defined in the condition (s 46(2)(b)). These
will be examined, as recently revised subsequently.
236
This separate listing of associated facilities in the Communications Act 2003 addressed a gap in the EU
framework. The Framework Directive defines its scope and aim at Art 1(1) as a ‘harmonized framework for
the regulation of electronic communication services, electronic communications networks, associated fa-
cilities and associated services’. It then proceeds to define these latter categories as ‘facilities associated with
an electronic communications network and/or an electronic communications service which enable and/or
support the provision of services via that network and/or service. It includes conditional access systems and
electronic programme guides’ (Art 2(e)). The Authorisation Directive’s aim and scope, however, states only
that it applies to ‘authorisations for the provision of electronic communications networks and services’ and
carves out as unnecessary conditional access system/services authorizations at Recital 6, provisions having
previously been made for the free movement of conditional access services in Directive 98/8 4/EC on the legal
protection of services based on, or consisting of conditional access. It appears implicit in this, therefore, that
associated facilities are services other than conditional access, while defined separately, are intended to fall
within electronic communications networks and services authorizations.
237
See Ofcom Consultation, ‘Proposal that all provisions continued from licences made under the
Telecommunications Act 1984 and all continued interconnection directions will cease to have effect except
for specific provisions in specific markets listed in this document as exceptions’, 9 September 2004, <http://
www.ofcom.org.uk/consult/condocs/P rop1984tele/provis_terminiate/>.
238
Oftel, Statement of DGT, ‘Guidelines for the interconnection of public electronic communications net-
works’, 23 May 2003.
37
and that the register added nothing since it did not itself create or condition the
exercise of rights.239 If the BEREC notification procedure under the proposed
EECC remains to foster cross-border market entry, these requirements may need
to be revisited.
6.4.4.3 Conditions
Section 45 of the Act authorizes Ofcom to set the general conditions of entitlement
(s 45(2)(a)) and specified individual conditions comprising: (i) universal service
conditions; (ii) access-related conditions; and (iii) significant market power condi-
tions (s 45(2)(b)).240 Each is considered in turn below.
See Consultation, n 237.
239
Section 45 also authorizes conditions on providers with exclusive and special privileges from other in-
240
dustries where relevant communications revenues exceed £50 million. None have been designated.
241
Ofcom recently grouped these into three main categories of ‘network functioning’ (Part A); ‘numbering
and technical conditions’ (Part B), and ‘consumer protection’ (Part C) in its consultation and revision of the
General Conditions of Entitlement, Statement and Consultation, n 235, at 2.2.
242
Italics represent implementation of the 2009 EU amendments.
38
Finally, the Act’s permitted general conditions for ‘scarce resources’ concern:
• access for end-users to numbers under the national numbering plan (s 57);
• the allocation to and adoption of numbers by providers and non-providers
(s 58); and
• the conditions for limiting any transfers of allocated numbers to another party
(s 56A).243
Falling within ‘consumer protection’ are those conditions under the Act regarding:
The Digital Economy Act 2017247 amended section 51(2) of the Act by adding a
new subsection (da) that specifies Ofcom’s power to set conditions requiring a
243
This section also details the obligation to justify any time limitations on number allocation as discussed
previously at Section 6.4.2.3.
244
This obligation together with the right granted under new s 146A of the Act to third parties to use any
published information for provision of an interactive guide or other technique to evaluate alternative service
usage costs implements Art 23, Universal Service Directive as amended by Directive 2009/136/EC.
245
Requiring Ofcom’s notification to the Commission and BEREC and that it take ‘due account’ of the
Commission’s comments and recommendations (s 52(2A)).
246
The italicized text indicates changes based on the 2009 reforms, implemented via The Electronic
Communications and Wireless Telegraphy Regulations 2011.
247
Digital Economy Act 2017, ch 30 (27 April 2017).
39
Sections 46–49C reflect the Act’s implementation of the EU procedural and sub-
stantive requirements for publication, consultation, approval of domestic conditions
and those with EU significance requiring Commission notification, and modification
and revocation of conditions, in light of the 2009 reforms.249
Following on from its 2015 Digital Communications Review, Ofcom conducted a
review and revision of the General Conditions (GCs) of Entitlement. One of its pri-
mary goals going forward from the Review was to ensure ‘a step change in quality
of service’ and the ‘empowering and protecting’ of consumers.250 Revisions to the
General Conditions, that will be reduced to seventeen with effect from October
2018, to address these include:
See Communications Act 2003, ss 48(A)–49(C) regarding notification to the Commission of imposition,
249
251
That being said, the September 2017 publication of the revised, consolidated Conditions is already no
longer complete in light of the November 2017 addition of GC 24 requiring enhanced transparency for SMEs
regarding service levels that was simultaneously revised as C2.16–2 .19. See Table 6.1.
314
The revised General Conditions are outlined in Table 6.1. These are worth re-
viewing, as they will comprise the bulk of the regulatory framework for many
providers of networks and services. As with the prior GCs, the revised scheme
generally distinguishes among three different categories of providers: providers
of electronic communications networks (ECN) or services (ECS), providers of
public electronic communications services (PECS) and networks (PECN), and
providers of publicly available telephone services (PATS)252 as well as PECN
networks over which PATS are provided (a seeming substitute for the PTN/
PCN previously used by Ofcom in particular conditions referencing telephony).
252
While it previously did so, the definition of ‘Publicly Available Telephone Service’ (PATS) no longer en-
compasses access to emergency services as a defining criteria and references only a service for originating and
receiving, directly or indirectly, national or national and international calls through a number or numbers in
a national or international telephone numbering plan. There are also provisions applicable to public internet
access services (PAIS). See C 3.
324
253
As noted by Ofcom, the definitions relating to Conditions reflect the change from ‘Public Telephone Network’
to ‘Public Communications Network’ (PCN), and the amendments to PATS and telephone number under the 2009
EU framework. See Consolidated Version of General Conditions of Entitlement as at 13 September 2011, at n
2. These continue but there are further divisions, eg Public Internet Access Service (PAIS) in C 3.4.
254
Very minor revision of GC 1.1.
255
Largely replicates GC 1.2; omits Ofcom disclosure exemption.
256
Replicates GC 2.1, 2.2; omits 2.3–2 .6, Ofcom powers re: standards.
257
Removes limitation to fixed networks.
258
Minor revision of GC 3.1.
259
Transposes para 11(a), former Annex 3, GC 14.
260
Replicates GC 4.2.
34
▪ If at fixed location, caller location information must at least include terminal equipment
location and full postal address (A3.6 (a)).261
▪ For mobile services, the cell identification of the cell from where the call is made and
radius of cell coverage where available
◆ zone code, exceptionally, if cell identification temporarily unavailable for technical
reasons (A3.6 (b)).262
▪ For VoIP Outbound Call Services at fixed location, providers to recommend that domestic/
small business users register address prior to service, keep updated (A3.6 (c)(i)).263
▪ Where VoIP Outbound Call Service reasonably expected to be accessed from multiple
locations, provider to recommend domestic/small business users register location data
associated with it update whenever accessed from new location (A3.6 (c)(ii)).264
A4: Emergency Planning
• Providers of PATS or PECNs over which PATS provided:
◦ to make arrangements to provide/restore rapidly reasonable and practicable services in a
disaster on request of/i n consultation with central and local government and EOs.
◦ to implement arrangements as requested by any designated person as is reasonable/practicable
◦ may seek compensation and be conditioned on indemnification.265
A5: Must carry obligation
• Regulated providers (designated broadcast network providers) to:
◦ comply with direction from Ofcom to transmit service from must-carry list under Section
64 of the Act.
◦ comply with any order of Secretary of State under Section 64 re: terms on which services
must be broadcast or otherwise transmitted.266
Part B: Numbering and Technical Conditions
B1: Allocation, adoption and use of telephone numbers267
• Provider of ECN, ECS:
◦ not to adopt, use or transfer numbers from national numbering plan unless allocated to it
or to another person who authorizes adoption, use.
◦ to comply with applicable restrictions, requirements of National Numbering Plan or in
Ofcom notifications recording specific number allocations to it.
◦ to ensure effective, efficient adoption/other use of allocated/t ransferred numbers; take
all reasonably practicable steps to secure that its customers’ use of numbers comply with
Condition, National Numbering Plan and Non-provider Numbering Condition.268
263
Essentially transposes 12(a), Annex 3, GC 14.
264
Essentially transposes 12(b), Annex 3, GC 14.
265
Largely transposes GC 5.1–5.3; adds specification of radioactive/toxic/other events with significant im-
pact on general public as disasters.
266
Transposes GC 7.1, 7.2.
267
Largely replicates GC 17; omits GC 17.11, 17.12 (allocation/w ithdrawal of numbers for limited period),
17. 20, 17.21 (pre-2015 application).
268
A 2013 transparency condition that requires that calls to non-geographic numbers be divided into their com-
ponent parts, the access charge by their communications providers and the service charge by company being called,
which must show the applicable service charges on all advertising and promotional material that includes the non-
geographic number using Ofcom mandated wording: ‘This call will cost you X pence per minute plus your phone
company’s access charge.’ Ofcom, ‘Telephone call chargers to be made simpler’, 12 December 2013, <https://www.
ofcom.org.uk/aboutofcom/latest/media/media-releases/2013/telephone-call-charges-to-be-made-simpler>.
34
269
Places where Ofcom has identified a likely potential number shortage in its consultation on the General
Conditions of Entitlement.
270
Omits obligation to provide directory enquiry/operator access but otherwise largely replicates GC
8.2–8.6, GC 19.
271
Replicates GC 18 with minor edit re: plain English.
272
Transposes GC 18.2.
273
Transposes GC 18.3.
274
Transposes GC 18.4.
354
Transposes GC 18.5.
275
Transposes GC 18.9.
276
277
Transposes GC 18.10.
278
Transposes GC 18.7.
279
Replicates GC 20 but removes GC 20.4 re: no longer existing EU Telephony Space.
280
Replicates GC 9.2–9.6, 9.7; adds provisions re: details of pricing information and material changes to core
pricing (in bold) (at C1.2 (i) and C1.7–C1.9, respectively) and; substitutes ‘fixed’ commitment periods for initial
commitment periods.
364
281
The service quality failure transparency condition is in addition to a voluntary Industry Scheme for auto-
matic compensation recently approved by Ofcom for 18 months as a trial in lieu of a regulatorily imposed
scheme. See Ofcom, ‘Statement: Automatic compensation—protecting consumers from service quality prob-
lems’, 10 November 2017.
282
Must be distinct for each commitment period and in a manner allowing for informed choice. See
Definitions, Revised General Conditions of Entitlement.
374
283
Minor clarification re: pricing details.
284
Largely transposes GC 10.1–10.2.
285
C2.4–C2.8 essentially transpose GC 14.8–14.12.
348
• Where different tariffs applied to small business customers, provider to ensure pricing is
transparent; inform if a business tariff (C2.9).286
• For controlled premium rate services (CPRS), providers to provide domestic/small business
customers, on request and free of charge, advice and information about:
◦ UK CPRS mechanisms, such as operator billing, premium rate Short Message Service
(PSMS) payments, CPRS number service, voice shortcode charges, and how applied to the
customer’s phone bill;
◦ Provider’s role regarding:
▪ general CPRS enquiries, requests for number checks via number-checker facilities
provided by Phone-paid Services Authority on its website; and
▪ dealing with formal complaints about service content abuses, non-c ompliance with
Phone-paid Services Authority’s code of practice, other alleged unlawful operation of
services/numbers (C2.10).287
• Provider to include information about:
◦ basics of CPRS, including whether routed to service providers hosted on own network or
different network; how revenue shared
◦ applicable tariffs for calls to any CPRS number range; any access charge
◦ individual service provider or hosting communications provider’s contact details; where
info available
◦ service providers’ customer service contact details; where consumers can get info about
services provided on CPRS numbers found on their bills
◦ Phone-paid Services Authority’s role in complaints; how to make formal complaint via
their website/helpline or in writing
◦ alternative dispute resolution schemes’ role in resolving CPRS-related disputes
◦ how consumers can bar access to all/specific range of CPRS numbers for cost/content
reasons
◦ consumer refund options for scams/abuses (C2.11).288
• Required information publication to be effected by:
◦ sending a copy to any end-user reasonably requesting it, free of charge
◦ placing plain English copy prominently/easily accessible, on provider’s website or as
Ofcom directs if no website (C2.12).289
• Providers to have:
◦ procedures to ensure enquiry/helpdesk staff aware of above requirements to respond to
complaint/enquiries and monitor compliance with requirements (C2.13).
◦ fully documented procedures ensuring customers, advice agencies aware
of requirements’ existence, eg, by referring to them in sales/marketing materials
(C2.14).290
286
New condition, requiring general transparency as to fact of business tariff but not detailed contrast with
consumer prices.
287
Transposes s 3.2 of Annex 1, GC 14 as direct information obligations. In light of these, the requirement for
Code of Practice regarding provision of information to consumers is removed as discussed above. (Removes
GC Condition 14.6.)
288
Transposes s 3.3, Annex 1, GC 14 as direct obligation.
289
Largely transposes GC 10.3, removes requirement for posting at major offices.
290
C2.13 and 14 transpose ss 4.1–4.2, Annex 1, GC 14 as direct obligations.
394
• PECN/PECS providers providing public pay phones to display/take reasonable steps to keep
displayed on/a round all public pay phones, notice of:
◦ minimum charge for call connection
◦ location info sufficient to enable EO’s swift location
◦ emergency calls to ‘112’ or ‘999’ are free with no coins/cards needed
◦ whether phone able to receive calls, and, if so, the phone number (C2.15).291
• PECN/PECS providers to publish, in plain English and reasonably prominent/easily
accessible on its website or other place as per Ofcom direction, information re: standard fixed
voice/other fixed services/broadband contracts for SMEs that includes:
◦ service level agreements, if any, regarding:
▪ activating the service on a confirmed date and for failing to do so;
▪ the event of a loss of service;
▪ keeping a pre-agreed appointment to the SME’s premises and for failing to do so.
◦ service level guarantees, if any of the above.
◦ whether no agreement/g uarantees exist.
◦ whether may be available on individual negotiation (C2.16–C2.17).292
• Where SME enters into an agreement for such services whether standard or bespoke, provider
to provide the above information with respect to the contract in a durable medium distinct
from the contract (C2.18–2.19).
C3: Billing requirements
• PECS providers not to charge/bill end-user for PECS provision unless every charge represents
true extent of provided service (C3.2).293
• PECS providers, subject to data protection requirements, to maintain records for at least 12
mos. to establish compliance (C3.3). 294
• Providers of PATS/P ublicly Available Internet Access (PAIS) with revenues not less than
£55 million to:
◦ comply with direction that Ofcom may issue on process/standards for approval of total
metering and billing systems (C3.4).295
◦ apply to approval body for approval of total metering/billing systems according to Ofcom
directed process, obtaining approval as soon as practicable and complying with approval
body direction for approval (C3.5).296
◦ take approval body recommended action where approval withdrawn/not granted or cease
use of system; inform Ofcom of either date (C3.6).297
291
Transposes GC 6.2; omits other payphone provision, accessibility, design requirements as either re-
dundant of general law (Equality Act 2010) or unnecessary in light of market developments (NGT Lite app on
smartphones obviating need for text payphones).
292
Transposes the new GC 24 that Ofcom recently set with effect from the period of 1 June 2018 to 1 October
2018 when the Revised Conditions are effective. See Ofcom, ‘Statement: Automatic Compensation—pro-
tecting consumers from service quality problems’, 10 November 2017, at Annex 2.
293
Transposes GC 11.1.
294
Transposes GG 11.2, directly specifies the minimum period.
295
Effectively transposes GC 11.7(e), extends metering and billing system obligations to data via inclusion
of PAIS in scope.
296
Transposes GC 11.4.
297
Transposes GC 11.5.
530
298
Largely transposes GC 12.1, in part.
299
Largely transposes GC 12.2.
300
Transposes GC 12.1’s ability to charge reasonable fees but limits to written bills.
301
Transposes GC 12.4, details ‘999’ and ‘112’ as free calls, specifies ‘SMS’.
302
Transposes GC 13.1, includes PAIS.
303
Transposes GC 13.2, specifies without charge, publication attributes.
304
Effectively transposes GC 14.4, 14.5, and Annex 4 to GC 14.
305
Annex, Condition C4 encompasses the Ofcom code for consumer service and complaints handling set-
ting out high-level minimum standards for accessible processing procedures (Section 1) and consumer com-
plaint codes (Section 2), including information provision requirements and standards, as well as obligations
to retain for at least 12 months from resolution/closing, accessible written records re: complaint, handling
and resolution for compliance monitoring purposes as well as complaint metadata (eg monthly complaints,
resolutions, ADR letters, etc.) (Section 3).
306
Contained in Annex to C3.4 requiring: timely complaints processing procedures with prompt handling
until resolved (where after 28 days after consumer advised of outcome, does not indicate dissatisfaction); ac-
cessibility by disabled, vulnerable customers; ability to make complaints by mail, email/webpage form free/
geographic phone numbers: staff training and posted procedures; prompt issuance of ADR letter in plain
English, durable medium where customer indicates not satisfied with outcome of provider’s investigation
and with details about independent ADR scheme contact info, and right to pursue without cost. Consumer
bills also to inform of rights to no-cost, independent ADR access for unresolved complaints ordinarily after 8
weeks, contact details, existence, location of Complaints Code (Section 4, Annex).
351
Enhances GC 15 to include requirements to consider and adequately address the needs of the vulnerable.
307
Includes circumstances such as age, physical or learning disability, physical or mental illness, low lit-
308
316
Transposes, enhances GC 22.3.
317
Transposes GC 22. 4.
318
Transposes GC 22.5, 22.6.
319
Transposes GC 22.7.
53
• For each contract entered into with switching customer, gaining provider, to create, keep for
not less than 12 months (irrespective of whether terminated before then):
◦ individually retrievable, direct record of consent to migrate services/begin acquiring
services via the target line.
◦ record of explanation that customer consent record required
◦ switching customer name, address
◦ time, means, place of consent
◦ salesperson, if applicable
◦ target address
◦ calling line information of target line (C7.7, C7.8)320
• Gaining provider to send to switching customer letter that clearly, intelligibly sets out: date,
fact that transferring services and relevant services to be transferred, estimated date
of migration, contract details, the calling line identification of all relevant transferred
communications services, right to terminate as above with specific applicable dates
(C7.10)321
• Losing provider to send letter, on paper or other durable medium and by post unless
otherwise explicitly agreed, advising clearly, intelligibly, in neutral terms that migration to
be effected without need for further contact to cancel existing services, date of migration,
bill to be sent after transfer, whether any contract early termination charges and relevant
explanation and estimate as of migration date, how to be paid, and the transfer’s impact on
any remaining services (C7.11, C7.12).322
• Where transfer of broadband and fixed line telecommunications services over same line,
gaining provider order to Openreach/KCom for simultaneous transfer to minimise loss of
service (C 7.13).323
• Where gaining provider elects to coordinate the CP migration on behalf of switching
customer and not involving a change of location,324
◦ Both GP and LP to adhere to Annex 1, (C7.14 (a)), requiring:
▪ GP to place transfer order in reasonable time
▪ LP not to issue ‘cancel other’ unless:
◆ verified slamming has occurred
◆ GP has failed to cancel transfer order at switching customer’s request as verified
◆ telephone line to be ceased in transfer period
◆ Ofcom directed circumstances
◆ industry forum agreed reasons unrelated to switching customer’s request to cancel,
agreed by Ofcom.
▪ LP to confirm order cancellation by durable medium to switching customer unless not
appropriate/possible
▪ LP to record reason in each case with appropriate code as approved by Ofcom for
such as:
◆ switching customer never had contact with GP or authorised a transfer
320
Transposes GC 22.8, 22.9.
321
Transposes GC 22.11.
322
Transposes GC 22.12, GC 22.13.
323
Transposes GC 22.14.
324
Essentially transposes GC 22.16–22.20.
534
325
Transposes Annex 1, GC 22.
326
Transposes GC 22.25.
327
Essentially transposes GC 22.22.
328
Transposes Annex 2, GC 22.
329
Transposes GC 22.26–22.29.
330
Essentially transposes GC 23.2 but with focus on accuracy of information.
35
▪ if acting as retailer, it creates, keep sales records for six months and related sales incentives
for 90 days after redemption date, but not less than 6 months with date, means and place of
contract (if available); not applicable to pre-paid or SIM only (C8.7);
• Providers to publish summary of C8 obligations on website, easily accessible and in
prominent manner, or other manner as Ofcom may order; provide free of charge copy to
customer on request (C8.3).331
• Providers to monitor, ensure own retailers aware, comply with condition;
make reasonable efforts to ensure third-p arty retailer compliance, sanction
non-c ompliance (C8.4). 332
• Providers to ensure retailers (not of prepaid/SIM only) appropriately trained (C8.8).333
• Before entering, amending contract (except for pre-paid, SIM only), providers to reasonably
endeavour to ensure customers authorized, intend to enter contract and have clear,
comprehensible, accurate information in durable medium (or if by phone for phone sales
shortly thereafter, in good time) about:
◦ contracting party’s legal identity, address, telephone, fax and/or email;
◦ description of service, key charges including: contract minimums, applicable early
termination; payment terms; any termination right and procedures; likely service date
if not immediate; any fixed commitment period; and for consumers, any relevant access
charges for calls to unbundled tariff numbers (C8.5).334
• Provider to ensure relevant services are available for customer to receive (C8.6).335
• Providers to ensure that it (reasonable endeavours to ensure that it or a person acting on
its behalf ) carries out and retains for its mobile service retailers (not including prepaid/
SIM only) a minimum of a check of credit references, director disqualification, director
of entity with bankruptcy/administration filing, ongoing checks for relevant updates of
this information, information provided by retailer to be kept confidential, used only for
monitoring, not given to anyone (eg, partners, subsidiaries) for whom it provides competitive
advantage (C8.9–8.10).
• Where customer to receive deferred sales incentive after contract entry, provider must ensure
terms & conditions not unduly restrictive, that customer receives in a durable medium
(unless by phone, durable medium to follow, in good time) clear, comprehensive, accurate
information that includes:
◦ Legal entity making sales incentive offer and undertaking obligations, its address, contact
detail (telephone, fax, email)
◦ Description of sales incentive and its terms & conditions; any process customer has to
follow to obtain the incentive (C8.11–8.12).336
331
Transposes GC 23.3.
332
Transposes GC 23.4.
333
Transposes GC 24.7.
334
Transposes GC 23.5 with access charge requirement added.
335
New requirement.
336
Transposes GC 23.10.
536
The current revisions have moved the definitions from each condition to a single
section at the end of the Condition Schedule but with the possibility that the
terms can still have a meaning particular to a specific condition if the context sug-
gests it.337 Each revised condition indicates the providers to which it applies in its
‘Scope’, the first section of each. A particular condition or part of a condition may
apply only to a subset of electronic communications service and networks pro-
viders. As noted, Ofcom has added a recital to each condition that explains what
it intends and to whom it applies but which has no legal effect. These are helpful,
however, as are the efforts to specify the actual requirements in the text rather than
mere cross-references.338 Yet, the GCs can still be somewhat difficult to under-
stand readily and there are often background issues that arise in consultations,
which give context. For example, the current revision is the product of a series
of consultations with the proposal and background rationales and set out in the
earlier documents that are cross-referenced but only partly explained. Similarly
frustrating is the failure to provide regularly updated consolidated versions with
any interim modification or at least a rolling index of all changes. The new condi-
tions that govern transparency about automated compensation schemes promul-
gated two months after the revision are just the latest example of changes that are
not readily apparent. Guidance and orders that can affect scope or interpretation
but which are not part of the GCs can as well create uncertainty.339 The GCs are
not models of clarity, therefore.340
The enhanced competition that flowed from early EU/U K liberalization reforms
produced some questionable sales and other marketing practices, such as slamming
(switching providers without customer consent), highly pressured sales pitches,
and retailer mobile cash-back schemes that defer its payment until much later and
then impose requirements not made clear to customers. Ofcom sought to address
these with conditions governing marketing transparency and sales practices.341 It
337
Ofcom, ‘Statement: Review of the General Conditions’, 19 September 2017, at Annex 14.
338
Not always adhered to. See eg definition of Controlled Premium Rate Services as having ‘the meaning
set out in the condition issued by Ofcom under section 120 of the Act’ with a footnote reference to the
2015 ‘Changing the implementation date of the new rules governing Freephone and revenue sharing ranges
from 26 June 2015 to 1 July 2015’.
339
See eg Ofcom, ‘Guidance on “Material Detriment” under GC 9.6 in relation to price rises and notification
of contract modifications’, 23 January 2014 (withdrawn as of the revision’s effect and some, but not all, of the
guidance specifications are now transposed to C1.7 and C1.8 and possibly as well the general transparency
requirements of C.2).
340
Nor is Ofcom’s website an aid to clarity.
341
Previously, GC 14, governing codes of practice, provided for the fixed-l ine marketing/sales code of prac-
tice to address ‘slamming’ or unauthorized transfers of accounts to another provider. With the need for a
mobile code, Ofcom promulgated both as distinct general conditions, then GC 23 and 24, removing the fixed
lines code from GC 14.
537
342
Annex 1, GC 14.
343
Ofcom, ‘Simplifying Non-G eographic Numbers—change in implementation date’, 26 February 2015.
344
Ibid. See also GCs 14, 17.
538
345
See GC 22; Ofcom, ‘Statement and Notification ‘Broadband migrations: enabling consumer choice’, 13
December 2006.
346
See Ofcom Media Release, ‘Easier broadband switching from tomorrow’, 19 June 2015.
347
In early 2003, Oftel set a list of key performance indicators (KPIs) as a checklist against which BT could
perform to attain relaxed retail price controls. Technically, therefore, these 15 KPIs against which perform-
ance was measured, while quality of service reporting, were voluntary and not conditions under the BT li-
cence. See Oftel Statement, ‘Wholesale Line Rental’, 11 March 2003.
348
See Ofcom, ‘Statement and Directions: Requirement on BT to publish Key Performance Indicators’, 23
September 2004, <http://w ww.ofcom.org.uk/consult/condocs/bt_ k pi/statement/statement_d irections.pdf>.
These comprised a range of month and/or quarterly reports regarding different performance parameters with
regard to end user access (data stream), wholesale line rental, virtual path facilities, FRIACO, and specified
interconnection circuits.
349
See ‘Our Undertakings: Key Performance Indicators’ (BT Group Plc London), <http://w ww.
btplc.com/ T heg roup/ R eg ulator yandPublicaf fairs/ O ur under ta k ings/ K eyPerformanceIndicators/
KeyProductPerformance Indicators/i ndex.htm>.
350
See eg Ofcom, ‘Quality of Service Direction for WLR: Direction setting further minimum stand-
ards for WLR provisions under the SMP condition imposed in the 2014 Fixed Access Market Reviews’, 22
November 2016.
351
Ofcom Notification of Direction, ‘A Statement on setting quality of service parameters’, 27 January 2005.
539
found the cost-b enefit was not justified. 352 The 2017 revisions of the General
Conditions eliminate GC 21. This however was in light of the new powers that
the Digital Economy Act 2017 grants to Ofcom to require and publish com-
parative quality of service information, broader than that in GC 21 and likely
rendering it unnecessary. 353
Despite not relying on GC 21 for quality of service metrics, Ofcom has pro-
duced annual consumer experience reports for nearly a decade and following
on from its Digital Communications Review, its first quality of service report
in 2017. 354 Ofcom found that consumers have experienced slow repairs and
installation delays and missed appointments for new and migrated services.
Although the Digital Economy Act 2017 has empowered Ofcom to impose an
automatic compensation for such service failures, while Ofcom was consulting
on such a scheme, the majority of fixed line/broadband providers proposed a
voluntary scheme in lieu of regulation. 355 The scheme will require them auto-
matically to pay residential service customers (that can include SMEs using
these services):
Ofcom will review its operation in a year to determine whether regulation will
still be needed. Ofcom, however, recently imposed additional SME transparency
regarding service level guarantees, adding GC 24 until October 2018 and then
within C2.16–C2.17 in the revised conditions.356
Ofcom’s quality of service concerns have also focused on general customer ser-
vice and complaint handling, finding that the sector trails behind others with
longer wait times, perceived lack of ease and flexibility, and consumer frustra-
tion.357 To address these, in the 2017 revisions Ofcom has honed and reinforced
354
See eg Ofcom, ‘Research Report: The customer experience’, 28 January 2015.
355
See ‘Communications Providers’ Voluntary Code of Practice for an Automatic Compensation Scheme
for service related issues relating to residential fixed-line telephony and broadband services’, 10 November
2017, at: <https://w w w.ofcom.org.uk/_ _d ata/a ssets/p df_ f ile/0 024/107691/A nnex-1-i ndustry-a utomatic-
compensation-scheme.pdf>.
356
See text accompanying n 292 above.
357
Ofcom, ‘Comparing Quality Service’, 12 April 2017, at 46–62.
630
the Ofcom Approved Code of Complaints Practice, the only GC 14 code of practice
retained (now Annex, C.4).358 Providers must provide, among other things, greater
signposting about complaint handling and faster access to alternative dispute
resolution once it is clear that the provider will not take further action to resolve
the complaint.359
Individual conditions
(i) Universal service conditions ‘Universal service’ is the first of section 45 of the
Communication Act’s permitted specific conditions that Ofcom may establish if it
deems appropriate for securing compliance with obligations set out in the ‘universal
service order’ by the Secretary of State for Trade and Industry (s 67).
In the Electronic Communications (Universal Service) Order360 the Secretary
of State originally defined the scope of the universal service obligation to in-
clude PATS, public pay telephones, directory and directory enquiry facilities,
special measures for disabled end-u sers, and special tariff and billing options,
including those for low income users. Pursuant to Communications Act, section
66(1), 361 Ofcom designated that universal service conditions apply to BT and
Hull (now Kingston Communications (KCOM)) but only within the latter’s geo-
graphical service area. Ofcom imposed specific USO obligations following the
Order362 that, although slightly modified after prior reviews, 363 still include the
obligation to:
358
Ofcom Consultation, ‘Review of alternative dispute resolution and complaints handling procedures’,
10 July 2008.
359
See text and accompanying nn 304–308.
360
2003 c. 21, SI 2003/1904.
361
Implemented by The Electronic Communications (Universal Service) Regulations 2003, SI 2003/33.
362
See Ofcom, ‘Strategic Review of Telecommunications Phase 1 Consultation, Annex G’, 2003.
363
eg in 2006, low-income schemes, including a pre-pay option, were approved, as was the ability to
modify some provision of public call boxes due to their cost and low utilization and the rules for removing
them, including a ‘local veto’ for qualifying boxes. See Ofcom, Statement, ‘Review of the Universal Service
Obligation’, 14 March 2006. None of these, however, altered the basic requirement in each of the areas, just the
extent of the obligation or how it may be satisfied.
364
Ofcom, in its 2006 review, determined that BT could charge non-u niform prices when the connection
cost was more than the standard charge of £3,400, although recommending that it use the standard charge
for particularly vulnerable customers. See ibid at 29. The Digital Economy Act’s broadband USO authorization
maintains this base cost limitation.
316
• provide at least one low-cost scheme for consumers with special social needs
who have difficulty affording telephone services;
• provide uniformly priced public call box services;
• ensure that tariffs for universal services do not entail payment for additional un-
necessary services;
• provide itemized billing at no extra charge;
• provide universal services that meet the defined quality thresholds;
• supply and maintain directories and databases for the provision of directory
services.365
The 2009 EU reforms required only limited changes to existing USO obligations.
Under BT’s revised condition 9 and KCOM’s condition 6 each must notify Ofcom
if it intends to dispose of all or a significant part of their local access network to a
separate legal entity under different ownership.
The Electronic Communications (Universal Service) (Amendment) Order made
few changes. For definitional consistency, it substitutes ‘public communications
network’ for ‘public telephony network’.366 The Order also limited USO special dis-
abled end-user obligations to where an ‘equivalence’ provision has been imple-
mented.367 In 2012, Ofcom removed Condition 4 regarding the provision of Next
Generation Text Relay from BT and KCOM in light of the modification to GC 15 that
required equivalent access of all communications providers.
Ofcom has also, to date, concluded that both BT and KCOM should continue to
bear the costs of the USO, in light of findings that the benefits to both of the USO
continue to equal or outweigh the costs. Therefore, no USO fund or other method
has been required for the current USO obligation.368
The Digital Economy Act 2017369 enables the adoption of a broadband USO with a
specified speed that must be at least 10mps.370 Both Ofcom371 and the government
365
Ofcom’s decision found that the USO Condition 7 requiring BT to provide any party the contents from
the OSIS database was not lawful as outside the scope of the Universal Service Directive’s obligation. This was
upheld in a March 2011 preliminary reference decision by the CJEU in C-16/10, The Number Ltd and Conduit
Enterprises Ltd. Thus, cost-orientated access to BT’s OSIS data set by other providers is beyond the scope of the
Universal Service Obligation 6.
366
See the Electronic Communications (Universal Service) (Amendment) Order 2011, SI 2011/1209, Art 5(a).
367
Ibid, at 4.
368
See Ofcom Statement ‘Review of the Universal Service Obligation’, 14 March 2006. Reportedly, BT would
like the obligations removed in connection with its NGA roll out commitments.
369
Digital Economy Act 2017 c. 30 (27 April 2017). 370
Ibid, s 1, Pt 1.
371
Ofcom, ‘Designing the Broadband universal service obligation: Call for inputs’, 7 April 2016.
632
(ii) Access conditions Ofcom is authorized by the Act to impose conditions con-
cerning the provision of network access and service interoperability appropriate
to secure provider efficiency, sustainable competition, and the greatest possible
benefit to end-users (s 73(2)). Where a person controls access to any electronic
communications network, that person may have an access condition imposed
on him without being a provider of a Public Electronic Communications Network
(PECN) or of associated facilities (s 46(6)). Otherwise, specific access conditions
must be imposed on providers of networks.375 Sections 73 and 74 specify the per-
mitted content of such conditions and include those relating to network access
and service interoperability considered appropriate by Ofcom in light of the
Framework Directive’s regulatory considerations (s 73(2)). These include specific
conditions to require interconnection of networks for the purpose of ensuring
end-to-end connectivity for end-users of PECNs (s 74(1)). In 2006, in order to en-
sure end-connectivity for telephony,376 Ofcom imposed an access condition on
BT. Before this no such condition had been imposed on BT, yet BT and the market
acted as if BT had such a connectivity obligation as a universal service operator,
following earlier guidance in this regard.377 Also included are obligations on a
372
Dept for Digital, Culture, Media & Sport, ‘A new broadband Universal Service Obligation: consultation
on design’ (July 2017).
373
Ibid.
374
Fildes, N, ‘BT’s £600m rural broadband offer rejected’ (Financial Times, 19 December 2017), <https://
www.ft.com/content/ebaf1ed6-e4e2-11e7-8b99-0191e45377ec>.
375
Section 65 requires that Ofcom impose access conditions of providers of conditional access services for
protected programmes.
376
End-to-end connectivity ensures that retail customers can make calls to other customers on that same
network or any other network.
377
T-Mobile et al v Ofcom [2008] CAT 12, 28 (citing Guidance issued by the former Director General of
Telecommunications on ‘End-to-end connectivity’ dated 27 May 2003).
36
person providing facilities for the use of application programme interfaces or elec-
tronic programme guides (s 74(2)).378
Section 73 was amended to permit an access-related condition to be set requiring
the sharing of infrastructure. Section 73(3A) indicates that this is to be exercised for
the purpose of ‘encouraging efficient investment in infrastructure’ and ‘promoting
innovation’, a balance that the 2009 framework and NGA recommendations require
of NRAs. The only NGA access conditions, however, are SMP conditions in relevant
wholesale access markets imposed on BT and KCOM.379
The non- SMP specific access- related conditions within the parameters
of the Act are conditions to provide conditional access and electronic pro-
gramme guide services on fair and reasonable terms that are published, on a
non-d iscriminatory basis, and maintaining accounting separation. 380 These
were originally applied to Sky entities with others applied as the pay TV market
evolved. 381 Other types of specific access conditions have been imposed in con-
nection with other PECS.
See Ofcom, ‘Review of the Wholesale Local Access Market’, 7 October 2010. These may be in addition
379
to the possible non-f ramework possible infrastructure sharing pursuant to the Communications (Access to
Infrastructure) Regulations 2016 that implement the Broadband Cost Reduction Directive 2014.
380
See Oftel Explanatory Statement and Notice, ‘The Regulation of conditional access; Setting of regulatory
conditions’, 24 July 2003; Ofcom Consultation, ‘Access regulation, regulation of electronic program guides’,
18 August 2005, at s 3, <http://w ww.ofcom.org.uk/consult/condocs/epg/epg/stat_provisions/>. Similarly, see
Ofcom Statement, ‘Technical platform services: Guidelines and explanatory statement’, 13 September 2006.
Also, see Chapter 8.
381
See Ofcom Consultation, ‘The setting of access-related conditions upon Top Up TV’, 15 February 2007.
The final statement expected in May 2007 has not been found on the Ofcom website.
382
2002/21/EC, Art 14(2).
634
The process is also one that market participants seem increasingly willing to
challenge. 384 There is a growing similarity to the US market, with its vast num-
bers of parties lined up on each side of an issue with seemingly perpetual chal-
lenges to the FCC’s regulatory measures that can take years to resolve finally,
and often after the market has reached another solution or the technology has
moved on.
(iv) Privileged operator The last of the section 45 specific conditions concerns
public communications providers with special or exclusive rights regarding
the provision of any non-c ommunications services (s 77(2)). 385 Conditions to
ensure transparency and service provision without cross-s ubsidies from the
privileged business must be applied but not where revenues from all com-
munications activities are less than £50 million (s 77(4)). The conditions may
include separate accounting, audit, and published financials and structural
separation (s 77(3)). Ofcom has not designated any ‘privileged operators’.
384
See eg BT v Ofcom [2017] CAT 25 (successfully challenging Ofcom’s definition of a single relevant market
for all bandwidths of contemporary interface symmetric broadband origination (CISBO) and imposing of a
SMP dark fibre access remedy); Talk Talk v Ofcom [2013] EWCA Civ 1318 (upholding unsuccessful challenge
that Ofcom’s application of charge control condition six months after SMP determination failed to comply
with s 86 requirement that either the condition be set upon or after determination that market conditions had
not materially changed, as OFCOM was aware of Talk Talk’s likely entry into relevant exchanges at the time
of determination).
385
This is not the case where this is solely in connection with associated facilities.
386
Ofcom has set out how its enforcement investigations will proceed, including the s 96A breach notifica-
tion in its ‘Enforcement Guidelines for regulatory investigations’, 28 June 2017.
387
With repeated, non-serious breaches, intermediate penalties must be sought.
63
388
See Ofcom, ‘Penalty guidelines’, 13 June 2011, <http://w ww.ofcom.org.uk/about/policies-a nd-
guidelines/penalty-g uidelines/>.
389
Ofcom, ‘Penalty Guidelines, Section 392 Communications Act 2003’, 14 September 2017, <https://w ww.
ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 022/106267/Penalty-Guidelines-September-2017.pdf>.
390
Ofcom Media Release, ‘BT to be fined £42m for breaching contracts with telecoms providers’, 26
March 2017.
391
See Ofcom Media Release, ‘Vodafone fined £4.6 million for failing customers’, 26 October 2016.
367
Ofcom has enforcement powers under the Communications Act 2003 beyond
the section 45 general and specific conditions. These include, inter alia, powers
concerning: the electronic communications code governing rights of way (ss 106–
119); premium rate services (ss 120–124); administrative fees payment (ss 38–43);
information provision (s 135–144); and network security requirements (ss 105A–D).
The enforcement powers were enhanced under the 2009 reforms to the framework
with increased financial penalties across the board for enforcement (potentially
up to 10 per cent of turnover for the relevant period), including for the section 45
conditions with daily fines possible for up to 1 per cent of the maximum lump sum
penalty where there are continuing contraventions.
There are also circumstances where the Act makes the failure to comply with a
condition or authorization requirement a criminal offence. These include, eg the
provision of a network, service, or associated facility when the entitlement to do so
is suspended or so restricted (s 103), and the failure to provide required informa-
tion392 (s 143).
The following considers the electronic communications code, premium rate
services, and administrative charges regulation.
392
Section 135 empowers Ofcom to require providers and other persons to provide justified, proportionate
information for specified purposes including determining a condition breach or to ascertain or verify a pay-
able charge, universal service reviews, relevant market and market power analyses, statistical purposes, etc.
This has been revised to encompass network security obligations. Here as well there are enhanced potential
penalties with the maximum increased from £50,000 to £2 million.
393
Currently Sch 2, Telecommunications Act 1984 retained via deeming provisions under Sch 18 of the
Communications Act 2003.
394
Digital Economy Act 2017, pt 2, s 4.
395
These transitioning provisions, inter alia, disapply new Code provisions concerning assignment, up-
grades and infrastructure sharing to agreements under the current Code. Digital Economy Act 2017, pt 2, s 4,
Sch 2.
638
2017.396 As previously, the Code will apply pursuant to a direction from Ofcom
under section 106(3)(a) of the Communications Act to operators for providing
their networks and providers of infrastructure systems to operators for use by
them to provide their networks for agreements with occupiers of land for these
purposes.397 The applicable rights provided are to:
Under the Code, operators can also automatically upgrade apparatus or share
its use with another operator where the resulting changes have no or only min-
imal adverse impact on its appearance and impose no additional burden on the
landowner. 399 The practical implications of these qualifications to the right re-
main to be seen since upgrades or sharing may require additional equipment
and/or site maintenance. An operator can assign its Code rights to another.400
The Code delimits the ability to contract out of these statutory rights where the
above conditions have been met or to impose additional conditions, including
conditions for further payment.401 Thus, landlords may have reduced control
over their land.
396
The Digital Economy Act 2017 (Commencement No. 3) Regulations, SI 2017/1286 (c. 119).
397
Digital Economy Act 2017, pt 2, s 4, Sch 1.
398
Schedule 3A, Communications Act 2003, s 5. 399
Ibid, s 17. 400
Ibid, s 16.
401
Ibid, s 17(5). This does not affect the ability to impose guarantor status on the operator for purposes of
ensuring the assignee meets its obligations.
369
402
In respect of England and Wales. The Electronic Communications Code (Jurisdiction) Regulations 2017,
SI 2017/1284 (14 December 2017) (also establishing original jurisdiction in the Lands Tribunal for Scotland
and continuing the jurisdiction in the country court for Northern Ireland; in England, the First Tier Tribunal
can hear cases referred to it by the Upper Tier Tribunal).
403
Communications Act 2003, s 21, Sch 3A. 404 Ibid, s 21(5).
405
Ibid, s 24.
406
Rathbone, D, Briefing Paper CPB7203 ‘Reforming the Electronic Communications Code’ (House of
Commons Library, 1 June 2016) 13.
407
See eg Crest Nicholson (Operations) Ltd v Crest Nicholson (Operations) Ltd v Arqiva Services Ltd and others
(Cambridge County Court, 28 April 2015) unpublished.
370
property law creates similar certainty.408 The Code continues existing provisions
that allow for tenure of rights with successors in interests without the need for
their registration.409
The new Code also changes the termination procedures. The landowner must
now provide eighteen months’ notice rather than the current twenty-eight days
and specify grounds for termination that must be one of the following:
The operator has three months to serve a counter-notice indicating that it does
not want the agreement to end and wants the owner to either continue to be bound
under the former Code or under the new Code.411 Within three months from the
date of service, it must file with the court for an order. The court may not grant the
order if it finds that the site provider has not established any of the above grounds
but must do so otherwise.412 The Code makes provision as well for temporary rights
pending a final order413 and for removal of equipment including after a court re-
fuses to enter an order for a new/extended agreement or where the equipment is
unused.414
The new Electronic Communications Code requires Ofcom to establish a Code
of Practice that addresses the provision of information by operators, the conduct of
negotiations and of operators in relation to persons with an interest in land under
the Code.415 Ofcom must also develop standard terms that the parties may use for
408
See eg Peel Land and Property (Ports No.3) Ltd v TS Sheerness Steel Ltd [2013] EWHC 1658 (Ch).
409
Communications Act 2003, s 4, Sch 3A. 410
Communications Act 2003, s 31, Sch 3A.
411
Ibid, s 32 (for prior Code agreements). 412 Ibid.
413
Communications Code 2003, ss 25–26, Sch 3A.
414
See Communications Act 2003, Pt 6, Sch 3A.
415
Ofcom Statement, ‘Electronic Communications Code: Digital Economy Act: Code of Practice, Standard
Terms of Agreement and Standard Notices’, 15 December 2017. The Digital Economy Act, however, does not have
a provision requiring operators to comply with it, rendering its status non-binding. The Code of Practice states
that it suggests ‘best practice’ and in defining its scope indicates that it ‘provides a reference framework’. Ofcom,
‘Electronic Communications Code: Code of Practice’, 15 December 2017.
731
agreements416 and for template notices that the parties must use where required
under the Code417 and may use, otherwise.418
Under Communications Act provisions not modified by the Digital Economy
Act 2017, Ofcom must maintain a public register of providers granted Code powers
(s 108), currently 125.419 Many on the current register previously held Code powers
under the former regime (retained via deeming provisions under Schedule 18 of
the Communications Act).
In order to obtain a grant of Code powers, a provider of networks or infrastruc-
ture systems for networks must still make a standalone application420 to Ofcom
with the following:
Guidance indicates that a business plan should evince the need for the
grant as part of the application.422 Applicants must also provide evidence of
ability to meet any fiscal liabilities under the Act.423 This can encompass let-
ters from potential guarantors or company directors indicating willingness
416
Ofcom Statement, ‘Electronic Communications Code: Standard Terms’, 15 December 2017 (a template
agreement).
417
Communications Act 2003, ss 88–89, Sch 3A.
418
Ofcom, ‘Electronic Communications Code template notices, December 2017 (file of various template no-
tices (eg requesting rights, assigning rights, requesting removal of apparatus, etc).
419
See Register, <https://w ww.ofcom.org.uk/phones-telecoms-a nd-i nternet/i nformation-for-i ndustry/
policy/electronic-comm-code>.
420
Those with grants of Code powers under PTO or other individual licences were deemed to have Code
powers without new application. See Statement DGT, ‘Statement: The Granting of Electronic Communications
Code by the Director General of Telecommunications’, 10 October 2003, at 2.4.
421
DGT, ‘Notification under Section 106(2) of the Communications Act 2003: Requirements with respect to
the content of an application for a direction applying the Electronic Communications Code and the manner
in which such application is to be made’, 10 October 2003.
422
Statement, n 420, at 2.64. 423
Section 109(e), Communications Act 2003.
732
424
Statement, n 420, at 2.85–2 .88.
425
Ibid, at 2.57 (noting the factors that will be considered in this balancing). See, eg, Ofcom Consultation,
‘Proposal to apply Code powers to IX Wireless Ltd’, 5 January 2018, at 2.13–2 .22.
426
Statement, n 420, at 2.101.
427
These are authorized by the Communications Act 2003, s 36(1)(d).
428
Ibid. See Ofcom’s Tariff Tables, 30 March 2017.
429
See Consultation, n 380.
430
See Section 6.2.2. Indeed, a significant number of code operators have requested that their code powers
be revoked. See the 20 Directions revoking these at the A–Z Document List, <http://w ww.ofcom.org.uk/atoz/
?letter=D&publication=All§or=All>.
73
Code operators must comply with obligations imposed under the Electronic Com
munications Code (Conditions and Restrictions) Regulations 2003434 (Regulations).
These replaced the former licence conditions and restrictions although are largely
unchanged substantively.
These, inter alia, detail requirements for providers exercising Code powers
in connection with local planning. An example of their interaction with the
planning regime is that while most apparatus installation does not require even
notice to local planning authority under the General Permitted Development
Order (GDPO), the Regulations require a month’s notice to local planning au-
thorities and compliance with their reasonable conditions where the provider
has not previously installed apparatus in the area or plans to install certain sized
cabinets and boxes that do not require planning permission.435 The Regulations
also provide for notice to local planning authorities of fifty-six days, and other
compliance requirements for works in connection with listed buildings and an-
cient monuments, conservation, and other protected areas.436 The Regulations
seek to minimize the aesthetic, environmental, and functional impact of the in-
stallation appropriate to the public or private land on which it is installed437 by,
eg requiring underground installations to be deep enough so as not to interfere
with its use, such as agricultural land. They require coordination between pro-
viders and others installing utilities,438 or coordinating public works such as
highway and road authorities, and impose inspection and maintenance obliga-
tions for safety to persons and property.439
The Regulations provide more detailed procedures for ‘funding liabilities’ than
the former licence conditions. These involve an annual section 16 certification on
1 April by the provider, if an individual, or its board, detailing amounts available
and how this determination was made.440 This encompasses ‘relevant events’ trig-
gering the need for the funds’ availability, including a specified insolvency level.441
Providers not exercising their Code powers must certify two weeks prior to doing
so. Ofcom maintains a list of filed certifications.
Ofcom’s predecessor had no powers to take any specific action for a breach of
a Code condition as such breaches would likely be in violation of private rights
actionable in the courts or comprise breaches of other statutes such as un-
authorized street works.442 The Communications Act 2003, however, authorizes
Ofcom to specify directions for remediation and to issue penalties (s 110(2)(e)),
including financial, under revised procedures similar to those for other con-
ditions which may include a daily penalty up to £100 per day for a continuing
contravention while a fixed penalty may not exceed £10,000 per contravention
(s 110A).
Suspension of the Code application to a provider is possible for repeated or ser-
ious contraventions of the Regulations (s 112), for urgent cases necessary to protect
health and safety or the economic or operational interests of others (s 111A),443 and
for serious or repeated failures to pay the administrative charge (s 113(1)). These
would appear to apply to the entire network.444
The grant exists as long as the network or conduit system provider does unless
suspended or revoked on request.
438
Ibid, at reg 14. 439 Ibid, at reg 10.
440
The annual certificate shall, in the case of a company state, that in the Code Operator’s Board’s reason-
able opinion, the Code Operator has fulfilled its duty to put funds in place in compliance with the Regulations,
the systems and processes which enabled the Board to form that opinion, and the amount of funds which
have been provided for. A copy of the relevant instrument that will provide the funds should accompany the
certificate.
441
SI 2003/2553 (as amended), n 434, at reg 16.
442
Consultation on the Draft Electronic Communications Code (Conditions and Restrictions)
Regulations, 3 (DTI, April 2003), <http://w ww.communicationsbill.gov.uk/i mplementation_c onsult-
ations.html>.
443
Requiring confirmation/removal within three months (plus one further extension of three months).
444
Except to the extent that they concern unconfirmed urgent cases.
735
Premium rate services Section 120 of the Communications Act 2003 author-
izes a condition that is general in nature as it may be imposed on all persons pro-
viding premium rate services (PRS) or to a person providing a specific description
of such services. PRS comprise those goods and services that people can buy like
chat lines, call-i n contests, access to ringtones and horoscopes, and more recently
calling card-l ike services providing access to a block of long-d istance minutes, in
exchange for an amount billed to their telephones as with pre-paid accounts or via
their communications service bill.445 The Act defines a PRS as one that provides the
user with access to content or a facility via an electronic communications service
where the charge paid to the communications service provider for that facility ac-
cessed or that content is included in the use of the service (s 120(8)). The section
120 condition applies to persons who provide the content, exercise editorial con-
trol, make available the facility, package the service, or provide the service over
their network under an agreement with the provider or retain part of the service
charge. It authorizes Ofcom to approve a code of conduct with which such pre-
mium service providers with a section 120 condition must comply (s 120(3)(za)), or,
if none is arrived at, to enter an order regulating the provision and content of such
services, including pricing and charge-sharing arrangements.
This intends continuation of a prior regulatory framework and its industry-
funded regulator, the former Independent Committee for the Supervision of
Standards of Telephone Information Services (ICSTIS). Although for a while this
was rebranded ‘PhonePayPlus’ apparently in an effort to create a higher profile
and eliminate the vagaries about its name, in 2016, this co-regulator with Ofcom
was renamed as the Phone-paid Services Authority (PSA). It regulates premium
rate services as an agent of Ofcom and pursuant to its Fourteenth Code of Practice
(2016) approved by Ofcom pursuant to sections 120 and 121, Communications
Act. This requires prior registration by all network operators and providers of all
non-exempt premium rate services (including various indirect providers) with
annual renewal.446 After investigation of a complaint, PSA can impose sanctions
for failure to comply with the Code that regulates such things as clear, accurate
rate information, truthful and appropriate advertising, unreasonable delays in
service, or service prolongation. The Code provides for an emergency investiga-
tion procedure with an immediate preliminary investigation for an apparent ser-
ious and urgent breach with the possibility to order the withholding of payments
As discussed above, revisions to the General Conditions will eliminate Ofcom’s Code of Practice re-
446
garding requirement and directly impose information publication and provision obligations in the consumer
protection conditions contained in C.2, including those that also govern premium rate services information
provision.
376
and the suspension of/blocking access to the provider service, as provided in the
revised EU consumer protections under the Universal Service and Users’ Rights
Directive. Sanctions for established breaches can include reprimands, imposition
of prior approval requirements, orders to reimburse complainants, fines, orders
limiting access to services with requirements for compliance advice sought, and
bans on named individuals from providing services.447 A database of sanctions ad-
judicated by a tribunal selected from an Adjudication Panel established by the PSA
code is available on the PSA website.448 Where the premium rate services involve
broadcasters, Ofcom may also address breaches under the Broadcast Code.449
The Communications Act sections 94–96 procedures, discussed above, apply to
breaches of section 120 condition (s 123) with a penalty of up to £250,000 possible if
proportionate and appropriate (s 123(2)). Section 124 that retains the ‘serious and
repeated’ wording of the Act governs suspension. Ofcom has additional powers to
promulgate orders necessary to address issues involving premium rate services for
which there is not a code (s 122).
447
For a list of barred service providers, see < https://psauthority.org.uk/for-business/prohibitions-f urther-
sanctions-a nd-suspensions>.
448
Tribunal Service Provider Adjudications (Phone-paid Services Authority), < https://psauthority.org.uk/
for-business/t ribunal-adjudications?date=>.
449
See Notice of Sanction ‘Square 1 Management Ltd., Smile TV’ (22 May 2007; 22:17), Broadcast Bulletin No
114-21/07/0 8, <http://w ww.ofcom.org.uk/t v/obb/prog_cb/obb114/>.
450
Code power charges are discussed above at Section 6.4.4.4.
451
Ofcom, ‘Statement of charging principles’, 8 February 2005, s 2.18. Also see, Ofcom’s Tariff Tables 2017/
2018, 30 March 2017, at s 1.10 (noting that it applies the 2005 Charging Principles).
73
There is some complexity to determining what falls within the categories, espe-
cially in the separation of content services that are excluded and transmissions
involving content layers that remain communications services. However, pro-
viders are to determine their revenue for ‘relevant activities’ and certify this infor-
mation to Ofcom within twenty-eight days of the publication of a general demand
for such information.454 Based on this information, Ofcom calculates the indi-
vidual administrative charge. Charges in excess of £75,000 per annum may be paid
in monthly instalments.455
While Ofcom has committed to lower its costs progressively, its current budget
for 2017–2018 is £121.7 million, an increase of £5.1 million or 2.5 per cent (stated in
‘real terms’) over the 2016–2017 restated budget with a resulting average increase
of administration fees of network and service provision, set out in Annex 1, of 12.9
per cent from those of the prior annual period.456 Ofcom cites continued work on
the implementation of the Digital Communications Review and the requirement
to conduct several market reviews going forward as the reasons for the sector
increase.457
Table 6.2 sets out the schedule of charges by revenue bands in light of the in-
creased budget for relevant activities for the year 2017–2018.
This detailed level of fiscal analysis in light of Ofcom work plans and with
retroactive adjustment suggests full compliance with Article 12, Authorisation
Directive requirements.
DGT Guidelines, ‘The definition of “relevant activity” for the purposes of administrative charging’, 29
453
0 5,000,000 0
5,000,000 10,000,000 5,000,000 5,635
10,000,000 25,000,000 10,000,000 11,270
25,000,000 50,000,000 25,000,000 28,175
50,000,000 75,000,000 50,000,000 56,350
75,000,000 100,000,000 75,000,000 84,525
100,000,000 150,000,000 100,000,000 112,700
150,000,000 200,000,000 150,000,000 169,050
200,000,000 300,000,000 200,000,000 225,400
300,000,000 400,000,000 300,000,000 338,100
400,000,000 500,000,000 400,000,000 450,800
500,000,000 600,000,000 500,000,000 563,500
600,000,000 750,000,000 600,000,000 676,200
750,000,000 1,000,000,000 750,000,000 845,250
For those new to telecommunications law, licensing and authorization might have
seemed merely an administrative exercise. However, as the above analysis has
demonstrated, while licensing and authorization involves the procedural aspects
of filling out the proper forms (if virtual), it is a complex area of telecommunications
law concerned not only with the structure and nature of a particular telecommu-
nications market but also the attainment of social policy objectives. Licensing and
authorization can be used as a tool to implement important national economic
priorities. This is true whether these are the preservation of a monopoly for the
time being in order to, inter alia, permit investors to recoup their expenditures or
continue a revenue stream for the government, to open the markets for equipment,
services, and networks to immediate or gradual competition or to adapt to techno-
logical developments in the market. The EU experience with the latter objectives
also shows that licensing is a tool that requires skill on the part of the regulator as
well as strong and appropriate conditions and sanctions to address market failures
inherent in networked product and services market where the former monopolist
still controls the access network. The proposed reforms to the EU framework that
will govern market entry and market conduct via general authorization are fairly
Ofcom found compliance with the 12-month limitation for two sets of notices to be difficult.
459
None of these considerations touches on Brexit and what changes it may produce, if any.
460
830
minor of themselves, but they reflect continued reliance on this regulatory tool to
hone the application and enforcement of other specific reforms.
This further fine-t uning of the EU licensing regime shows that, despite the fact
that licensing is a regulatory tool with old legal roots, those roots continue to
underlie regulatory foundations of telecommunications markets today, even if not
always readily apparent. However, the law is an organic thing. It grows, evolves,
and adapts as the societal, technological, and economic conditions that produced
it change, with the laws of authorization/l icensing no exception.
The evolution in telecommunications licensing, especially in the UK, shows this
clearly. The earliest providers, after the invention of the telephone, entered the
market and sold their apparatus and services, without the need for any formality.
When telecommunications was deemed to be a service with a public interest, it was
reserved to a single provider either under a licence or by requiring any others to
have a licence. When this monopolist provider could no longer meet the economic
and social needs in an increasingly computerized world that required creative and
competitive communications networks and services, licensing was used as a tool
to pry open markets and control the level of play. Finally, the removal of individual
licence requirements in the UK for everything but access to spectrum brings us al-
most full circle since no licence is needed or justified under the common law free
market principles regarding limitations placed on a person’s economic freedom.
At the same time, licensing law has also evolved in the EU to try to address the
concerns about convergence in technologies and to provide a consistent, harmon-
ized, and technology-neutral framework for any electronic communications net-
work or service. More specifically and recently, its authorizations policies have
sought to meet the market’s demand for new technologies in a way that does not
limit their development or entry but that at the same time seeks to impose ex ante
reasonable conditions for their provision and use. While it is to be hoped that the
proposed reforms will address the EU’s continued and self-identified weaknesses
in the implementation of authorization, particularly for cross-border market de-
velopment, the continued Member State opposition to the enhanced and more
centralized powers possibly essential to achieve the EU’s cross-border and pan-
European digital agenda is understandable but regrettable.
Whatever changes loom in the future, the ‘student’ of EU licensing and author-
ization law for electronic communications providers will likely continue to wit-
ness this dynamic evolution.
381
7.1 INTRODUC TION
1
The author thanks Geoff N Chapman (MA, MSc, PhD) and Renee Greenberg (BA, JD) for their research
assistance.
2
Bachman—Turner Overdrive ‘You Ain’t Seen Nothing Yet’ (Mercury Records 1974).
3
Enter, R, ‘The Wireless Industry: Revisiting Spectrum, The Essential Engine of US Economic Growth’
(Recon Analytics April 2016).
4
See eg Smith, A, ‘U.S. Smartphone Use in 2015’ (Pew Research Center, Pew Hispanic Center, 1 April
2015) (noting that while that growing numbers have access to digital technology in the US with 64% of
Americans adults owning a smartphone, that non-w hites (12% of Black and 13% of Hispanic Americans) have
access to the internet only on their mobile phones, compared to 4% of whites. Also, penetration rates in Africa,
while significantly lower than much of the world, are increasing annually although cost and quality issues
382
remain, see ITU, ‘Percentage of Individuals Using the Internet 2000–2016’, 2017, <https://w ww.itu.int/en/I TU-
D/Statistics/Pages/stat/default.aspx>.
5
See eg White, B, Keysight Blogs: ‘FCC 5G spectrum allocation demands 3 breakthrough innovations’
(Keysight Technologies, 29 July 2016) (noting the FCC’s recent allocation of 11Ghz combined licensed and
unlicensed spectrum for 5G use is 200 x that allocated for the first cellular analogue communications),
<https://c ommunity.keysight.com/c ommunity/k eysight-blogs/i nsights-outside-t he-box/blog/2 016/0 7/2 9/
fcc-5g-spectrum-a llocation-demands-3-breakthrough-i nnovations>.
38
the unprecedented and growing demand for spectrum, including notably for mo-
bile data broadband services, will shortly outpace its availability.6 To address this
problem so that further innovation and the economic and social growth possible7
with mobile technologies are maximized, policymakers and regulators are exam-
ining their current policies, allocations, and uses of allocated spectrum.8 Various
complex strategies are being used or, at least, considered to promote more effi-
cient and effective use of spectrum and further its potential availability, including
the 2016–2017 US ‘voluntary incentive auction’ that required several stages of a
reverse auction and then a forward auction after the FCC’s ‘repacking’ of spectrum
returned by broadcasters for its reallocation and licensing in contiguous blocks.9
This US effort, proposed in its 2010 National Broadband Plan, is a regulatory op-
tion available for currently underused but allocated spectrum: its repurposing,
or ‘refarming’ to more valuable uses or to the same use but using more efficient
technology that requires less bandwidth or has less risk of adjacent bandwidth
bleed. Refarming has been of considerable recent significance with the realloca-
tion in various jurisdictions of what is called the ‘digital dividend’, or, the spec-
trum used by bandwidth-hogging TV analogue broadcasts, including in the 700
and 800 MHz bands10 discontinued with the switchover to digital television that
uses digital compression technologies permitting at least four channels to op-
erate in the same spectrum bandwidth as one analogue channel. These more ef-
ficient digital TV services take up less bandwidth freeing the remainder for new
uses. However, exploiting this ‘digital dividend’ has not been problem free. In the
UK, Ofcom estimated that as many as 2.3 million households would be affected
6
See eg FCC Staff Technical Paper, ‘Mobile Broadband: The Benefits of Additional Spectrum’, October
2010 (noting that demand for mobile data is expected to grow between 25 and 50 times current levels within
five years in light of take up of smart devices, producing a spectrum availability deficit of at least 300 MHz).
Also see Bazelon, C and McHenry, G, ‘Substantial Licensed Spectrum Deficit (2015–2019): Updating the FCC’s
Mobile Data Demand Projections’ (Brattle Group 23 June 2015) (estimating the deficit as of 2019 at 366 MHz
with two thirds unmet by interim re/a llocations of licensed spectrum as of 2016), <http://fi les.brattle.com/
files/5 927_ s ubstantial_ l icensed_ s pectrum_ d eficit_(2015-2 019)_-_u pdating_t he_ f cc’s_ mobile_ d ata_ d e-
mand_projections.pdf>.
7
One estimate is that in 2035, 5G value chain will globally have USD 3.5 trillion in output and support
22 million jobs. ‘How 5G Technology Will Contribute to the Global Economy’ (Communications Today,
April 2017) (based on Qualcomm Report), <http://w ww.communicationstoday.co.in/reports/10255-how-5g-
technology-w ill-contribute-to-t he-g lobal-economy>.
8
See FCC Staff Technical Paper, n 6. Also, see Decision 243/2012/E U of the European Parliament and of
the Council establishing a multi-a nnual radio spectrum policy programme (RSPP Decision), OJ L 81/7, 14
March 2012; Decision 676/2002/EC of the European Parliament and of the Council on a regulatory framework
for radio spectrum policy in the European Community (Radio Spectrum Decision), OJ L 108/1, 7 March 2002.
9
Hazlett, TW, ‘FCC “Incentive Auction” marks progress and pitfalls toward freeing wireless spectrum’
(The Brookings Institute, 24 May 2017).
10
In Europe, the bandwidth freed comprised 470–862 MHz or the UHF band which has long wave coverage
with building penetrating capabilities suitable for mobile networks.
834
by the new LTE 4G signals in the 800 MHz band of which 900,000 UK households
with only Freeview likely to lose all or part of this Digital Terrestrial Television
(DTT) service.11 Although this estimate has likely proved well beyond actual ex-
perience to date, the possible problems could not be ignored.12 The government
determined that the spectrum winners establish a private entity, Digital Mobile
Spectrum Limited, (pursuant to an imposed licence condition and key perform-
ance indicators) with a budget of up £180 million (added to the spectrum licence
fees) to manage these problems. It has an oversight board of mobile operators
and broadcasters, several independent board members as well as Ofcom and the
Ministry of Digital, Culture, Media & Sport. This entity, aka ‘at800’, appears to be
addressing most problems by sending and/or installing filters for the interfering
bandwidths,13 although not all homes are projected to be able to use this fix with
some possibly requiring a new platform for which a payment of up to £10,000 was
authorized.14 The company is also now beginning to address issues arising from
the UK clearance of DTT from the 700 MHz band for mobile data use that Ofcom
has indicated it is accelerating by eighteen months to 2020.15
The need to free harmonized bands of spectrum for wireless broadband also
drove the repurposing of spectrum bands to allow 3G and then 4G in bands origin-
ally allocated for 2G, the earlier generation of digital communications, comprising
eg in the EU, the 900/1800 MHz bands.16 This involves technical co-existence in
parts of the bands over which 2G services are still provided (and possibly will
11
See Ofcom, Second Consultation on coexistence of new services in the 800 MHz band with digital terres-
trial television (23 February 2012), <http://stakeholders.ofcom.org.uk/binaries/consultations/949731/sum-
mary/condoc.pdf>.
12
Although the ‘actual’ numbers do not reflect the nearly 1 million filters sent out in advance proactively
by at800 and not in response to a complaint. See Letter from Chair 4G/T V Coexistence Oversight Board to
Ofcom proposing revised scheme and trial (Dept for Culture, Media & Sport, 18 December 2013) (requesting
enforcement forbearance for a pilot to explore varying the licence conditions and KPIs of 4G spectrum win-
ners in their operation of at800 to permit greater flexibility in light of experience to date and much lower actual
numbers of involved households with a view to permanent variance, if successful).
13
See eg, Matthews, C, ‘New 4G masts in Camborne could cause interference for Freeview users’ (Cornwall.
live, 7 March 2018), <https://w ww.cornwalllive.com/news/cornwall-news/new-4g-masts-c amborne-could-
1308423>. Not all households have been fully compensated or their problems proactively addressed by
at800 as new 4G masts become operational. See Corr, S, ‘Cookstown residents out of pocket as 4G signal kills
Freeview TV’ (Mid-U lster Mail, 18 December 2015) (noting that couple in their 80s only offered £50 reimburse-
ment of £105 spend to fix interference problem, notification of which they did not receive), <https://w ww.
midulstermail.co.uk/news/cookstown-residents-out-of-pocket-a s-4g-signal-k ills-f reeview-t v-1-7121102>.
14
Department for Culture, Media and Sport, News Release, ‘Eliminating Interference with TV signals from
4G’, 2 April 2012, <http://w ww.culture.gov.uk/news/media_ releases/8 865.aspx>.
15
Ofcom, ‘Statement: Maximising the benefits of 700 MHz clearance, Enabling acceleration of 700 MHz
clearance and use of the 700 MHz centre gap’, 10 October 2016.
16
See eg Commission Decision 2011/251/E U amending Decision 2009/766/EC on the harmonization of the
900 MHz and 1800 MHz frequency bands for terrestrial systems capable of providing pan-European elec-
tronic communications services in the Community, [2011] OJL 106/9, 18 April 2011.
358
continue to be as more advanced mobile handsets still have this capability and
these bands allow for other uses such as existing machine-to-machine systems
that would require costly upgrades).17 Mitigation of adverse effects has been ad-
dressed via mandatory channels where technically necessary18 as well as cooper-
ation among providers. Regulatory policy in refarming also often needs to address
compensatory issues where the refarming constitutes a taking by government or
where existing equipment is no longer usable.19 The US voluntary incentive auction
was an innovative way to do this although it did not achieve the targeted 120 MHz
of returned spectrum, with only 70 MHz procured for reallocation.20
Concerns about competitive disadvantage may arise from refarming. For ex-
ample with 2G spectrum, these could include that it was obtained under dif-
ferent allocation processes and at lower costs than that incurred for 3G and most
recently, 4G. When the EU directed that 2G bands be liberalized,21 rather than
reauction them, the UK government sought to address these concerns by directing
that Ofcom allow current spectrum incumbents to retain the 2G bands of 900/1800
MHz they were allocated in the 1980s but pay current full market value for their
liberalized use.22 Ofcom set the market value prices using benchmarks such as the
2013 UK 4G spectrum auction results and other markets’ 2G refarming auction
prices but converted to annual fees (as these 2G were never auctioned) and further
adjusted, by taking into account the delay in the spectrum’s availability. Ofcom
ultimately raised the fees by nearly quadruple their original amount. The incum-
bents appealed. The Court of Appeal struck down Ofcom’s pricing decision that
was originally upheld by the High Court as will be discussed further.23
Other competitive concerns arise since 2G spectrum was likely to be held by
the original mobile market entrants, often a duopoly. Allowing them to change
17
Chambers, D, ‘The mobile industry focuses on spectrum refarming through 2017’ (ThinkSmallCell, 2
February 2017), <https://w ww.thinksmallcell.com/Technology/t he-mobile-i ndustry-focuses-on-spectrum-
refarming-t hroughout-2017.html>.
18
See eg Commission Decision 2009/766/EC on the harmonisation of the 900 MHz and 1800 MHz fre-
quency bands for terrestrial systems capable of providing pan-European electronic communications services
in the Community (in allowing for the use of these 2G bands spectrum by UMTS terrestrial systems, requiring
separation channels for neighbouring UMTS systems of 5MHz and neighbouring GMS and UMTs systems of
at least 2.8 MHz), OJ L 274/32, 16 October 2009.
19
See Ofcom Statement, ‘Maximising the benefits of 700 MHz clearance, Enabling acceleration of 700 MHz
clearance and use of the 700 MHz centre gap’, 10 October 2016 (noting that the UK government will fund a
grant in 2019 for programme-making and special events (PMSE) users whose equipment will no longer be op-
erational when the 700 MHz band is cleared before the expected date).
20
See Hazlett, n 9.
21
Decision 243/2012/E U of the European Parliament and of the Council establishing a multiannual radio
spectrum policy programme, [2012] OJ L 81/7, 21 March 2012.
22
The Wireless Telegraphy Act 2006 (Directions to OFCOM) Order 2010, SI 2010/3024.
23
Decision 243/2012/E U, n 21.
386
their licences to include its use for 4G could provide their earlier ability to move
to the new technology if liberalized before other 4G spectrum is auctioned or at a
lower cost than other 4G spectrum. Such time and cost advantages could serve to
entrench the market position of these earliest mobile market entrants. A UK tri-
bunal interpretation of the EU ‘Refarming Directive’,24 however, found that the
original 2G GSM spectrum grantee had no directly enforceable right to the auto-
matic revocation of any restrictions in its 2G licence to enable 3G use with only
ex post competition law to address any anti-competitive effects, a compulsory EU
regulatory objective.25 Rather, the Directive required a technical harmonization
to be effected by making the bands ‘available’ for 3G use (although unnecessary in
the UK) with specific authorization decisions to follow after the mandatory frame-
work consultation and taking into account on an ex ante basis any distortions in
competition.26
Another key issue with refarming is that significant amounts of spectrum is al-
located for public sector purposes, eg in the EU noted to comprise up to 40–50 per
cent of usable frequencies below 15 GHz.27 For some uses this may not only need to
be repurposed at the national level but also require regional and international har-
monization for effective cross-border use such as for mobile networks.28 In the UK,
for example, since 2010 various public agencies have identified bands that could
be freed up or shared with new uses under the Public Sector Spectrum Release
Programme’s goal to free up 500 MHz of bandwidth from public uses. According
to the government, this has been 80 per cent achieved. The Ministry of Defence, for
example, released 200 MHz of bandwidth in the 2.3 GHz–2.4 GHz and 3.4 GHz–3.6
GHz bands that was to be made available by 2016 for public mobile uses including
likely 5G in the 3.4GHz bands. This spectrum auction, however, was delayed until
April 2018 by legal challenges to Ofcom’s proposed caps on the amount of spec-
trum that any one holder might control. These were to address competition con-
cerns in light of BT’s acquisition of EE that gave it the largest share of available
spectrum at 45 per cent, although not market share as BT had not provided mo-
bile services since its 2006 sale of O2 to Telefónica.29 As noted, the US similarly
24
Directive 2009/114/EC amending Directive 87/372/E EC on the frequency bands to be reserved for the
coordinated introduction of public pan-European cellular digital land-based mobile communications in the
Community.
25
See Telefonica O2 Ltd v Ofcom [2010] CAT 25. 26
Ibid, at 85, 90–102.
27
Commission Discussion Paper EU Spectrum Summit, at 6 (22–2 3 March 2010).
28
See eg UK Dept for Culture, Media and Sport, ‘Enabling UK Growth: Releasing Public Spectrum Update
on Progress’, December 2011 (noting mobile network operators’ preference for spectrum harmonized at the
international level over spectrum that was not).
29
Torrance, J, ‘Ofcom accuses Three of holding up spectrum auction after failed legal bid’ (The Telegraph,
20 December 2017), <http://w ww.telegraph.co.uk/business/2017/12/20/ofcom-accuses-t hree-holding-spec-
trum-auction-failed-legal-bid/>.
387
committed to repurposing at least 500 MHz of bandwidth from both federal gov-
ernment agencies and private entities for mobile and fixed broadband uses. As of
2015 over half had been made available.30 The US auctions addressed competition
concerns via ‘set asides’ for entities with sparse holdings of spectrum below 1 GHz,
overall caps of the amount to be auctioned and a cap for a single entity serving
sparsely populated areas.31 The EU in its newly adopted Multi-A nnual Spectrum
Policy Programme clearly contemplates a harmonized approach across the EU to
repurposing public sector spectrum.32
A regulatory option for making spectrum more broadly available is the deregu-
lation of spectrum frameworks from existing ‘command and control’ models
that dictate use according to the technical characteristics of the spectrum, ex-
plored briefly in Section 7.2, to allow a market-d riven determination of best use
and continued innovation and as well, economically efficient pricing, and es-
sentially service and technological neutrality. The concern here is how to en-
sure non-i nterference and safe use that the designated allocation was intended
to control in the first place.
Finally, spectrum licensing liberalization may enable those with grants of spec-
trum to participate in secondary spectrum markets enabling under-utilized spec-
trum to be shared or used exclusively by others but more efficiently.
A full exploration of all these issues is beyond the scope of this chapter which
intends to provide an overview of spectrum used for communications and its
regulation, historically and today, so that the reader will appreciate the differences
from telecommunications licensing generally addressed in the prior chapter. To
do this, it first considers at a fairly basic level the nature and characteristics of
spectrum used in communications with inherent relevance to spectrum regu-
lation. It then provides an overview of historical spectrum regulation including
at the international level and national level generally for a sense of how and why
the frameworks are what they are. It will then examine the EU framework which
seeks to harmonize Member State approaches to spectrum regulation, generally
within national competence, in order to foster the continued economic and social
development of the Single Market. National policy priorities could undermine this
given that spectrum interference does not respect national boundaries and mobile
communications are a significant growth sector and emergent platform for new
30
Priztger, P and Strickling, L, ‘Six Interim Progress Report on the Ten-Year Plan and Timetable’ (Dept of
Commerce, June 2016). Also see, FCC Staff Technical Paper, n 6, at 2.
31
Meyer, D, ‘FCC sets aside 30 megahertz for auction set to begin March 29, 2016’ (RCR Wireless News, 6
August 2015), <https://w ww.rcrwireless.com/20150806/policy/fcc-releases-600-m hz-auction-r ules-stomps-t-
mobile-i ncreased-reserve-request-t ag2>.
32
Decision 243/2012/E U of the European Parliament and of the Council establishing a multiannual radio
spectrum policy programme, [2012] OJ L81/7, 21 March 2012.
83
commercial and government services. The chapter will also examine some recent
reforms by the EU to drive this harmonization. It will finally examine how the UK
has implemented the EU framework, an overlay on its own legislative scheme that
has been in place since the mid-1800s.
33
Electromagnetic spectrum can also be measured according to the length of the wave and its energy with
the three elements having a mathematical relationship, so that low frequency waves are long with low en-
ergy or power and high frequency waves are very short and have high energy. See Imagine, ‘Electromagnetic
Spectrum: Introduction’ (NASA GSFC, February 2010), <http://i magine.gsfc.nasa.gov/docs/science/k now_ l1/
emspectrum.html>.
34
Usually measured from the highest point of one wave to the same point in the next, although a standard
without a technical difference as the measure of repetition will be the same no matter what point is used.
35
See Lapthorn, R, ‘Sub-9kHz Amateur Radio’ (March 2011) (noting sub-9 kHz range radio communica-
tions an impossibility as little as five years ago), <https://sites.google.com/site/sub9khz/>.
36
See ibid.
37
See eg Rysavy, P, ‘Low Versus High Radio Spectrum’ (High Tech Forum, 5 March 2012), < http://
hightechforum.org/low-versus-h igh-radio-spectrum/>.
38
There is also correspondence between these and a wave’s height, called ‘amplitude’. See Imagine,
‘Electromagnetic Spectrum: Introduction’ (NASA GSFC, February 2010), <http://i magine.gsfc.nasa.gov/docs/
science/k now_ l1/emspectrum.html>.
39
The power of the transmitting device is also relevant to distance that radio waves can travel.
389
geographic areas and can carry more data.40 As basically illustrated in Table 7.1,
different frequencies’ propagation characteristics are relevant to different com-
munication uses and technologies. While waves at the same frequency in the
Rysavy, n 37.
40
930
same space can interfere with each other, waves at different frequencies do not.
Filters and smart antennas can eliminate those that are not wanted, leaving the
receiver to decode the information carried as energy without mass on the wave.
Radio waves comprising photon energy travel at the speed of light and can often
travel through non-conductive materials without being absorbed,41 also relevant
to various wireless communications uses.
That different frequency characteristics make certain uses more likely, combined
with the fact that radio waves don’t stop at international borders and the possi-
bility of interference with simultaneous but competing uses at the same frequency
in the same geographical location led to national, international, and multilat-
eral cooperation in spectrum policy and allocation management, notably via a
framework established within the International Telecommunication Union (ITU),
perhaps the oldest intergovernmental body. Coordination of frequency has also
had the benefit of allowing the manufacture of equipment that can operate cross-
borders and, optimally, internationally, with ensuing economies of scope.
International spectrum ‘regulation’ also comes from the World Trade Organiza
tion to the extent that the General Agreement on Trade in Services (GATS) and
the Reference Paper impose transparency and other obligations regarding spec-
trum licensing, as well as from international standards bodies like the Institute
of Electrical and Electronics Engineers (IEEE) that work to develop technical spe-
cifications for radio equipment. One example is the 802.11 wireless local access
network family of standards that permit multiple non-interfering uses within
ITU bands, requiring limited regulatory oversight.42 Although important, a dis-
cussion of these sources of spectrum regulation is also beyond the scope of this
chapter.43
41
This, as noted, is a very basic analysis. There are many things that affect the propagation characteristics of
radio waves at various frequencies, including weather, the earth’s curvature and topography, the ionosphere,
line of sight reception, solar flares, absorbing (eg water) and reflective (the ground) or diffractive (roof edges)
surfaces. These of course depend on the frequency/w ave length involved. For a good overview of wave propa-
gation, see Toronto Emergency Communications Group, ‘Basic Amateur Emergency Radio Course: Module
7: Radio Wave Propagation’, October 2010, <http://w ww.emergencyradio.ca/course/>.
42
The WiFi 802.11 is a grouping of IEEE standards for low-powered high frequency communications sys-
tems intended for non-i nterfering use within the International Telecommunication Union (ITU) Industrial,
Scientific, Medical bands designated by ITU-R in 5.138, 5.150, and 5.280, Radio Regulations. Use of WiFi is
often unlicensed or exempt or subject only to interference tolerance requirements. See generally, eg Negus,
K and Petrick, T, ‘History of Wireless Local Area Networks (WLANS) in the Unlicensed Bands’, (2009) 11(5)
Journal of Policy, Regulation and Strategy for Telecommunications, Information and Media 36.
43
But see Chapter 16, at Section 16.4.
391
Schmahl, S, ‘The United Nations: Facing the Challenges of the “Information Society” ’, (2007) 11 Max
45
Wireless telegraphy, like many other technologies, evolved from a series of theor-
etical and practical advances over time. Marconi is largely credited with ‘inventing’
wireless telegraphy in the late 1890s. While the extraordinary achievement of his
experimentation to apply theory and produce workable equipment as well as his
efforts to commercialize wireless technology cannot be minimized, Marconi was
standing on the shoulders of giants, as the expression goes. These include, among
others, James Clerk Maxwell, a Scottish physicist and mathematician, who in 1865
formalized a cohesive electromagnetic wave propagation theory underpinned by
his mathematical equations and which organized prior disparate theories, ex-
periments, and practical observations about light, electricity, and magnetism.48
Heinrich Hertz proved Maxwell’s electromagnetic wave theory by constructing an
apparatus detecting their presence. In 1895, inspired by Hertz’s discovery, Marconi
began experimenting with Hertzian waves, attempting to transmit signs and sym-
bols without connecting wires.49 In 1895, he successfully transmitted a signal a
distance of 2km.50 As Italy lacked interest in his experiments, Marconi moved to
England at the Post Office’s invitation51 where he continued his experiments. In
1896 he was granted a patent in his invention and started the Marconi Wireless
Telegraphy Co, to develop his worldwide patent monopoly and commercialize the
invention fully via equipment manufacture and the construction of wireless tel-
egraphy networks.52 His continued experimentation progressively extended the
distance over which the signals were conveyed until 1901 when he successfully
transmitted a message from southwestern England to Newfoundland, a distance
of 2100 miles53 and in early 1902, between a wireless telegraph station in Cornwall,
England and the SS Philadelphia, an American ship.54
Marconi realized that transmission of readable messages over long distances had
major implications for maritime safety and naval operations, having conducted
a range of experiments with the British Navy.55 In 1898, the Marconi Wireless
48
They also include Michael Faraday on which Maxwell’s work expanded. See Biography: Michael Faraday,
Institution of Engineering and Technology, <http://w ww.theiet.org/resources/l ibrary/a rchives/biographies/
faraday.cfm>.
49
Marconi, G, ‘Wireless Telegraphic Communication: Nobel Lecture 1909’, <http://w ww.nobelprize.org/
nobel_prizes/physics/laureates/1909/marconi-lecture.pdf.>.
50
Braga, GM, ‘Marconi Family History’ (Marconi Family Society), <http://marconisociety.org/about/mar-
coni-family/>.
51
Events in Telecommunications History, BT Archives, <http://w ww.btplc.com/Thegroup/BTsHistory/
1881to1911/1896.htm>.
52
Marconi, n 49. 53
Ibid.
54
According to Marconi who was onboard the SS Philadelphia, ‘readable messages were received by means
of a recording instrument up to a distance of 1,551 miles and test letters as far as 2,099 miles from Poldhu.’
Marconi, n 49.
55
For example, in 1899, on two separate occasions when the East Goodwin Sands Lightship encountered
problems at sea, lives were saved because the vessel had been equipped with radio installation and could send
distress messages, allowing assistance to be quickly dispatched, see Howeth, LS, (Capt., USN (Retired), ‘Birth
39
of Science of Radio and Development of Usable Components’ History of Communications-E lectronics in the
United States Navy (Bureau of Ships and Office of Naval History, 1963), Section 9: ‘First Uses of Radio as an Aid
to Safety of Life at Sea’, <http://earlyradiohistory.us/1963hw02.htm#2footnote>.
Ibid, Section 10.
56 57
Ibid. 58
Ibid. 59
Ibid.
Radiotelegraph and Radiocommunications Conferences, ‘Preliminary Conference on Wireless Telegraphy
60
64
Ibid.
65
ITU Library and Archives, ‘Art V, Final Protocol –Preliminary Conference on Wireless Telegraphy (Berlin
1903)’, <https://www.itu.int/en/history/Pages/RadioConferences.aspx?conf=4.35>.
66
Conference Documents, Procés-Verbaux, Conférence Internationale Concernant La Télégraphie Sans Fil
(German Department of Post, Empire, 3 October 1906 Berlin), 39–43, ITU Radiocommunication Conferences,
<http://www.handle.itu.int/11.1004/020.1000/4.36>.
67
See Radiocommunications Sector, ‘100 Years of ITU Radio Regulations (1906–2006)’, <https://www.itu.
int/en/history/Pages/100YearsITURadioRegulations.aspx>.
68
These have since been expanded and revised by numerous radio conferences, and are now known as the
Radio Regulations. They are part of the Administrative Regulations and the legal basic framework of the ITU
with treaty status. During the 1906 Convention, the protocol was only twelve pages long. The Radio Regulations,
now 122 years old, generally harmonizing how frequency spectrum may be used and shared among various
services, comprise over 2300 pages. See ITU History Portal: Radio Regulations—A n Introduction (ITU 2008),
<http://itu150.org/h istorical-t imeline/>. See also Chapter 16, at Section 16.3.
69
Timofeev, V, ‘How ITU processes and regulations have helped shape the modern world of
radiocommunications’, (ITU News Magazine, 2006) 5–9, <http://search.itu.int/h istory/H istoryDigitalCollec
tionDocLibrary/12.26.71.en.pdf>.
70
Ibid. 71
Ibid.
72
See eg McPhail, TL, ‘The Medium: Global Technologies and Organizations’, in Global Communications:
Theories, Stakeholders, and Trends (Wiley-Blackwell, 2011), 270–271.
395
Subsequent interfering uses were prohibited with the result that ‘first-come, first
served’ became the operational norm with most spectrum allocated to certain
North American and European nations.73 The issue was not divisive when the pri-
mary allocations were for maritime activity but with the emergence of commercial
broadcast and other public radio this became a growing problem. In 1929, there
was agreement for further formal coordination by allocating bands or groups of
bands for specific services.
In 1932, the International Telegraph Union and the International Radiotelegraph
Conference merged into the International Telecommunication Union. Its workings
continue to this day via ITU-R.74 Other technical standards were added. A table
of tolerances and another giving the acceptable bandwidths for various types of
emissions were added into the regulations75 as guides for national administrations
to measure the technical conformity/efficiency of radio stations and to focus their
attention on the need to develop effective controls on transmitting stations.76 To
combat harmful interference and ensure that countries followed the harmonized
allocated radio bands that evolved over time with new discoveries such as short
waves, registration requirements were strengthened. These require countries to
inform the ITU prior to using a new frequency and/or changing the power of a fre-
quency already in use.77
In 1947, the ITU became a specialist agency of the United Nations. At that year’s
International Radio Conference, the International Frequency Registration Board
was created. Its role was to formalize, administer, and oversee a master frequency
register to track notifications and usage and ensure the compliance of a new fre-
quency use registered with the requirements of the Radio Regulations. This work
in connection with international coordination of spectrum, allocation, and inter-
ference control is accomplished via the ITU-R. Its Recommendations address
technological developments that may, inter alia, enable new uses, enhance cap-
abilities, and address the desirability of old uses.78 These can be adopted as binding
Radio Regulations at ITU Radiocommunication Conferences. Starting in 1947, the
ITU divided the world into three regions for the coordination and registration as
well as for the preparatory work for its Radio Conferences and regulation promul-
gation. This allows for regional variation where international harmonization is not
73
Ibid. 74
See Chapter 16, at Section 16.3. 75
Ibid.
‘50th Anniversary of the Madrid Conferences’, (1982) 49(9) Telecommunication Journal 510–511, <https://
76
itu.tind.io/record/13682?ln=en>.
77
Ibid.
78
It is noted that sometimes allocations are determined purely on technological and not political consid-
erations. For example, spark gap wireless technology used in ship transmitters was simple to use but wasteful
of spectrum and with considerable signal interference. It was just phased out of usage over time. See Krasner,
SD, ‘Global Communications and National Power: Life on the Pareto Frontier’, (1991) 43(3) World Politics 351.
936
essential or based only on technical merits and permits a structure to help con-
sensus building, not an easy process with so many different interests.79 The div-
ision comprises: Zone 1, Europe, Africa, the Middle East; Zone 2, the Americas;
and Zone 3, Australia, China, Japan, and other Asian countries.
The ITU international coordination process of allocating bands and promul-
gating harmonizing recommendations and regulations for efficient and non-
interfering uses of radio spectrum continues to this day. The process has not
been without criticisms, such as the ITU’s failure to intervene until congestion
and interference had already occurred, and to allocate only according to present
needs and technological capability to the detriment of developing nations.80 The
continuity of purpose and process of this oldest claimed intergovernmental or-
ganization, however, remains notable. For example, the most recent World Radio
Communications Conference took place in November 2015 and the next will be in
2019. On its agenda is regulatory support of Global Marine Distress Safety Systems’
modernization, including evaluation of how adding other satellite systems, eg mo-
bile satellite systems, can be supported as per Resolution 359 (rev.WRC-15), alloca-
tions and how possible modifications of the Regulations impact compatibility and
sharing with other services,81 an issue blending the regulatory and technical con-
cerns of efficient spectrum management with systems begun over 100 years ago.
79
Ibid, considering an economic analysis of pareto outcomes to the international spectrum allocation
framework which, post-1971, is noted to be different from that based solely on equal need for coordination
that existed previously.
80
See McPhail, n 72. These issues are considered further in Chapter 16.
81
See WRC 2019 Agenda Item Details, Agenda Item 1.8, (Transfinite Systems), <https://w ww.transfinite.
com/content/w rc2019list>.
82
As has been noted, ‘The ITU is really a gentlemen’s club. It depends on the goodwill of its members. There
is no mechanism for forcing an administration into compliance with the rules’, de Selding, PB, ‘France seeks
ITU help to Halt Satellite Signal Jamming by Iran’ (Space News, 1 August 2010) (quoting Francois Rancy, dir-
ector of France National Frequencies Agency), <http://spacenews.com/f rance-seeks-itu-help-halt-satellite-
signal-jamming-i ran/>.
83
See ERC Report 25, ‘European Table of Frequency Allocations and Applications in the Frequency Range
8.3 kHz to 3000 GHz (ECA Table)’ (CEPT 2017) (noting its 2002 principle of adopting a ‘harmonised European
Table of Frequency Allocations and Applications to establish a strategic framework for the utilisation of the
radio spectrum in Europe’), <http://w ww.erodocdb.dk/docs/doc98/official/pdf/ercrep025.pdf>.
397
These uses would be reflected in a national allocation table identifying the services
for which specific allocated bands could be used and which likely encompassed
the international allocation obligations. Allocations and controls have until fairly
recently largely been effected via command and control administrative processes
using the tool of licensing as a means to specify usage rights, including term, ser-
vice, geographic area, configuration, apparatus, etc,86 and conditions for that use
that could include payment of an annual fee, power limitations, and requirements
for conformity to what in the EU are called ‘essential requirements’ for such things
as electromagnetic compatibility and efficient spectrum use so as to prevent
interfering operation of relevant radio spectrum.87
With the competing commercial demands for its use, nations began to seek
ways to allocate fairly the spectrum among the many applicants. The US first used
a comparative process to assign licences for various cellular wireless services to
84
This is the case with intellectual property rights, see Case 16-74, Centrafarm v Winthrop [1974] ECR 1183;
Case 15-74, Centrafarm v Sterling Drug [1974] ECR 1147.
85
Ofcom, ‘Spectrum Framework Review’, 2004.
86
See eg Australian Cordless Class Licence 2014, <https://w ww.legislation.gov.au/Details/F 2014L01800>.
87
See recitals 4–8, Directive 2014/53/E U on the harmonisation of the laws of the Member States relating to
the making available on the market of radio equipment and repealing Directive 1999/5/EC, [2014] OJL 153/
62, 22 May 2014.
938
which it had allocated frequency. It set what it considered were appropriate stand-
ards for awarding the available, free licences according to the FCC’s statutory
charge of ‘public interest, convenience and necessity’.88 The licences were awarded
to the candidate considered most qualified after hearings, with decisions often ap-
pealed for years in the US courts. The comparative processes, labelled ‘beauty con-
tests’, were time and resource consuming for both applicants and the FCC. They
were criticized as too slow, costly, and serving as an impediment to new service
entry.89 Also, questions raised about comparative processes concerned the ob-
jectivity of their selection criteria and their transparency. The US then employed
lotteries to address these concerns. Pre-lottery screening, however, to ensure that
only qualified applicants participated in the lottery was similarly time and re-
source intensive with screening for the first lottery lasting twenty months with the
same concerns about the pre-selection criteria.90 Open lotteries followed. These,
however, introduced speculation and ensuing windfall profits for applicants who
had won licences with no intention of providing service to the public and who
quickly traded them on then emergent secondary markets. The additional time
inherent in a second transaction and the process to reassign the licence as well as
the necessity to aggregate licences necessary to use the spectrum efficiently were
also suboptimal and delayed service roll-out.91
The US in 1993 authorized a market mechanism for spectrum allocation, com-
petitive bidding. This decision was premised on modern economic theory that ef-
ficient use of scarce resources such as spectrum required allocation other than
via traditional command and control since these mechanisms would ensure that
the spectrum went to those who valued it the most and would ensure its most pro-
ductive use.92 While these can be structured in different ways, the US, the UK, and
many other countries now employ a form of auction where price alone, typically,
dictates outcome. Other criteria or limitations can apply for participation to fur-
ther social or competition policy, such as set asides to permit new entrants or spec-
trum caps to ensure relative competitive holdings of spectrum.93 While pure price
auctions may be more objective than comparative processes, they also have flaws.
88
47 USC §309. See further Chapter 5.
89
See Goodman, E, McCoy, S, and Kumar, D, ‘An Overview of Problems and Prospects in US Spectrum
Management’, Telecommunications Convergence: Implications for the Industry & for the Practicing Lawyer, 698
PLI/Pat 327 (Practising Law Institute, New York, 1 May 2002).
90
See ‘In the Matter of FCC Report to Congress on Spectrum Auctions’, FCC 97-353 (Federal Communications
Commission Washington DC, 30 September 1997), 7.
91
Ibid.
92
Brito, J, ‘The Spectrum Commons in Theory and Practice’, (2007) Stan Tech L Rev 1, paras 5–7, <http://
jerrybrito.com/pdf/2007StanTechLRev1.pdf>.
93
See regarding recent US caps/set asides, text accompanying n 31; regarding the recent UK applied overall
and bandwidth spectrum caps for 5G auctions, see text accompanying n 29.
93
For example, concerns were raised about the massive overbidding for 3G spec-
trum94 and the high debt levels the successful undertakings incurred long before
they could realize the value of the spectrum.95 It has also been shown that auctions
may not necessarily be efficient. This can occur, eg where overbidding, whether
collusive or not, is intended to foreclose the market to new entrants,96 or where
intimidatory collusion occurs via threats of retaliatory bids, or where bidders co-
ordinate and reduce their demand to lower prices.97 The emphasis on maximizing
government revenues from spectrum auctions has also been called into question
for failing to ensure that policy objectives underlying auctions including efficient
and most productive use are, in fact, occurring.98 It is further criticized for failing
to allow for spectrum use for broader social objectives that would be undertaken
ordinarily by organizations unlikely able to compete in a price-based auction.99
Another criticism of pure price auctions is that they can inhibit the roll-out of in-
novation beyond existing technology, eg the long-term licences viewed as neces-
sary to recoup costs and justify infrastructure investment.100
Not all market economies, however, have exclusively used a highest bidder approach. For example, Finland,
allocates certain blocks of spectrum for development, research, and teaching in geographical areas, See
FICORA, Regulation 2500–2690 MHz Auction (Helsinki, 2009), s 7. The US has unlicensed bands available
for use as a ‘commons’, subject to Part 15 device rules that require they operate on the principle that interfer-
ence must be tolerated. The UK as well has allocated blocks of spectrum for low-powered unlicensed use as
a commons. Also recently the UK applied overall and bandwidth spectrum caps for the 5G auctions. See text
accompanying n 31.
94
See Ozanich, G, Hsu, C, and Park, H, ‘3G Wireless Networks as an Economic Barrier to Entry: The Western
Experience’, (2004) 21(3) Telematics and Informatics 225.
95
See generally, Rose, G, ‘Spectrum Auction Breakdown: How Incumbents Manipulate FCC Auction
Rules to Block Broadband Competition’, Working Paper 18 (New America Foundation 2007), <http://w ww.
newamerica.net/fi les/ WorkingPaper18_ FCCAuctionRules_ Rose_ F INAL.pdf>.
96
Ibid. Verizon Wireless (perhaps the largest US wireless carrier) sold the ‘A’ and ‘B’ Block spectrum it
bought at auction in 2008 but never used in exchange for FCC permission to buy licences from a former cable
company wireless venture for spectrum that would better enable its 4G national network. There was consid-
erable industry criticism of Verizon’s alleged ‘warehousing’ of spectrum to keep it out of competitors’ access
and, although the sale/exchange was finally approved, it was subject to regulatory scrutiny. See FCC, Letter to
Verizon Wireless (Washington DC, 15 May 2012), <http://t ransition.fcc.gov/Daily_Releases/Daily_Business/
2012/db0515/DOC-314071A1.pdf>.
97
See generally, Bajari, P and Fox, JT, ‘Measuring the Efficiency of an FCC Spectrum Auction’, NBER
Working Paper No. 11671 (2005, rev 2009) (and other works cited therein, 1), <http://fox.web.rice.edu/pub-
lished-papers/fox-a nd-bajari-aej-m icro.pdf>.
98
See generally, Hazlett, TW and Muñoz, R E, ‘What Really Matters in Spectrum Allocation Design’,
(George Mason Law & Economics Paper No.11-4 8 27, October 2011), <https://papers.ssrn.com/sol3/papers.
cfm?abstract_ id=1961225>.
99
Industry groups have suggested, however, that restrictions/set asides for other uses have only ultim-
ately produced delays, enhanced costs, lower state revenues, and proved to the ultimate detriment of con-
sumers over unrestricted pure price auctions. See Position Paper: ‘The Case for Inclusive Spectrum Auction
Rules: How Failed International Experiments with Auction Bidding Restrictions Reveal the Strength of
Inclusive Rules that Put Consumers and Innovation First’ (MobileFuture.org, September 2013).
100
See Milgrom, P, et al, Working Paper 17-028 ‘Redesigning Spectrum Licences to Encourage Innovation
and Investment’ (Stanford Institute for Economic Policy Research, October 2017) (suggesting perpetual but
04
depreciating licences where annual fees are based on a declared value at which the licensees would be willing
to sell the licence and that must be sold at that price).
101
Telegraph Act 1863, <http://w ww.legislation.gov.uk/u kpga/ V ict/26-27/112/contents>. See further
Chapter 3.
102
Telegraph Act 1868, <http://w ww.legislation.gov.uk/u kpga/ V ict/31-32/110/contents/enacted>.
103
Records of the Post Office and British Telecommunications public corporation: 1849–1984, BT Digital Archives,
<http://www.digitalarchives.bt.com/Calmview/Record.aspx?src=CalmView.Catalog&id=BTA%2f3+BT1>.
104
Ibid. 105
Ibid.
106
Telegraph Act 1870, <http://w ww.legislation.gov.uk/u kpga/ V ict/33–3 4/8 8/contents>.
107
Attorney-G eneral v The Edison Telephone Co of London, Ltd [1880–81] LR 6 QBD 244.
410
The Postmaster General’s exclusive privilege in telegraphic systems did not ex-
tend to communications exchanged by wireless telegraphy with foreign countries
or with ships beyond its territorial waters.108 Also, Lloyd’s operated its own wireless
systems, the provisioning of which was involved in the Marconi contract. A grant
of powers was needed to effect controls over spectrum use for wireless systems
without disturbing the framework already established for telegraph services under
the current national law. In 1904, the UK passed the first Wireless Telegraphy Act,
granting the Postmaster General the power to license the use of radio spectrum.109
In 1908, the General Post Office built its first ship-to-shore wireless coast station,
and licensed others. In 1909, the General Post Office acquired most of Marconi’s
wireless coast stations110 and, as with telephony systems, continued to take over
others throughout the country including those operated by Lloyd’s. The UK, in 1913,
ratified the International Radio Conference—on the heels of wireless telegraphy’s
role in saving over 700 lives onboard the Titanic in 1912 with its Marconi systems
able to radio for assistance.111 Wireless telegraphy and coastal stations operations
by the GPO continued well into the twentieth century.
The national frequency management infrastructure that exists in the UK today
began in 1918 with the establishment of the Wireless Telegraphy Board112 to manage
interference problems. The Post Office represented non-government users’ inter-
ests throughout the board’s various reconfigurations until it was disbanded in
1948. The Wireless Telegraphy Act of 1949 vested powers in the Postmaster General
generally to license all apparatus using radio frequencies.113
The Post Office Act 1969 abolished the GPO and moved spectrum management
authority to the former Ministry of Posts and Telecommunications.114 Responsibility
passed in 1974 to the Home Office, and in 1983 to the former Department of Trade
and Industry’s Radio Regulatory Division. In 1990, it became an executive agency
within the former DTI called the Radiocommunications Agency (RA) operating
under the Wireless Telegraphy Act as amended over time.
The RA merged with Oftel and three other agencies in 2003 to form Ofcom, a
converged regulator intended to better address convergence in electronic com-
munications networks and services. Ofcom now regulates spectrum under the
Wireless Telegraphy Act of 2006 and the Communications Act 2003, which were
Preliminary Conference, n 60.
108
110
Ibid. 111
Ibid, 1912.
112
See DEFE 59, Record Summary, ‘Ministry of Defence and predecessors: Defence Signal Board and pre-
decessors: Minutes and Papers’ (National Archives), <http://d iscovery.nationalarchives.gov.uk/details/r/
C15205>.
113
Wireless Telegraphy Act 1949, s 1. 114
Post Office Act 1969, s 3(1).
420
amended not only to reflect that change but also to pursue a more market-d riven
approach to spectrum regulation in contrast to the administrative allocation pro-
cesses with command and control oversight. Some amendments to these Acts also
reflect changes required by European Union telecommunications frameworks
that have evolved with respect to spectrum. Although the UK wireless framework
precedes the EU’s by nearly eighty-five years, it must comply with the EU’s require-
ments for licensing radio spectrum. It may be helpful therefore now to examine
the EU regime in this regard.
115
Directive 96/2/EC amending Directive 90/388/E EC with regard to mobile and personal communica-
tions, OJ L 020/59, 26 January 1996.
116
Directive 87/371/E EC on the frequency bands to be reserved for the coordinated introduction of public
pan-European cellular digital land-based mobile communications in the Community.
117
Directive 86/361/E EC of 24 July 1986 on the initial stage of the mutual recognition of type approval for
telecommunications terminal equipment.
118
The Directive stated that it intended to enable the exclusive occupation of the 890–915 and 935–960 MHz
frequency bands for digital cellular communications.
430
119
Directive 1999/5/EC on radio equipment and telecommunications terminal equipment and the mutual
recognition of their conformity, OJ L 91/10, 7 April 1999.
120
Directive 2014/53/E U on the harmonisation of the laws of the Member States relating to the making
available on the market of radio equipment and repealing Directive 1999/5/EC, OJ L 153/62, 22 May 2014.
121
Devices using wave propagation to determine position, eg RFID or radar.
122
Arts 1, 2 RED (with the exception of devices exclusively used for military, state security, amateur radio,
and civil aviation).
04
among other things.123 The RED continues the prior self-certification for harmon-
ized standards as one of the possible mechanisms for conformity. A notified body
must assess if these requirements are met where harmonized standards don’t
exist before self-certification is possible. The CE mark of such technical compli-
ance is still required. The RED requires that where equipment has restrictions on
putting into service or of requirements for authorization of use, information on
the packaging must allow identification of the applicable Member States as set by
the Commission.124
The EU’s continuing approach of technical and other harmonization for open
network provision based on cross-border and pan-European considerations to-
gether with its ‘essential requirements’-only licensing mandates comprises the
essence of the EU spectrum regulation.125 The harmonizing GSM Directive was
amended to permit technological neutrality and refarming of spectrum for 2G,
under harmonized technical conditions set out in Commission decisions, to be
used for 3G and 4G technologies.126
The Licensing Directive 97/13/E C, in the first phase of EU regulation, con-
tained a stated default for general authorizations but clearly permitted in-
dividual licences to accord rights to use spectrum, impose conditions, and
limit the number of licences where necessary to ensure spectrum’s efficient
use as a scarce resource.127 The Directive also only seemed to contemplate
market-pricing structures such as auctions within the context of an indi-
vidual licence.128 These factors, combined with the vast discretion permitted
to the National Regulatory Authorities (NRAs), resulted in the varying indi-
vidual licences and processes remaining the rule across Europe rather than
123
Art 3(1)(a) and (b) (incorporating by reference the essential safety and electromagnetic compatibility
requirements of respectively Directive 2014/35/E U on the harmonisation of the laws of the Member States
relating to the making available on the market of electrical equipment designed for use within certain voltage
limits (the Low Voltage Directive but without the low voltage limits) and Directive 2014/30/E U on the har-
monisation of the laws of the Member States relating to electromagnetic compatibility) and Art 3(2) (adding
‘efficient use’) and (3), RED.
124
Art 10(10). See also Commission Implementing Regulation (EU) 2017/1354 specifying how to present the
information provided for in Article 10(10) of Directive 2014/53/E U, OJ L 190/7, 21 July 2017.
125
This is reflected eg in the Roaming Regulation which sets a mandatory glide path of wholesale and retail
pricing for cross-border roaming service provision which harmonization has been extended to voice and SMS
in order to force down prices in these, to date, monopoly cross-border call-termination markets.
126
Directive 2009/114/EC amending Directive 87/372/E EC on the frequency bands to be reserved for the
coordinated introduction of public pan-European cellular digital land-based mobile communications in the
Community, OJ L 274/25, 20 October 2009. Two Commission Decisions have provided the harmonized tech-
nical rules for the extension of the 900/1800 bands to 3G and more recently 4G uses.
127
See Directive 97/13/EC, Recitals 3, 7, 13, on a common framework for general authorizations and indi-
vidual licences in the field of telecommunications services.
128
Ibid, at Art 11.
450
129
Commission Communication, ‘Toward a new framework for electronic communications infrastruc-
ture and associated services: The 1999 Communications Review’ (1999 Communications Review), COM(1999)
539, 10 November 1999. Accord, Commission Communication, ‘5th Report on the implementation of the
Telecommunications Regulatory Package’ (1999).
130
Directive 2002/20/EC on the authorization of electronic communications networks and services, OJ L
108/21, 24 April 2002.
131
This will include denser networks and at higher bands, mass M2M communications with over 100 x the
number of connected devices, ubiquitous connectivity, and ultra low latency.
132
A potentially significant issue for 5G uses. See eg Presentation, ‘5G Spectrum Sharing’ (Qualcomm, 2
December 2016), <https://w ww.qualcomm.com/i nvention/technologies/5g-n r/spectrum-sharing>.
406
• requiring Member States in all other cases to set out in advance the conditions
for use of spectrum in a general authorization (Article 46(1)).133
The proposed EECC would also further refine the circumstances of individual
grants of rights. The Directive now states that Member States can only require in-
dividual rights where necessary to: avoid harmful interference, safeguard efficient
use of spectrum, ensure technical quality of service, or fulfil other general interest
objectives (Article 5(1)). The proposed Code rewords this, perhaps tilting away
from a possible single decision defaulting to individual grants where potential
challenges to a general authorization exist. It would require that Member States
decide on the most appropriate regime for permitting the use of radio spectrum,
whether by general authorization or individual grant, taking into account factors
that include the:
The proposed EECC thus changes slightly the criteria for deciding whether indi-
vidual grants of use can be permitted. It enhances the technical considerations
(arguably objective) that must be explored seemingly with a view to resolve them
on the technical merits before a weighted decision of which regime is appropriate
in each case can be determined. The Code however permits the Commission to
adopt implementing acts on how Member States apply the above criteria, including
governing issues relating to sharing, receiver resilience, and protecting against
harmful interference (Article 46). The proposed Code would similarly allow
the Commission, as a technical implementing measure, to determine whether
rights in harmonized bands are subject to a general authorization or individual
rights of use (Article 45(2)). BEREC has objected to both of these harmonizing
implementing measures as encroaching on NRAs’ ability to determine needs ac-
cording to national conditions, possibly where a general authorization was man-
dated but individual licences might be more appropriate, such as existing users
in the band or the adjacent band use differs. Also, they contend that harmonized
Adequate receiver resilience is critical to protect against harmful interference. This will be important for
134
denser networks as will be likely in 5G using higher bandwidth as well as for shared spectrum. With enhanced
requirements for receivers under the RED, the issue there and here is getting enhanced regulatory focus.
470
requirements for individual grants of use where not needed in the national market
could ‘sterilise valuable spectrum resources’.135
The inclusion of suitable sharing agreements within the decision criteria for
granting individual versus general authorizations evidences the proposed Code’s
support of spectrum sharing. This support is further reflected in: the Code’s spe-
cific definition of sharing indicating that spectrum can be shared on a licensed or
unlicensed basis and under both the general authorization or individual licence or
combination of the two (Article 2(26)); the NRA harmonization duty to maximize
spectrum sharing by ensuring the least onerous authorization system possible
(Article 45(2) and; the inclusion of the ability to set access conditions for neces-
sary spectrum sharing among the harmonized competences that NRAs must have
(Article 35 (1)(f)). Notable in this regard are, however, the network operators’ ob-
jections to the greater use of spectrum sharing, general authorizations for spec-
trum rights, and other ‘deregulation’ such as allowing third parties to provide
RLAN at the edge of fixed networks (Article 55) in order to preserve their status
quo on markets.136
The proposed EECC would continue the specific time frames for the process
within which individual grants of rights must be awarded, including for spectrum
use, under the Authorisation Directive’s specified procedures. As considered in
Chapter 6 these include a time limit of six weeks for grants of radio frequencies
that have been allocated for specific purposes under the national frequency plan
(Article 5(3)). For allocations by competitive/comparative procedure, a further ex-
tension of no longer than eight months is permitted to ensure that the process is
fair, reasonable, and open (Article 7(4)). Under the proposed EECC, however, time
frames allow for the possibility of a harmonized date set by the Commission for
completion of the specific frequency allocation (Article 54(8)).
The Authorisation Directive permits restrictions on the numbers of persons
granted individual rights to use spectrum only where ‘unavoidable’ and dictated
by scarcity and the need to ensure efficient use137 following procedures for con-
sultation with interested parties and publication of NRA decisions with reasons
135
BoR (17) 91 ‘BEREC’s Paper on the Commission’s Proposals for an EECC Spectrum Provisions—
Implementing Acts’ (BEREC, 27 April 2017).
136
See Orange Position Paper, ‘Spectrum Management: The European Electronic Communications Code’,
December 2016, <https://www.orange.com/en/Group/Committed-to-Europe/The-new-Code-EECC>; European
Telecommunications Network Operators’ Association Position Paper on the European Electronic Communica
tions Code (ETNO, January 2017), <https://etno.eu/datas/positions-papers/2017/ETNO%20Position%20Paper%20
on%20the%20EECC>.
137
Recital 11, Directive 2002/20 as amended 2009. While the recital addresses both spectrum and numbers,
it and Art 5, also addressing both, fail to make clear whether these two criteria for limitation apply both to
spectrum and numbers, the other individual grant, possibly due to unfortunate wording. However, this would
make sense as both are considered scarce public resources, for which ensuring efficient use would seem com-
mensurate under the ‘public trust’ theory discussed in Chapter 6 at Section 6.2.2.
048
justifying the limitation. The grant of such limited rights must be on the basis of
selection criteria that are objective, transparent, non-d iscriminatory, and pro-
portionate (Article 7(3)). The Directive requires review of the grant limitation at
reasonable intervals for continued justification. Where not, the NRA must publish
that decision and invite applications for further grants of such rights. Both users
of communications services and providers of networks and services are eligible to
obtain spectrum use grants (Article 5(2)).
The proposed EECC does not directly refer to selection criteria for limited grant.
It rather first requires Member States to state the reasons for the limitation on rights
of use, giving due weight to the need to maximize benefits for users and to facilitate
the development of competition (Article 54(1)(a)). It then requires Member States to
clearly define and justify the objectives pursued with the selection procedure, and
where possible quantify them, giving due weight to the need to fulfil national and
internal market objectives (Article 54(2), proposed EECC). Possibly referring to the
same ‘objectives’, the proposed Code then specifies that the objectives that Member
States may set out for the grant, with a view to design the specific selection procedure,
must be limited to one or more of:
• promoting coverage;
• required quality of service;
• promoting competition;
• promoting innovation and business development; and
• ensuring that fees promote optimal use of radio spectrum in accordance with
Article 42 that requires they be objectively justified, transparent, non-d iscrim-
inatory, and proportionate in relation to their intended purpose and take into
account an extensive list of policy objectives.138
138
Art 54(2), proposed EECC. The objectives referenced in Art 42 include: the Art 3 general regulatory ob-
jectives of promoting regulatory consistency and predictability, non-d iscrimination, technological neutrality,
promoting high capacity data connectivity, promoting competition in provision of networks, including effi-
cient infrastructure-based competition and services, contributing to the development of the internal market,
and promoting the interests of citizens; the Art 4 spectrum coordinating ‘aim of optimising the use of radio
spectrum and avoiding harmful interference’; and the Art 45(2) objective of spectrum harmonisation of use of
radio spectrum, consistent with the need to ensure effective and efficient use thereof and in pursuit of bene-
fits for the consumer such as economies of scale and interoperability of services and networks by, inter alia,
(a) ensuring coverage of their national territory/population at high quality and speed, both indoors and
outdoors, including along major transport paths, including the trans-European transport network;
(b) ensuring that areas with similar characteristics, in particular in terms of network deployment or
population density, are subject to consistent coverage conditions;
(c) facilitating rapid development in the Union of new wireless communications technologies and appli-
cations, including, where appropriate, in a cross-sectorial approach;
(d) ensuring the prevention of cross-border or national harmful interference in accordance with Articles
28 and 46 respectively, and taking appropriate pre-emptive and remedial measures to that end;
409
The proposed Code requires the further layering that the Member State clearly de-
fine and justify the selection procedure choice, including any preliminary phase
in order to be able to access the selection procedure. Member States must also
state the outcome of any related assessment of the competitive, technical, and
economic situation of the market and provide reasons for the possible use and
choice in adopting any measure under the required NRA competences under pro-
posed Article 35.139 Any exercise of these would be subject to the previously noted,
mandatory but non-binding prior ‘peer review’ by BEREC, the Commission, and
other NRAs, designed to ensure better harmonization of spectrum management
(Article 35(2), proposed EECC). This requirement has been criticized for various
reasons including in the context of the award process that it is unnecessarily com-
plex, delaying, and likely unworkable, eg, since the review would occur at the final
design stage of the award process after rounds of public consultation with any pro-
posed changes further delaying the award and requiring further consultation. The
practical feasibility of the regulators to conduct reviews of the very complex award
processes within the fairly short time frames, especially under harmonized dead-
lines where all twenty-eight Member States would have award processes, is also
questioned.140
Individual rights to use spectrum under the Authorisation Directive can cur-
rently be subject to conditions that may only include those regarding:
(e) promoting shared use of radio spectrum between similar and/or different uses of spectrum through
appropriate established sharing rules and conditions, including the protection of existing rights of
use, in accordance with Union law;
(f) applying most appropriate/least onerous authorisation system possible in accordance with Article 46
in such a way as to maximise flexibility, sharing and efficiency in the use of radio spectrum;
(g) ensuring that rules for the granting, transfer, renewal, modification and withdrawal of rights to use
radio spectrum are clearly and transparently defined and applied in order to guarantee regulatory
certainty, consistency and predictability;
(h) ensuring consistency and predictability throughout the Union regarding the way the use of radio
spectrum is authorised in protecting public health against harmful electromagnetic fields.
139
These would include, in keeping with relevant proposed Code’s requirements: the selection process;
bidder eligibility criteria, parameters of spectrum economic valuation measures; rights duration and renewal
conditions; measures necessary to promote competition; conditions for transferring and assigning spectrum
(including by leasing and trading), sharing spectrum or wireless infrastructure; and limiting individual spec-
trum accumulations. Peer review as per Art 35 (2), proposed EECC.
140
BoR 17/129 ‘Peer Review Process (Article 35)’ (BEREC, 30 May 2017).
140
The only provision in the Code specifically governing spectrum under the gen-
eral authorization is that under Annex I, B (for the provision of networks) that al-
lows, as with the Authorisation Directive, conditions for the use of radio spectrum,
in conformity with ‘Article 7(2)’143 of Directive 2014/53/EU where such use is not
made subject to the granting of individual rights of use in accordance with Articles
46(1) and 48 (the procedures for granting individual rights under the Code). This
would suggest, therefore, that only conditions limited to essential requirements
under the RED may be imposed. Possibly the RED’s efficient/effective use restric-
tions and avoidance of harmful interference requirements can encompass general
141
Part B, Annex.
142
Art 13 ‘Conditions attached to the general authorisation and to the rights of use for radio spectrum and
for numbers, and specific obligations’; Art 47 ‘Conditions attached to general authorisations and to rights of
use for radio spectrum.’ The omission of the additional categories of numbers and specific obligations in Art
47 indicates its likely status as lex specialis.
143
There is no subsection (2) in Art 7 of the Radio Equipment Directive, likely a drafting oversight here.
Article 7, Radio Equipment Directive limits restrictions for radio equipment to those concerning efficient/ef-
fective use, avoidance of interference, and electromagnetic disturbances and public health.
41
144
European Commission, ‘Review of the Electronic Communications Regulatory Framework: Executive
Summary 3: Wireless Networks and Spectrum’, at s 2.2.
145
See Recital 32, Art 13, Authorisation Directive 2002/20/EC as amended.
146
See text and accompanying n 138.
412
of these rights to use this public resource must have a coherent approach in set-
ting fees that should not provide an ‘undue financial burden’ linked to the rights
of use for undertakings providing electronic communications networks and
services (not limited to public) (Article 42(4); Recital 93). They also indicate that
Member States are to ensure that fees not only reflect the economic and tech-
nical situation of relevant markets and any other significant factor determining
the rights’ value but also that they be set in a manner enabling innovation in
network and service provision as well as competition, while also ensuring that
award processes provide safeguards against distorted fees resulting from rev-
enue maximization policies, anticompetitive bidding, or similar behaviours
(Recital 94). The proposed EECC further addresses pricing reserves in any award
process, requiring that these reflect the additional costs associated with ful-
filling conditions imposed to further policy objectives not reasonably met under
normal commercial standards, such as territorial coverage obligations (Article
42(2); Recital 95). This universal service-like cost consideration is not however
required to be counterbalanced with any benefits that might derive from the
obligation such as the benefits of ubiquity for brand such as are built into the
current USO cost/benefits analysis. Proposed fee reserves and other spectrum
economic valuation measures would also be subject to the peer review process
(Articles 42, 35, proposed EECC).
The Commission intends that these proposed changes ensure a greater har-
monization in the allocation and management of spectrum throughout the EU,
a problem it has sought to address in various ways since 2002. The 2002 Radio
Spectrum Decision147 established a framework for EU coordination of spectrum
management approaches across the EU working with the European Conference
of Postal and Telecommunications Administrations (CEPT) to establish the
technical parameters. Under the Decision, the Commission can harmonize the
technical conditions for the use of spectrum to ensure its efficient use, its access
conditions at the EU level, and the interoperability of radio equipment. A Radio
Spectrum Policy Group comprising expert members from the Member States and
chaired by the Commission assists in the work via reports and opinions on stra-
tegic spectrum policy and coordination issues. Since 2002, the Commission has
made over two dozen decisions harmonizing spectrum band uses and setting
harmonized conditions for use and updated or amended these over time to reflect
technological developments and permit refarming. These include decisions to
permit public mobile radio access networks based on low power licence-exempt
WiFi technologies at 5 GHz148 and the refarming of the 900 MHz and 1800 MHz
bands for 4G.149
That the Commission continues to perceive this as inadequate to develop an
EU-w ide harmonized market on the scale of that of the United States is clear from
its other various proposals to enhance EU-centralized control over spectrum
management. These have included the 2007 proposal for creation of a European
Telecommunications Market Authority with delegated powers to oversee spec-
trum regulation and allocation to avoid what it perceived as fragmented national
regulation and uneven roll-out of advanced mobile services. The Member States
resoundingly rejected this, guarding their prerogatives over this valuable resource
and their regulatory competences (as they similarly did with a 2013 ‘Connected
Continent’ proposal for a Commission ‘veto’ over national regulatory decisions).150
They will likely also reject the Commission proposals to change BEREC to op-
erate on a more EU agency-l ike basis, discussed in Chapter 6 and the ‘double lock’
that would allow the Commission ultimately to veto NRA proposed actions in an
Article 7 (proposed Articles 32(3), 33) review of market definitions and remedies
if both the Commission and BEREC agreed that the NRA proposed action was in-
appropriate, further discussed in Chapter 9.
BEREC has not only decried the double veto but also what it calls the ‘hard
harmonisation’ of spectrum in the proposed EECC.151 This has been a significant
balance of competence issue for over the last decade. The development of pan-
European markets using spectrum has been noted to be threatened by the varying
national approaches to spectrum management and regulation, including trading
and refarming of spectrum. In the latter context this was true for the ‘digital divi-
dend’, the vast blocks of spectrum that were being freed up, essentially globally,
for other use by the switch from analogue broadcast television to digital terrestrial
television. This was a one-off opportunity to repurpose spectrum that had been
tied to a specific use and technology for many decades. A lack of EU-w ide harmon-
ization here might have resulted in purely local and possibly non-technological
considerations being applied to valuable spectrum with propagation characteris-
tics that made it suitable for wide area delivery, eg 3G and 4G wireless broadband
148
Commission Decision 2005/513/EC on the harmonised use of radio spectrum in the 5 GHz frequency
band for the implementation of Wireless Access Systems including Radio Local Area Networks (WAS/R LANs),
OJ L 187/22, 19 July 2005.
149
Commission Decision 2009/766/EC on the harmonisation of the 900 MHz and 1800 MHz frequency
bands for terrestrial systems capable of providing pan-European electronic communications services in the
Community, OJ L 274/32, 20 October 2009.
150
See Commission Memo 08–0 4 ‘Commission welcomes European Parliament vote to strengthen the EU’s
Single Market for Telecoms but important questions remain open’ (Brussels, 8 July 2008).
151
BEREC Press Release BoR (17) 95 ‘BEREC Papers on the Review’ (BEREC, 11 May 2017).
41
services or enhanced broadcast and mobile services, among others. Such di-
vergent decisions could have implications for decades since available spectrum
below 1 GHz is rare with previous allocations having occurred half a century ago
in the UK.
With an EU centralized regulator shot down, the Commission’s reforms pro-
posed in 2007 and adopted as amendments to the Framework Directive via the
2009 ‘Better Regulation Directive’ contained specific provisions that, if not full
Commission control, enabled higher levels of harmonization and cooperation
among the Member States and the Commission in strategic planning for spectrum
use. Building on earlier harmonization measures such as the Radio Spectrum
Policy Decision, these arguably were a new order of spectrum regulation within
the EU and with the Commission more in the driver’s seat as a result of: (1) en-
hanced Member State duties of cooperation and spectrum management re-
quirements and, notably, (2) legislative proposals authorized to be made by the
Commission in a multi-a nnual programme of spectrum objectives and adopted
under the co-decisional procedure (now the ‘ordinary procedure’), both con-
sidered as follows.152
The duties of cooperation arose under Article 8a, Framework Directive that im-
posed somewhat amorphous obligations of enhanced planning, coordination,
and harmonization by Member States at the EU level regarding spectrum policy.
Relevant factors for their consideration in optimizing the use of radio spectrum
and avoiding harmful interference include:
152
See Decision 243/2012/E U of the European Parliament and of the Council establishing a multi-a nnual
radio spectrum policy programme (RSPP), 2010/0252 (COD), 15 February 2012. See text accompanying nn
153–158 below for a further discussion of the current programme.
451
Article 9b, Framework Directive requires spectrum transfers or leases for bands
determined by the Commission under the implementing procedures, subject to
continuing application of any attached conditions unless otherwise specified
by the NRA. Other transfers and leases are permitted in other bands but not for
uses not conforming to designated harmonized uses under the Radio Spectrum
Decision. Proposed Article 51, EECC continues these provisions. How the above
Member State discretion regarding continuing condition applicability and non-
compulsory band transfer/leasing interacts with a new provision intended to
ensure consistency and clarity, Article 51(3), is not clear. It states that ‘Member
States shall allow the transfer or lease of rights of use for radio spectrum where
the original conditions attached to the rights of use are maintained.’ Article 51(3)
requires that Member States, without prejudice to the need to ensure undistorted
competition, submit leasing/transfers to the least onerous procedure possible
and, on notification by the lessor, not refuse the spectrum lease unless the lessor
refuses to remain accountable for the original conditions and, on request by the
parties, approve transfer of rights of use unless the transferee is unable to meet the
original conditions for use. The competent authorities must facilitate leases/t rans-
fers by timely considering requests to adapt the conditions and by ensuring that
the rights and the spectrum attached to those rights may best be partitioned or
disaggregated. Under both the current and proposed frameworks, transfer/lease
rights may be delimited where the original grant was not paid for.
Also related to the amorphous strategic planning and cooperation obligations
is the provision for what is essentially secondary legislation by the Commission.
In 2009, Article 9(3) (Framework Directive) authorized it to propose legislation for
multiannual spectrum policy programmes taking utmost account of the Radio
Spectrum Policy Group’s opinion. Such legislation will specify the policy object-
ives for the relevant strategic planning and harmonization of the use of spectrum
(arguably the substance of what the Member States are cooperating in/w ith/for)
under the Framework and the specific Directives and the common policy object-
ives for coordinating EU interests at international bodies competent in spectrum
matters. This purposive and significant planning programme appears to place
much greater control over the direction of EU spectrum allocation and policy with
the Commission, if not to the same level of the rejected supranational regulator of
spectrum.
In pursuit of this agenda, the Commission in 2010 introduced a multi-a nnual
spectrum policy programme, approved by the Council and the Parliament.153 This
153
Decision No 243/2012/E U establishing a multiannual radio spectrum policy programme, OJ L 81/7, 21
March 2012.
416
set the regulatory principles and policy objectives to be applied for various spec-
trum use determinations and other actions through 2015, that included:
• Ensuring that at least 1200 MHz spectrum are identified by the end of 2012 to
meet the Digital Agenda’s stated 30 Mbps target for wireless broadband in light
of demand and that the need for additional harmonized spectrum bands is
assessed;
• Allowing spectrum trading throughout the EU in all harmonized bands where
flexible use has already been introduced;
• Making available sufficient harmonized spectrum for the development of the
internal market for wireless safety services and civil protection;154
• Fostering different modes of spectrum sharing in Europe since there is great and
still growing demand for these bands generally used in licence exempt WiFi ac-
cess with core role in broadband mobile technologies;
• Ensuring that the radio spectrum can be used to support more efficient energy
production and distribution in Europe with wireless promoting a low-carbon
society;
• Finding appropriate spectrum for wireless microphones and cameras
(PMSE);155 and
• By mid-2013 defining details for the EU’s radio spectrum inventory and an ad-
equate analysis of the efficiency of spectrum use, particularly in the 400 MHz
to 6 GHz range which inventory and analysis will serve as the basis for further
harmonization and coordination in appropriate bands.156
154
2016/6 87/E U Commission Implementing Decision on the harmonisation of the 694–790 MHz frequency
band for terrestrial systems capable of providing wireless broadband electronic communications services
and for flexible national use in the Union, OJ L 118/4, 4 May 2016 (harmonizing technical conditions for Public
Protection and Disaster Relief use in various 700 MHz bands).
155
2014/6 41/E U Commission Implementing Decision on harmonised technical conditions of radio spec-
trum use by wireless audio programme making and special events equipment in the Union, OJ L 263/29, 3
September 2014.
156
2013/195/E U Commission Implementing Decision defining the practical arrangements, uniform for-
mats and a methodology in relation to the radio spectrum inventory established by Decision No 243/2012/
EU, OJ L 113/18, 25 April 2013.
157
O’Donohue, P, Presentation, The Radio Spectrum Policy Programme & the Spectrum Inventory (ITU
Regional Development Forum Warsaw, 7 May 2012).
417
158
The Commission in 2014 issued a Report to the European Parliament and the Council on the implemen-
tation of the Radio Spectrum Policy Programme, COM/2014/0228 final where it noted that a final 2015 report
on the particular programme would be forthcoming; the author is unable to locate this or a further proposed
multiannual programme plan. It is difficult to track, therefore, the status of the various programme elements
beyond the interim report in the absence of implementing decisions as noted.
159
Decision, n 1 at Article 10(1). 160 See Case C246/07, Commission v Sweden (20 April 2010).
148
contested this before the Court. The CJEU agreed with the Commission and found
that these conclusions were not legal acts with the form required by the TFEU pro-
vision and did not indicate the legal basis that must underpin EU actions for them
to have legal effect.161 As the Court noted, the legal basis controls the powers of the
Council and the Commission, here requiring a qualified majority on the part of the
Council, for approval.162 It is suggested that the practical significance of the deci-
sion is: enhanced influence for the Commission in WRC negotiations in light of its
right of initiative; reduced influence and veto power for individual Member States
in light of qualified majority Council voting and; generally reduced manoeuvring
room in negotiations due to the requirement for formal, legally binding Council
decisions.163
The outcome of the proposed EU Electronic Communication Code continues the
tug of war as to institutional (and intra-i nstitutional)164 and Member State compe-
tences. The Council and the Parliament committees have adopted positions for the
trilogue negotiations that are underway with a view to agreement in Spring 2018.
Spectrum is clearly a key issue on the table.
In the UK spectrum use is regulated under the Wireless Telegraphy Act 2006
(WTA).165 This Act combined into one statute the legislation under which Ofcom
manages radio spectrum.166 Coming into force on 8 February 2007, this Act replaced
the Wireless Telegraphy Acts 1949, 1967, and 1998, the Marine, etc Broadcasting
(Offences) Act 1967, Part 6 of the Telecommunications Act 1984, and certain pro-
visions of the Communications Act 2003 regarding regulatory obligations with
161
Case C-6 87/15, (25 October 2017), at paras 47–55. 162
Ibid, at para 51.
163
Legal Case Note: ‘CJEU Decision on EU Negotiation Positions in International Bodies’, (European, 18
December 2017).
164
See reported comment by rapporteur Dita Charanzová, European Parliament Committee on the
Internal Market and Consumer Protection (IMCO), that while the Committee on Industry, Research and
Energy (ITRE) leads the Parliament negotiations, IMCO has ‘exclusive competence’ over one-third of the
text, Internet Society, EU: Feedback on the negotiations on the European Electronic Communications Code
(Internet Society European Regional Bureau Newsletter, 18–2 4 November), <https://w ww.internetsociety.
org/resources/doc/2017/european-regional-bureau-newsletter-18-nov-2 4-nov-2017/>.
165
Wireless Telegraphy Act 2006.
166
Practical Law ‘Wireless Telegraphy Act receives Royal Assent’ (Thompson Reuters, 2006), <https://
uk.practicallaw.thomsonreuters.com/4-205-7993?transitionType=Default&contextData=(sc.Default)&firstP
age=true&bhcp=1>.
419
respect to the management of spectrum.167 Parts of the Act have been amended to
address the 2009 reforms and other subsequent changes.
167
See Joint Committee on Consolidation of Bills, First Report of Session 2005–2006, Wireless Telegraphy
Bill [HL], Vol II Minutes of Proceedings and Minutes of Evidence (House of Lords, House of Commons, London,
23 May 2006).
168
See Radiocommunications Agency, ‘Licensing Policy Manual’ Section A: Impact of UK Legislation
(Archived 12 July 2008), <http://webarchive.nationalarchives.gov.uk/20051027120000/http://w ww.ofcom.
org.uk/static/a rchive/ra/rahome.htm>. The specific acts governed by these are ‘to instal or use wireless tel-
egraphy apparatus’ and ‘to establish or use a wireless telegraphy station.’ WTA 2006, s 8(1)(a)–(b)). A licence
under the Broadcast Act may also need to be obtained for certain TV and radio broadcasters.
169
WTA (2006), at ss 8, 35(1). See R v Blake [1997] 1 Cr App R 209.
170
SI 2003/74. The criteria for exemptions are discussed at text and accompanying nn 183–184. These are
routinely amended and updated. See Ofcom, Wireless Telegraphy Exemption Regulations, <https://w ww.
ofcom.org.uk/spectrum/radio-spectrum-a nd-t he-law/l icence-exempt-radio-u se/w ireless-telegraphy-regu-
lations>. Also see, eg The Wireless Telegraphy (Exemption and Amendment) (Amendment) Regulations 2017,
SI 2017/4 6 (amending the Wireless Telegraphy (Exemption and Amendment) Regulations 2010, SI 2010/2512 as
amended by SI 2011/3035, SI 2013/1253, and SI 2014/1484). There would not seem to be an up-to-date summary
list of exempt devices.
171
The spectrum policy role was transferred from BIS to the Department for Culture, Media and Sport
(DCMS) in January 2011. The Secretary of State has residual powers to consult on policy and make a direction
or order, in consultation with Ofcom and other persons, concerning reserving spectrum for specified uses
and licensing exemptions and charges. Communications Act 2003, ss 156, 157; Wireless Telegraphy Act 2006,
ss 1–5. The government has exercised the former extensively. The ability to exercise the latter is questionable
in light of recent case law. See text and accompanying nn 181–186.
172
SI 2011/1210. Other EU reforms not generally involving spectrum were made via other instruments. Eg
some minor amendments were made by The Electronic Communications (Universal Services) Amendment, SI
2011/1209 (eg removing any GC requirement from the scope of the USO). Another not so minor amendment was
240
made by The Communications Act 2003 (Maximum Penalty for Contravention of Information Requirements),
SI 2011/1773 (increasing the maximum penalty to £2,000,000 from £50,000).
173
WTA, s 2. Ofcom must also publish as part of the plan, what spectrum is available and whether these can
be traded. Ibid. The Plan is online at <http://spectruminfo.ofcom.org.uk/spectrumInfo/>.
174
See also Chapter 16.
175
These Communications Act duties include: s 3, General Duties; s 4, Duties for the purpose of fulfilling
Community obligations (implementing Art 8, Framework Directive); s 5, Directions in respect of networks and
spectrum functions and; s 6, Duties to review regulatory burdens.
241
of access, including by bid (ss 12–14, ss 21–23), the Secretary of State may require
Ofcom to exercise their powers in such cases, in such manner, subject to such re-
strictions and constraints, and with a view to achieving such purposes as is speci-
fied in the order (WTA, s 8(3), (4)).
All of these duties were the subject of a recent Court of Appeal decision, EE Ltd
v Ofcom.176 Here the Court found that Ofcom, following the Secretary of State’s
2010 Direction177 set Annual Licence Fees in 2015 to reflect the market value for
the 900 MHz and 1800 MHz spectrum bands (liberalized for other uses by, inter
alia, the EU repeal of the GSM Directive) that it had reallocated to existing users
also pursuant to the Direction for an indefinite period and varied their licences
accordingly. In doing so, therefore, Ofcom had failed to exercise its Article 8,
Framework Directive obligations required to be performed by it in carrying out
its radio spectrum functions by both section 4(1) and (2), CA and section 5(3),
WTA.178 While WTA, section 5 (1), and 5(3) and (4) authorize the Secretary of
State to direct Ofcom in performing its spectrum functions to exercise its power
in such manner as the Secretary may specify and subject to such restrictions and
constraints, with a view to achieving the purposes specified in the Order,179 the
Court found that nothing in the WTA or the Communications Act transferred that
power to the Secretary of State or allowed Ofcom to delegate to the Secretary its
duties under section 4(2) ‘to act in accordance with the six Community objectives
(which give effect, amongst other things, to the requirements of Article 8 of the
Framework Directive and are to be read accordingly)’.180 Thus, the 2010 Direction
could not have this effect. Therefore, as the duties remained with Ofcom, in fol-
lowing the Direction it failed to meet its duties under both Acts and the Framework
Directive.181
The outcome of this decision is not yet clear as the Court of Appeal has given
Ofcom leave to appeal. Practically speaking, the decision effectively renders the
power of the Secretary to give directions meaningless. Although the Court of
Appeal bent over backwards to find that the Direction itself was not ultra vires,
the fact is that the Direction itself was not challenged. Rather, only the 2015 an-
nual fee decision by Ofcom. Had the other Ofcom acts pursuant to the Direction
178
[2017] EWCA Civ 1873, para 54.
179
The speedy reallocation of the liberalized spectrum to 3G use to avoid delay likely via the regulatory pro-
cess and the litigation challenges thereto likely, no matter the outcome. Ibid, para 47.
180
Ibid, para 19.
181
Indeed, the Court found that WTA, s 3(5) gave priority to the CA, s 4(2) duties (Art 8, Framework Directive
duties) in the event of a conflict with Ofcom’s WTA, s 3 powers.
24
been subject to the same analysis as here (reallocating the spectrum to the same
users and making them indefinite), Ofcom would have had to follow its full regu-
latory processes of consulting and making a decision in light of the merits of the
Article 8 considerations that these actions were warranted as meeting the object-
ives. A Secretary of State’s order regarding issues with the EU framework would,182
therefore, always seem a moot act subject to Ofcom’s review under Article 8 as
these apply across the board to all regulatory functions, calling the CA and WTA
provisions into question.
The WTA mandates the use of spectrum without licence183 where the conditions of
the use are unlikely to:
Here, conditions, if any, may only be those permitted under Annex A, Authorisation
Directive (s 8(3A)), ie general authorization conditions. Included within exempt use
are, eg terminal equipment for GSM, UMTS, and now LTE and WIMAX (technologies
for 4G services).184
Where a licence is granted it may be subject to conditions or limitations (WTA,
s 9). While the Act states that these can be any kind Ofcom deem fit, a 2011 re-
vision limits these to areas specified in Annex B of the Authorisation Directive
(WTA, s 9(1A)).185 If the condition limits technological neutrality, the nature of the
182
Not all directions might concern such issues. Eg Ofcom notes that a grant of recognized spectrum access
could be revoked immediately under a WTA, s 5 Direction. As this is regulation outside the framework except
for the application of the Radio Equipment Directive, the Art 8 duties would not apply. See Procedures Manual
for Recognised Spectrum Access for Receive Only Earth Stations (Ofcom, July 2017).
183
WTA, s 8. Ofcom has committed to exempting spectrum whenever possible. See Ofcom, ‘Licensing
Exemption’, Licensing Policy Manual, 2007, <http://w ww.ofcom.org.uk/radiocomms/i fi/l icensing_policy_
manual_2/>.
184
See eg Wireless Telegraphy (Exemption) (Amendment) Regulations 2011. Also see, Ofcom, Licence
Exempt Radio Use, <http://stakeholders.ofcom.org.uk/spectrum/spectrum-management/l icence-exempt-
radio-u se/>.
185
Annex B, Authorisation Directive comprises a nine-item list of the types of conditions that can be im-
posed on the use of spectrum. These include conditions regarding usage fees; technical/operational condi-
tions necessary for safe, non-i nterfering use if different from general authorization obligations; effective and
efficient use; maximum duration; transfer of rights; undertakings made in the course of a comparative/com-
petitive process to obtain the spectrum; obligations to provide a service or use a technology for which the right
to use the spectrum was granted, including coverage/quality requirements where appropriate; obligations
243
service or the type of equipment that can be used, it can be imposed only where
justified by the essential requirements and general interest objectives set forth in
section 9ZA, that comprise:
These implemented the 2009 reforms to the Directive and reflect its layers of safety/
efficiency/international requirements. A review is required for licences granted
for longer than ten years with non-transferrable individual conditions to deter-
mine whether they meet the section 8 exemption criteria, so as to make it eligible
not to be subject to the requirement for a licence (WTA, s 8A).
A decision to grant a licence for exclusive rights to use a frequency, nationally or
otherwise, may not be made by Ofcom except where necessary to protect safety of
life services or other exceptional circumstances exist that Ofcom believes justify
the exclusive grant to ensure a general interest objective, as above listed. Where
the limitation has a significant impact on the market for the use of electromag-
netic spectrum for wireless telegraphy, Ofcom must consult and publish a no-
tice of its intention to limit a grant, specifying the reasons why and the period for
which representations can be made to Ofcom, but that can be no less than a month
(WTA, s 8C).
Limitations of either kind must be reviewed for continuing necessity with the
consultation outcome published (WTA, ss 8B(5); 9ZA(7)). The review/publica-
tion requirements do not apply in the context of technology/service limitation
where the user can opt for a different spectrum frequency without the limitation
(WTA, 9ZA(8)). The precise interaction with the section 8A review for licences and
under international agreements governing frequencies; and obligations specific to experimental use of a fre-
quency. See Annex B, Directive 2002/20/EC as amended by Directive 2009/140/EC.
42
non-t ransferrable conditions of more than ten-year duration for possible exemp-
tion is not clear.
How the section 8A unique licensing provisions practically relate to section 29
of the WTA 2006 is also not precisely clear. Section 29 is intended to implement
Article 7(1)(c) of the Authorisation Directive as amended in 2009. This requires that
where a Member State is considering whether to limit the number of rights of use
to be granted for radio frequencies, it publish any decision to limit the granting
of rights of use, stating the reasons. In keeping with Article 5(5) of the Directive,
section 29 states that it applies to situations where Ofcom considers ‘it appro-
priate to impose limitations on the use of particular frequencies for the purpose
of securing the efficient use of the electromagnetic spectrum’. Spectral efficiency
suggests this is only where limited numbers of users can utilize certain frequen-
cies due to the spectrum’s inability to support more users without interference. As
the section 8A exclusive use decision can apply for general objectives like cultural
diversity or social cohesion as well as efficient use of spectrum, whether section 29
applies at all or only in the last instance is unknown.
Section 29 requires Ofcom to issue an ‘order’ and publish the criteria it will
use to determine the number of available licences and the category of persons to
whom they will be made available.186 The criteria must also be objectively justi-
fied, proportionate to the use, objective, and non-d iscriminatory in keeping with
the requirements of the Framework Directive. Efficiency, or scarcity practically
speaking, is likely to mean specific band or pairs of bands that must be allocated
nationally or regionally in a way so as to avoid interference. The likelihood exists
that scarcity could be co-extensive with exclusive use in some instances where
other interference-a meliorating conditions will not suffice such as spectrum
masks, filters, or smart technologies such as ‘listen before you speak’ transmit-
ters, etc. However, the section 29 requirements may just apply where multiple but
limited numbers of users can share the same spectrum bands, perhaps with prior-
ities or other rationalizing conditions to manage interference.187
186
WTA, s 29. See also, The Wireless Telegraphy (Limitation of Number of Licences) (Amendment) Order
2006, SI 2006/2786 (as further amended).
187
See eg The Wireless Telegraphy (Licence Award) Regulations 2012, SI 2012/2 817 specifying the criteria for
the 800 MHz and 2.6G band awards, cohesive draft Regulation, <https://w ww.ofcom.org.uk/_ _data/a ssets/
pdf_ fi le/0 016/41344/condoc.pdf/>.
245
where a licence is not required but where transmissions occur within the UK may
apply for such grant. Ofcom is to consider an RSA to the same extent as it would
consider an existing licence when it allocates spectrum (WTA, s 20). Ofcom in-
dicated its intent to limit interference with uses and areas covered by RSAs, as it
would with licensed grants.188 The RSA may be given for any use and equipment
specified in the grant and subject to any conditions that Ofcom may consider ap-
propriate,189 including for strength of signal and equipment, but cannot duplicate
any conditions that are already imposed under general conditions. RSA grants
may be converted to a licence or a licence to a grant.190
That RSA conditions could be also imposed under a general condition indicates
that the Act here contemplates an exempt or similar use in the nature of a general
authorization and where the exempt user might wish to preserve the use for fu-
ture allocation. A grant effectively reserving a particular usage could be sought by
government agencies191 whose spectrum use is currently largely ‘licensed’ only by
a voluntary agreement called a ‘side letter’.192 It may also be sought by providers
of networks using equipment that is exempt from licensing, as its risk of harmful
interference is small, such as WiFi.193 Radio astronomy was identified as a use
where RSAs could be valuable to help limit interference. In 2007, Ofcom issued
regulations for the granting of RSAs in connection with radio astronomy uses in
existing bands and in the six locations where presently carried out.194
The Act also contemplates RSAs for uses unlicensed as they are outside the reach
of the Wireless Telegraphy Act. The references to emissions from outside the UK (s
159(1)) as well as the use of a ‘station’ suggests a ‘receiving only’ earth station oper-
ator which although outside of licensing jurisdiction under the Act,195 might wish
to ensure continued frequency availability and interference minimization. Ofcom
188
Ofcom Statement, ‘Spectrum framework review for the public sector: Extending market mechanisms to
improve how spectrum is managed and used’, 31 January 2008, s 2.8.
189
RSAs and conditions could not, however, constrain the uses of public sector agencies, such as the
Ministry of Defence. See Ofcom Consultation, ‘Spectrum framework review for the public sector: Notice of
Ofcom’s Proposal to make regulations on Recognized Spectrum Access for public bodies and consultation on
technical conditions’, 20 June 2008, s 4.7.
190
WTA, s 27.
191
This is limited to Crown bodies. That the Act contemplates this purpose for RSA is reinforced by the au-
thorization to Crown agencies to pay for, inter alia, recognized grants of spectrum use. See WTA, s 28.
192
This would seem a fairly unique UK example of a contract for a licence. See Ofcom Licensing Policy
Manual, ‘Authorisation of radio use for Crown bodies’, 2007, <http://w ww.ofcom.org.uk/radiocomms/i fi/l i-
censing_policy_manual_2/c rown>.
193
Ofcom Licensing Policy Manual, ‘Licence exemption’, 2004.
194
Ofcom, ‘Statement on regulations for recognised spectrum access as applied to radio astronomy’, 28
February 2007.
195
See eg ‘Procedures manual for recognised spectrum access for receive only Earth stations’, July
2017, <https://w ww.ofcom.org.uk/m anage-y our-l icence/r adiocommunication-l icences/s atellite- e arth/
earth-stations>.
426
has promulgated regulations providing for RSAs, their conditions, charging, and
trading in connection with ‘receive only’ earth stations used for fixed satellite or
meteorological satellite services in certain bands, noting that the scheme remains
voluntary.196 With RSAs, conditions would be imposed only under the grant noti-
fication procedures.
RSAs may have a charge, including one determined by auction.197 RSA charging
regulations provide that these can vary according to the nature of the use and
the frequency bands involved.198 With radio astronomy, Ofcom determined that
Administrative Incentive Pricing (AIP) based on the opportunity cost of denying
the spectrum to alternative services, eg broadcasting, was appropriate.199 With
Receive Only Earth Stations, Ofcom has set an annual £500 plus a calculation
based on a rate charge for the bandwidth involved.200 The fee applies to a single
REOS or all REOS within 500 metres.
The Wireless Telegraphy (Register) (Amendment) Regulations 2007 provided for
the inclusion of RSA grants in the registry created by Ofcom to enable spectrum
trading.201 As public entities hold a significant allocation of spectrum estimated to
have a value of over £20 billion, Ofcom was seeking ways to improve its manage-
ment and efficient use, including by trading and licensing of traded RSAs where
possible.202 The latter is required as RSA eligibility often derives from the non-
licence status of the user as a public agency. Therefore use by another undertaking
would not be subject to an RSA and would have to be licensed. Starting in 2007,
Ofcom consulted on RSA tradability by public sector entities.203 In 2009, it adopted
regulations to permit trading of an RSA or conversion to a licence or from a li-
cence to an RSA, both where either all of the rights and obligations are transferred
196
Ofcom, ‘Decision to make Regulations for Recognised Spectrum Access (RSA) for receive only Earth sta-
tions in the bands 1690–1710 MHz, 3600–4200 MHz and 7750–7850 MHz’, 30 November 2011.
197
WTA, ss 21, 23.
198
The Wireless Telegraphy (Recognised Spectrum Access Charges) Regulations 2007, SI 2007/392 as
amended by The Wireless Telegraphy (Recognised Spectrum Access Charges) (Amendment) Regulations 2011,
SI 2011/2762, The Wireless Telegraphy (Recognised Spectrum Access Charges) (Amendment) Regulations
2015, SI 2015/1399.
199
See Ofcom Consultation, ‘Notice of Ofcom’s proposal to make regulations for Recognised Spectrum
Access (RSA) for radio astronomy’, 10 November 2006, ss 5.38–5.60.
200
The Wireless Telegraphy (Recognised Spectrum Access Charges) (Amendment) Regulations 2015, SI
2015/1399; Ofcom, Fees for Grant of RSA for ROES (2015), <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/
0038/6 6899/fees_for_g rant_of_rsa_for_roes.pdf>.
201
The Wireless Telegraphy (Recognised Spectrum Access and Licence) (Trading Regulations) 2009, as
amended (the ‘RSA Trading Regulations’).
202
Ibid.
203
See Ofcom Consultation, ‘Spectrum framework review for the public sector: Notice of Ofcom’s proposal
to make regulations on Recognized Spectrum Access for public bodies and consultation on technical condi-
tions’, 20 June 2008); Ofcom Consultation, ‘Crown Recognised Spectrum Access in the 3400 MHz -3600 MHz’,
17 December 2011.
247
(surrendered to Ofcom which then issues a new RSA/l icence) or they remain con-
current.204 The RSA Trading Regulations also allow for partition of spectrum or
geographically. Various transactions have resulted, enabling public bodies such
as the Ministry of Defence that holds 75 per cent of the public spectrum to share
frequencies for other uses.205
204
The Wireless Telegraphy (Recognised Spectrum Access and Licence) (Trading Regulations) 2009, SI
2009/17, as amended (the ‘RSA Trading Regulations’).
205
Ofcom Consultation, ‘Crown Recognised Spectrum Access in the 3400 MHz– 3600 MHz’, 17
December 2011.
206
RSAs can now be auctioned as well. 207
Wireless Telegraphy Act 1998, s 3, ch 6.
208
See eg Ofcom’s efforts to make regulations for the auction of spectrum in the 2.3 and 3.4 GHz bands in
the face of Three’s continuing legal challenge to the overarching spectrum caps the Ofcom has imposed for
the auction. Ofcom, ‘Notice of intent to make regulations: auction of spectrum in the 2.3 and 3.4 GHz bands’,
17 January 2018, <https://w ww.ofcom.org.uk/spectrum/spectrum-management/spectrum-awards/awards-
in-progress/2-3-a nd-3-4-g hz-auction>.
209
Communications Act 2003, s 167.
248
been given the right to appeal to the Supreme Court from the Court of Appeal’s
decision in EE Ltd v Ofcom.210
The ‘4G’ auction comprising the 800 MHz spectrum digital dividend and 2.6 GHz
and labelled by Ofcom as the ‘largest ever’ UK single auction of ‘internationally har-
monized mobile spectrum’ was also plagued with delays.211 Ofcom had difficulty
deciding whether and how the prime, low frequency 800 MHz bandwidth should be
reserved for bidders other than O2 and Vodafone, both 2G licensees who were even-
tually allowed to use their 2G 900 MHz band spectrum that they were given (origin-
ally for nothing) for refarmed 3G use at lower fees than other 3G spectrum went for
at auction. Below 1 GHz, spectrum is considered to have potentially lower network
roll-out and operating costs due to its propagation characteristics, described earlier,
requiring fewer base stations, lower power, less backhaul, etc. The potential other bid-
ders, Hutchinson 3(3) and Everything, Everywhere (Orange and T-Mobile (Deutsche
Telekom) (EE)), the UK’s other national mobile providers who did not hold any UK
‘sub 1 GHz’ spectrum, contended that the promotion of competition, a Framework
regulatory objective, required that either they should have allocations reserved for
their bidding or that caps should apply to the amount of spectrum that the others
could acquire.212
Possibly to avoid the litigation hinted at in ‘veiled threats’ by O2 and Vodafone if
the rules were not changed,213 Ofcom removed the reservation of sub-1 GHz spec-
trum for EE, the nation’s largest network. In early March 2012, Ofcom notified that
it proposed to grant EE’s petition to refarm in 2012 its existing 1800 MHz spec-
trum to LTE and WiMAX usage. Ofcom indicated in the Notice that it did not con-
sider the licence variation to distort competition in light of the nascent state of the
market and the availability to other operators of the 2 x 15 MHz of 1800 MHz spec-
trum that EE was required to sell as a condition for EU approval of the merger of
T-Mobile and Orange to create EE. Ofcom had earlier concluded that there would
only be a small difference in EE’s ability to deliver a comparable 4G product with
the large amount of 1800 MHz spectrum that it already held for 2G and a network
with more base stations.214 Ofcom varied EE’s licence to permit 4G use before the
210
See text and accompanying notes 176–181.
211
Ofcom, ‘Second Consultation on assessment of future mobile competition and proposals for the award of
800 MHz and 2.6 GHz spectrum and related issues’, 12 January 2012.
212
See Garside, J, ‘4G Spectrum Auction: Time for the Networks to Grow Up’ (Technology Blog, The Guardian,
11 October 2011), <http://w ww.guardian.co.uk/technology/blog/2011/oct/11/4g-spectrum-auction>.
213
See Garside, J, ‘London Becomes 4G High Speed Internet Hot Spot’ (The Guardian, 13 November 2011),
<http://w ww.guardian.co.uk/business/2011/nov/13/4g-h igh-speed-mobile-i nternet-t rial-i n-london?INTCM
P=ILCNETTXT3487>.
214
See Second Consultation, n 211, at 1.24.
249
800 MHz auction.215 Although the other mobile operators argued that this gave EE
an unfair advantage (ultimately less than seven months) in the 4G market, 216 they
did not further challenge the decision when Ofcom agreed to advance the auction
so that the other operators could launch their 4G offers by early summer 2013.
The 800 MHz auction regulation did reserve sufficient spectrum to ensure a
fourth national wholesaler, ultimately Hutchinson, Three’s parent. Ofcom pro-
posed a special condition for one block of 2 x 10MHz bands to be imposed on one
national provider to provide coverage to 98 per cent of the country by 2017 inclu-
sive of ‘not spot’ areas for which Ofcom had £150 million designated for infrastruc-
ture, assuming that others would seek to compete with their network roll-outs.
Vodafone acquired this block.
Litigation challenges also delayed Ofcom’s auction of spectrum in the 2.3 GHz–
3.4 GHz spectrum bands valuable for, respectively, current 4G uses and future ‘5G’
uses (both freed up by the Ministry of Defence for non-m ilitary use in keeping with
a government initiative to make available 500 MHz of spectrum by 2020). The April
2018 auction made available 190 MHz of spectrum, 40MHz in the 2.3 GHz band,
all acquired by O2 and 150 MHz in the 3.4 MHz band, awarded to Vodaphone, O2,
EE, and Three who acquired respectively 50, 40, 40, and 20 MHz each of this 5G
spectrum for a total auction spend of over £1.4 billion. In a second set of auctions,
targeted for 2020, Ofcom will make available spectrum in the 700 MHz band that
the government plans to clear from digital terrestrial use,217 moving it to 470–690
MHz bands and from wireless microphones. Ofcom also plans to auction 116 Mhz
of spectrum in the 3.6GHz–3.8GHz bands, that will be available for 5G use seem-
ingly alongside its current use for fixed links and satellite services.
The first auction originally intended for 2016 was delayed by the O2’s proposed
merger with Three, rejected by the Commission. Rescheduled for late 2017, in its
July 2017 Decision, Ofcom imposed layered spectrum caps to foster competition
in the market in light of the current spectrum holdings of the providers. It capped
215
Ofcom, ‘Notice of Proposed Variation of Everything Everywhere’s 1800 MHz spectrum licences to allow
use of LTE and WiMAX technologies’, 13 March 2012, at 4.30.
216
See Webster, A, ‘UK Regulators delay Orange/T-Mobile LTE plan, give competitors time to react’ (The Verge,
27 March 2012), <http://www.theverge.com/2012/3/27/2905627/ofcom-delays-everything-everywhere-lte-network/
in/2671145>.
217
The same challenges faced decisions to make this available. The 600 MHz frequency is the lower part
of the digital dividend. The decision to move Digital Terrestrial Television (DTT) down to the 600 MHz band
from the 700 MHz band where it is currently operating was considered to be the most beneficial since there is
an emergent international trend to harmonize this frequency for this use with a proposal tabled for this at the
ITU 2015 conference with the US, Africa, and parts of Asia rolling out LTE in this band. This would promote
international interoperability and economies of scope for consumers in terminal equipment. At the same
time, however, the 600 MHz band will not be harmonized for DTT with possible ensuing costs for television
receiving equipment.
340
the amount of overall spectrum holding post-auctions by any one provider to 340
MHz (or 37 per cent share with the added spectrum) and immediately usable spec-
trum post-auction to no more than 255 MHz. EE and Three, for different reasons,
challenged these in court. The immediate use cap meant that EE could not bid
for spectrum in the 2.3 GHz band with its current possible 4G use. BT’s acquisi-
tion of EE gave it a combined pre-auction holding of 45 per cent of total spectrum,
meaning that its auction bid for future use 3.4 MHz was limited to a maximum of
85 MHz. Vodaphone could acquire no more than 160MHz in both auctions but O2
and Three would have no caps.
Three objected to the 37 per cent cap, arguing that it should be lowered to 30 per
cent. BT/EE objected to any caps and the phased implementation, or contiguity, of
5G spectrum auctions, contending that spectrum in all relevant bands for 5G use
should be combined into a single auction. The High Court found for Ofcom218 but
granted Three leave to appeal to the Court of Appeal that it rejected on an exped-
ited basis, allowing the auctions to proceed.
Ofcom’s role is not to be envied. In awarding spectrum, it must anticipate the
long-term technological possibilities (while remaining technologically neutral
where possible), yet also deal with the short-term reality while coordinating with
the EU and possibly internationally for the medium term, all pursuant to legal and
procedural obligations. Even where it seeks to consider all of these required rele-
vant and very complex factors, it faces legal challenge. When it follows government
direction, it faces legal challenge. At the same time, the providers’ challenges are
understandable. Decisions now could serve to preserve their interests for decades.
Their objections to novel ways of proceeding to allow for future developments as
rules and frameworks evolve mean that the spectrum regulatory landscape is
likely to be fraught with uncertainty and future legal challenges to stop or slow
down regulatory developments that they do not like and lobbying in the press and
the political arena to bring pressure on the regulator to backpedal after having
reached a reasoned, if not perfect, decision based on competition, social policy,
and technological considerations.
One area, however, where the industry participants appear to have worked to-
gether more fruitfully in recent years involves another market liberalizing effort,
the trading and leasing of spectrum. The 2003 Act authorized spectrum trading
(s 168).219 With the Wireless Telegraphy (Spectrum Trading) Regulations 2004,
Ofcom began a phased implementation of spectrum trading combined with a pro-
gramme of increasing liberalization of spectrum via de-licensing bands and/or
218
Hutchinson 3G v Ofcom [2017] EWHC 3376 (Admin).
219
Repealed and replaced by the WTA, s 30.
431
220
Ofcom Statement, ‘A Statement on spectrum trading: Implementation in 2004 and beyond’, 6 August 2004.
221
See Ofcom, ‘Trading Guidance Notes’, July 2015, at 1. 222
Ibid.
223
The Wireless Telegraphy (Mobile Spectrum Trading) Regulations 2011, SI 2011/1507, at s 8.
224
Trading Guidance Notes, n 221 at Table 4. 225
Ibid, at s 3.
342
were promised when it was convenient for Ofcom to do so, but have not yet been
done. Ofcom possibly wishes to continue to observe some actual developments
to take into consideration. This would likely need to be done in the event that the
proposed EU reforms requiring the least onerous regime are implemented. Ofcom
has, however, established a Spectrum Trading desk to facilitate these.
Exceptions to the ability to trade spectrum include failure to pay the charges for
the licence and also where Ofcom has not yet made a requested variation or revo-
cation of the licence.226 Ofcom can refuse consent for a transfer where necessary
in the interests of national security, compliance with EU and international obliga-
tions, or pursuant to an order of the Secretary of State under the Communications
Act 2003’s spectrum policy powers under sections 5 and 156.227
The mobile age is as exciting today as it was in 1901 when the new wireless achieve-
ments led a London newspaper to conjecture that people would someday carry
their wireless telephones with them.228 It took almost ninety years for that to be-
come an almost global reality. The inventions that lie at the heart of today’s mo-
bile devices have their roots in scientific achievement that spans two centuries.
Whether 100 years from now, someone marvels at their cochlear ‘brainplants’ that
allow them to ‘hear’ from anyone or anything, everywhere, must be left to imagin-
ation. However, if the Commission’s vision is any indicator, with spectrum to be
allocated for the Internet of Things and new science,229 we cannot rule it out.
The law is often in a catch-up mode. Sea changes in technologies and market
trends will occur with regulation needing to react. The regulator as ‘seer’ is part of
today’s job description. The benefit of such expertise and vision can be witnessed
in a number of the EU’s early decisions to mandate a harmonized EU standard
for mobile communications. The GSM Directive allowed for the quick roll-out and
take-up of wireless communications technology with providers, manufacturers,
and ultimately end users able to benefit from the derived economies of scale and
certainty about interoperability and technical specifications. This view is not
226
See Statement, n 220, at s 7.
227
The Wireless Telegraphy (Mobile Spectrum Trading) Regulations 2011, s 8(5).
228
White, TH, ‘United States Early Radio History: Personal Communications by Wireless (1879–1922)
(noting 4 November 1901, Los Angeles Times, at 6, quotation of London Spectator: ‘Some day men and women
will carry wireless telephones as today we carry a card case or camera’), <http://earlyradiohistory.us/1901age.
htm>.
229
See Decision, n 8.
34
universal,230 and perhaps, in light of the intransigence of mobile rates, NGN roll-
out delays, and other spectrum bottlenecks in the EU which have caused and con-
tinue to cause the Commission to seek new regulatory powers, it is not deserved.
However, not every regulator’s decoder ring always receives across all frequen-
cies and sometimes seers do not have 20/10 vision. Although not perfect, the EU
legal infrastructure in place to analyse and agree standards for mutual imple-
mentation of spectrum management has been workable. The review and proposed
reforms are a good thing if only to step back and take a look at what is needed
for future developments and what is working well enough in light of possible
alternatives.
Spectrum regulation is a truly complex exercise. The allocation of these sig-
nificant blocks of spectrum are decisions with potential impact for generations,
and require engineering and economic expertise and the administrative ability to
meet all the policy objectives that must be mashed into the market mechanism the
regulator must use for decisions that, ultimately, will not make everyone happy.
How the proposed reforms shake out in 2018 are not likely to make everyone happy
either.
230
See Sutherland, E, Paper: ‘European Spectrum Management: Successes, Failures & Lessons’ (ITU
Workshop on Market Mechanisms for Spectrum Management Geneva 22–2 3 January 2007). Sutherland, how-
ever, does not seem to question the GSM Directive itself but the lack of sufficient competition introduced at
the time continuing to today with a further failure to anticipate the high pricing issues and the disadvantage
fixed networks had in termination rates.
34
435
8.1 INTRODUC TION
For the purposes of this chapter, and under the European Union’s Access Directive,2
the term ‘access’ encompasses all kinds of contractual (private law) arrangements
under which an operator or service provider acquires services from another op-
erator in order to enable it to deliver services to its own customers. The issues dis-
cussed in this chapter relate to the regulated (public law) rights of operators to
access each others’ networks and services at a wholesale level, not the rights of end
users to access telecommunications services, at a retail level.
The primary rationale in mandating different kinds of access in a liberalizing
market is to reduce barriers to market entry, so a new operator will not have to
replicate every network element that the incumbent has before being able to offer
a competing end-to-end service. Once liberalized, however, there will generally
1
This chapter was originally written for the 2nd edition by Emma McCormack.
2
Directive 2002/19/EC of the European Parliament and of the Council of 7 March 2002 on access to, and
interconnection of, electronic communications networks and associated facilities, OJ L 108/7, 24 April 2002
(the ‘Access Directive’). The definition of ‘access’ is in Art 2(a).
346
continue to be a need to mandate access from those market players that con-
trol facilities and services that a competitor cannot feasibly, economically or
technically, duplicate. In a networked industry like telecommunications, com-
petitors are usually also your customers; creating so-c alled ‘frienemies’. Access
rights are also a means of reducing the environmental impact of competition
by, for example, reducing the need for road works or the installation of masts.
Access rights which are commonly regulated include network access (eg access
to the ‘local loop’3), the provision of wholesale products for resale (eg access to
wholesale DSL products), and access to services and infrastructure necessary
for the provision of a service (eg access to co-location or number translation
services). Broadly speaking, access services can be distinguished into ‘active’
and ‘passive’ elements; the former including any access to the operator’s trans-
mission network and associated operational systems, while the latter would in-
clude access to ducts and poles.
‘Interconnection’ is a type of access right.4 At its most basic level, interconnec-
tion involves the physical means of linking two different networks for the exchange
of traffic, so that users on one network may communicate with users on the other.
Typically, interconnection arrangements provide for two networks to be joined to-
gether at a ‘point of interconnection’ and require each operator to carry messages
received from the other operator at the point of interconnection across their net-
work and to either ‘terminate’ them with the relevant user or pass the messages
onto another network operator.
There are obvious incentives for operators to enter into interconnection ar-
rangements with operators in other territories. Incumbent operators will have
had interconnection arrangements with incumbent operators in other coun-
tries, in order that its users could make and receive calls from users in the other
country. 5 Interconnection means the ability to extend an operator’s reach and
provide a wider range of (sometimes high-cost and very profitable) services
to users.
There is, however, little commercial incentive for an incumbent operator to
interconnect with an operator who wants to compete in the same geographic
market. Incumbent operators with a large number of customers may well deter-
mine, in the absence of appropriate regulation, that they bear little commercial
risk if their users cannot contact the (initially very small number of) users on a
new competitor’s network. The new entrant, however, cannot survive without
3
The ‘local loop’ is sometimes referred to as the ‘local access network’ and refers to the part of a telecommu-
nications network that connects end-u sers premises with the nearest telecommunications exchange.
4
Access Directive. The definition of ‘interconnection’ is in Art 2(b).
5
International interconnection arrangements are examined further in Chapter 16, at Section 16.3.5.
437
6
See Telecom Corporation of NZ Ltd v Clear Communications Ltd (1992) 4 NZBLC. Such issues are now sub-
ject to an ex ante access regime under the Telecommunications Act 2001, implemented by the Commerce
Commission of New Zealand.
7
See also Chapter 11.
348
8.2.2.2 Transit
The basic scenario described above, where the two networks are directly inter-
connected, could be varied in a number of ways. For example, it may not be effi-
cient for a small fixed-l ine operator to establish direct interconnection with every
other operator. Instead, a small operator may rely on a third operator, often a large
incumbent operator, to transit traffic across that other operator’s network to the
terminating operator’s network. Provided that both the originating operator and
the terminating operator are interconnected with that third party and the third
party has agreed to transit calls across its network, the calls will be connected even
though the parties have not established direct interconnection arrangements.
Much more complex arrangements have evolved. In the case of international calls
or the exchange of internet traffic (discussed further in Section 8.2.3), messages or
data may pass through many telecommunications networks.
8
See also Chapter 2, at Section 2.13 and Chapter 16 at Section 16.3.5. See further Chapter 15.
9
40
See European Commission decision in Case IV/M.1069 WorldCom and MCI, OJ L 116/1, 4 May 1999.
10
41
The peering agreement will sometimes indicate a ratio (eg 1:4) within which the relative volume of traffic
12
flows between the networks may vary, but if they fall outside it can trigger an option to levy charges or renego-
tiate the agreement.
42
One of the most intrusive examples of network access lies in local loop access.
The ‘local loop’ is defined in the following terms:
the physical circuit connecting the network termination point to a distribu-
tion frame or equivalent facility in the fixed public electronic communications
network.13
This ‘last mile’ of an incumbent’s fixed network has generally been the last com-
ponent of its network to be upgraded to enable high bandwidth transmission
capacity. Where replacement with optical fibre has not occurred, it has been pos-
sible, through digital subscriber line (DSL)14 technologies, to upgrade the trad-
itional copper-based local loop to provide high speed, ‘broadband’ internet access
services to users. By obtaining access to the local loop, operators obtain the right
to locate equipment at a telephony operator’s local exchange and to physically
connect that equipment to end-users’ local lines, in order to provide DSL-en-
abled broadband internet services to customers. Where access to the whole line is
obtained, the broadband operator controls the provision of telephony services to
the customer as well. Alternatively, ‘shared access’ is where the original operator
continues to provide voice telephony services, with the alternative operator pro-
viding broadband services to the user.
Access arrangements may relate to facilities as well as network elements. An ex-
ample of access to non-network facilities is the arrangement between the mobile
operators O2 and T-Mobile to share mobile telephony masts and sites in Germany
and the UK, as well as to provide reciprocal roaming services to each others’ cus-
tomers. These facility-sharing arrangements were achieved through commercial
negotiations, and were subsequently cleared by the European Commission.15
The majority of access arrangements are not as intrusive as the ones described
above, and are more like straightforward interconnection arrangements. An ex-
ample is the provision of carrier pre-selection, which allow users who are customers
of one network to select an alternative operator in advance for particular calls (eg
all national calls) without dialling additional digits on their telephone. Unlike
local loop access, carrier pre-selection only involves the interconnection of the two
13
Access Directive, Art 2(e).
14
There are many variants of DSL technology. The variant most commonly used for upgrading the local
loop has been asymmetrical digital subscriber line (ADSL) technology. ADSL provides fast download speeds,
but comparatively slower speeds for uploading data. To obtain higher speeds over copper, other protocols are
deployed, such as VDSL or G.fast.
15
See O2 UK Limited/T-Mobile UK Limited (UK network sharing agreement) (2003) OJ L 200/59, 7 August
2003 and T-Mobile Deutschland GmbH/Viag Interkom GmbH (Germany network sharing agreement) (2003) OJ
L 75/32, 12 March 2004.
43
networks; it does not involve the alternative carrier taking physical control of the
incumbent’s network infrastructure.16
Some access arrangements are even less intrusive than the ones described
above. For example, the provision of wholesale line rental requires the operator
who owns a local access line to provide that line on a wholesale basis to another
operator. This product is particularly useful for operators who have obtained car-
rier pre-selection, as it allows them to bill their customers for both calls and for line
rental. In this case, the incumbent operator will continue to service the customer’s
line, although the customer’s contract will be with the other operator.
8.3.1 Introduction
The interests of new entrants in the telecommunications sector may be said to be
protected by two distinct tiers of regulation in the EU. Firstly, general competi-
tion law prohibits certain anti-competitive agreements and the abuse of a dom-
inant position.17 Secondly, sector-specific ex ante measures under the Access
Directive. Operators are required to comply with both competition law and the
sector-specific rules.
Most legislators and regulators recognize that general competition law rules
are inadequate for fostering the emergence of competition in telecommunications
markets. This is because the telecommunications sector may be said to have spe-
cial characteristics which justify a more interventionist approach than is involved
where general competition law is applied. These characteristics include the preva-
lence of previously state-owned and state-funded operators who have historic-
ally enjoyed a legal monopoly. These operators have often maintained very high
market shares, even many years after the introduction of competition. Secondly,
operators who wish to enter the market by building competing infrastructure face
very high barriers to market entry; it may not be possible economically, for ex-
ample, to build out an entire competing communications network. Thirdly, the
fact that cooperation among competitors, in the form of access and interconnec-
tion arrangements, is essential for the successful operation of competitive com-
munications markets. In this sector, your competitor is also your customer.
16
Carrier pre-selection was required of all SMP operators under the Universal Services Directive, Article
19, but this was deleted by the 2009 Reforms and under the Access Directive, NRAs may now require an SMP
operator to give access to any network elements or facilities that allow carrier selection or pre-selection (Art
12(1)(a)).
17
See further Chapter 10.
4
18
Directive 97/33/EC of the European Parliament and of the Council on interconnection in telecommu-
nications with regard to ensuring universal service and interoperability through the application of the prin-
ciples of open network provision (ONP), OJ L 199/32, 26 July 1997 (the ‘Interconnection Directive’).
19
The functions of the Director General of Telecommunications (DGT) now rest with Ofcom.
20
Telecommunications Act 1984, s 7(5), (6).
21
See Beesley, ME, and Laidlaw, B, ‘The British Telecom/Mercury interconnect determination: an expos-
ition and commentary’, in Beesley, ME, Privatisation, Regulation and Deregulation (2nd edn.) (Routledge,
1997) pp 299–327.
45
Further demands for interconnection arose when the first post-duopoly PTO
licences were issued in 1993, and when the first international facilities licences
were issued in 1996.22 The new licensees were required to show that they had rele-
vant connectable system (RCS) status in order to be entitled to interconnection.
In practice, RCS status was defined in such a way that most PTO licensees were
entitled to interconnection.
In 1994, Oftel commenced a major review of interconnection pricing. The review
was needed to take account of the growing level in sophistication of the intercon-
nection products needed and a growing requirement on the part of operators to
purchase disaggregated interconnection services. In 1996/97 Oftel required that
BT’s interconnection charges be based on the forward-looking incremental cost of
replacing capital assets, rather than the historic cost of what the assets cost when
originally purchased.23 This type of cost modelling in respect of interconnection
pricing was finally adopted in a Commission Recommendation in 1998.24
24
Commission Recommendation 98/195/EC ‘on Interconnection in a liberalized telecommunications
market’, OJ L 73/42, 12 March 1998.
46
SMP operators were also subject to a range of other obligations under the
Interconnection Directive. 25 These included adherence to the principle of
non-d iscrimination, and a requirement to make interconnection agreements
available to the national regulatory authority and to interested parties. SMP
operators providing fixed-line networks and leased lines (but not SMP op-
erators providing mobile networks) were also required to set transparent
and cost- orientated interconnection charges, and to publish a reference
interconnection offer.
The second tier of regulation under the Interconnection Directive required all
operators authorized to provide the public telecommunications networks and
services set out in Annex II of the Directive to negotiate interconnection with each
other on request (Article 4(1)). The categories in Annex II were:
(i) organizations which provided fixed and/or mobile public switched tele-
communications networks and/or publicly available telecommunications
services, and controlled the means of access to termination points identified
by numbers in the national numbering plan;
(ii) organizations which provided leased lines into users’ premises;
(iii) organizations which were authorized in a Member State to provide inter-
national telecommunications circuits between the Community and third
countries, for which purpose they had special or exclusive rights; and
(iv) organizations which provided telecommunications services which were per-
mitted to interconnect in accordance with relevant national licensing or au-
thorization schemes.
26
Directive 97/13/EC on a common framework for general authorizations and individual licences in the
field of telecommunications services; OJ L 117/15, 7 May 1997 (the Licensing Directive).
27
DGT, Determination of a dispute between BT and MCI Worldcom concerning the provision of a Flat Rate
Internet Access Call Origination product (2000).
28
Such as at its tandem exchanges. See DGT, Determination relating to a dispute between British
Telecommunications and Worldcom concerning the provision of a Flat Rate Internet Access Call Origination
product (FRIACO) (2001).
48
Another direction made by the DGT under the Interconnection Regulations con-
cerned what are known as radio base station (RBS) backhaul circuits. RBSs are the
base stations that transmit signals to and from mobile handsets. RBS backhaul circuits
are functionally identical to PPCs, but they are used to link RBSs with the main part
of a mobile operator’s network. A dispute arose between BT and Vodafone as to the
provision of RBS backhaul circuits. In June 2003 the DGT, using its powers under the
Interconnection Directive and the Interconnection Regulations, directed BT to pro-
vide RBS backhaul circuits to Vodafone on terms similar to those applying to PPCs.32
BT challenged the DGT’s right to investigate the dispute on the basis that RBS
backhaul circuits do not fall within the definition of ‘interconnection’ under the
Interconnection Directive and the Interconnection Regulations. In May 2004, the
Competition Appeals Tribunal (CAT) handed down a decision holding that RBS
backhaul circuits are not interconnection products, and, accordingly, the DGT
had no jurisdiction over the Vodafone/BT dispute.33
The principal reason for the Tribunal’s decision was that RBS backhaul circuits
are, in effect, used by Vodafone to construct its own network: they link a Vodafone
RBS with the main part of Vodafone’s network. RBS backhaul circuits do not
29
Retail minus pricing does not involve setting an absolute level of charge; it allows the operator to set the
level of charges according to its commercial judgment. However, the operator is required to ensure that a
sufficient margin exists between the charge in question and the relevant downstream price so as to allow the
necessary additional costs of providing the downstream product to be recovered. Setting prices on a retail
minus basis should ensure that no discrimination takes place between the downstream arm of the operator
providing the product and competing operators.
30
DGT, Direction to resolve a dispute between BT, Energis and Thus concerning xDSL interconnection at the
ATM switch (2002).
31
DGT, Direction under condition 45.2 of the public telecommunications operator licence granted to BT under
Regulations 6(3) and 6(4) of the Telecommunications (Interconnection) Regulations 1997 (2001), DGT, Phase 1
Direction to resolve a dispute concerning the provision of partial private circuits (2002), DGT, Partial private cir-
cuits, Phase 2—a Direction to resolve a dispute (2002).
32
DGT, Direction to resolve a dispute between BT and Vodafone regarding wholesale connections between
BT’s and Vodafone’s networks (radio base station backhaul circuits) (2003).
33
[2004] CAT 8. Although the DGT’s determination was made before the Communications Act 2003 came
into force, the appeal was made after that time, and so proceedings were brought before the CAT rather than
the High Court.
49
34
Pursuant to the Telecommunications (Licence Modifications) (Standard Schedules) Regulations 1999,
SI 1999/2 450 and the Telecommunications (Licence Modification) (Mobile Public Telecommunications
Operators) Regulations 1999, SI 1999/2 453.
35
Oftel, Wholesale line rental: Oftel’s conclusions—statement (2003).
540
loop networks. Local loop unbundling was seen as the shot in the arm necessary
for stimulating the broadband market.
The Commission adopted a Recommendation36 in May 2000 recommending that
Member States should mandate access to the local loop by the end of that year. It
became clear, however, that many Member States were unlikely to meet this target.
The European Parliament and Council then adopted a Regulation37 requiring
‘notified operators’ to meet reasonable requests for unbundled access to their local
loop and related facilities under transparent, fair, and non-d iscriminatory con-
ditions, and to publish a reference offer for such access. NRAs were given powers
to intervene to ensure non-d iscrimination, fair competition, economic efficiency,
and maximum benefit for users, and to settle disputes. The Local Loop Regulation
was repealed as part of the 2009 Reform.
Under the Local Loop Regulation, ‘notified operators’ were those designated
by NRAs as having significant market power in the provision of fixed public tele-
phone networks and services under the Interconnection Directive. This meant
that BT and Kingston were ‘notified operators’ in the UK, and, accordingly, licence
conditions were imposed on them in August 2000.
In the months after the imposition of the licence condition a large number of
operators expressed interest in obtaining access to BT’s local access network.
These operators signed confidentiality agreements with BT and joined an oper-
ator interest group established by Oftel to facilitate progress. However, the vast
majority of these operators never obtained local loop access and withdrew their
interest. This can be attributed to many factors, including financial strains on the
telecommunications industry at the time. Some in the industry, however, criti-
cized the DGT and Oftel for failing to take swift and appropriate action against BT
when faced with complaints by operators seeking access. To address this concern,
Ofcom established a Telecommunications Adjudication Scheme in 2004, operated
under the auspices of a Telecommunications Adjudicator, to facilitate competitor
access to BT’s local loop (which has since become operated by Openreach38).
Over recent years, the availability and take-up of broadband internet access has
boomed, mainly through resale of wholesale DSL products obtained from BT.
36
Recommendation 2000/417/EC of 25 May 2000 on unbundled access to the local loop, OJ L 156/4 4, 29
June 2000.
37
Regulation 2887/2000 of the European Parliament and Council of 18 December 2000 on unbundled ac-
cess to the local loop, OJ L 336/4, 30 December 2000 (‘Local Loop Regulation’).
38
See further <http://w ww.offta.org.uk>. 39
See further Chapter 14.
451
40
Directive 95/47/EC on the use of standards for the transmission of television signals, OJ L 281/51, 23
November 1995.
41
The Class Licence was issued under the Telecommunications Act 1984, s 7 in January 1997 and re-issued
in August 2001.
42
August 1999.
43
DGT, Decision as to the status of Sky Subscriber Services Limited as a regulated supplier in the market for
access control services for digital interactive TV services (2000).
44
See Commission Communication, ‘Towards a new framework for Electronic Communications infra-
structure and associated services: The 1999 Communications Review’, COM(1999) 539, 10 November 1999.
542
The New Regulatory Framework of 2002 includes the Access Directive and the
Framework Directive.45 The Access Directive defines Member States’ duties in
relation to imposing access obligations. The Framework Directive is relevant in
understanding these duties, in particular because it sets out the market analysis
process which must be undertaken when imposing access obligations based on
an undertaking’s market power. Both Directives were subsequently amended in
2009, by Directive 09/140/EC.46
45
Directive 2002/21/EC on a common regulatory framework for electronic communications networks and
services, OJ L 108/33, 24 April 2002 (‘Framework Directive’).
46
Directive 2009/140/EC amending Directives 2002/21/EC on a common regulatory framework for elec-
tronic communications networks and services, 2002/19/EC on access to, and interconnection of, electronic
communications networks and associated facilities, and 2002/20/EC on the authorization of electronic com-
munications networks and services, OJ L 337/37, 18 December 2009.
47
See further Chapter 6, at Section 6.4.4.4.
543
‘impose’ in all situations where operators exercise Code Powers, subject only to
the principle of proportionality (Article 12(1)). Second, the no ‘viable alternative’
threshold for intervention has been removed. Third, the persons who may be
required to share has been broadened from operators with Code Powers to any
owner of ‘wiring inside buildings or up to the first concentration or distribution
point where this is located outside the building’, if such sharing can be justified ‘on
the grounds that duplication of such infrastructure would be economically ineffi-
cient or physically impracticable’ (Article 12(3)). Fourth, a transparency obligation
has been inserted whereby undertakings are required to provide, on request, in-
formation about the nature, availability and geographical location of any facilities
referred to in the first subsection, to enable an NRA to establish an inventory of
such facilities (Article 15(4)).
Access has been defined in the very broadest term, catching not only network ac-
cess but access to physical infrastructure such as ducts and masts, and to related
facilities such as software. The inclusion of ‘virtual network services’ in the def-
inition also seems to imply that access obligations can be imposed on those who
do not own the underlying network, but have other rights to use it, such as Mobile
Virtual Network Operators.48
• Member States must impose a general obligation on all providers of public elec-
tronic communications networks to ‘negotiate interconnection’ with other such
providers on request.
• NRAs are to encourage and ensure adequate access and interconnection, and
the interoperability of services in a way that promotes efficiency, sustainable
competition, and gives maximum benefit to end-users.
• NRAs may impose additional obligations on operators designated as having
SMP on a specific market, from a list of remedies detailed in Articles 9 to 13a.
• NRAs must impose specific access obligations in relation to conditional access
services.
49
The right of operators to negotiate is similarly enshrined in the Authorisation Directive, Article 4(2)(a).
50
See Case C-277/07, Commission v Poland, 13 November 2008; [2008] ECR I-8 403, at para 36.
54
51
Case C-192/0 8, 12 November 2009; [2009] ECR I-10717. Inserted in 2009.
52
546
Article 5(1) goes way beyond anything under the 1997 Interconnection Directive.
It should be emphasized that it accords Member States the right to impose access
obligations even on operators who do not have market power, where the NRA takes
the view that such obligations are needed to ensure ‘adequate access and inter-
connection, and interoperability’.
Conditions imposed under Article 5(1) must be notified to the Commission
and other NRAs in accordance with the procedures under Articles 6, 7, and 7a of
the Framework Directive, as a check on how these powers are used.53 Under the
original proposal for the Access Directive, the Commission would have been en-
titled to require conditions set under Article 5(1) to be withdrawn, but this right
was not included in the final wording. In 2006, the Commission again proposed
an amendment to require prior authorization from the Commission for Article
5(1) obligations,54 but it was absent from the 2009 Reform; which is illustrative on
the ongoing political tussle over the best approach towards harmonization in the
European Union.
SMP designation A key difference between the access and interconnection re-
gime under the Interconnection Directive and that under the Access Directive
concerns the test that is applied to determine whether an undertaking is con-
sidered to have SMP. Whilst the Interconnection Directive created a presumption
of SMP where an operator had 25 per cent market share, the Access Directive sets
a higher hurdle by adopting a definition which is consistent with European com-
petition case law:
. . . either individually or jointly with others, it enjoys a position equivalent to
dominance, that is to say a position of economic strength affording it the power
to behave to an appreciable extent independently of competitors, customers and
ultimately consumers. (Article 14)
Commission Staff Working Document ‘on the Review of the EU Regulatory Framework for electronic
54
The process that NRAs are to undertake to determine whether any undertaking
in a given market has SMP is set out in the Framework Directive.55 NRAs are re-
quired to take the ‘utmost account’ of the Commission’s guidelines for market
analysis and the assessment of SMP56 (EC Guidelines), and the Commission’s
Recommendation on the relevant product and service markets.57
There is arguably an underlying tension between the requirement that NRAs
must on the one hand define markets according to general competition law, and
the requirement on the other hand that they must take ‘utmost account’ of the
four markets defined in the Commission’s most recent Recommendation. If a na-
tional regulatory authority undertakes a study of part of the industry to determine
the relevant market, and such study is undertaken strictly in accordance with the
tests set out in general competition law, it is hard to see how taking the ‘utmost
account’ of the Commission’s list of markets is meant to change or influence that
analysis. However, if NRAs define markets that differ from those set out in the rec-
ommendation, they are required to undertake a consultation process with other
national regulatory authorities and the Commission. The Commission may ultim-
ately require a market definition which departs from its Recommendation to be
withdrawn.58
Based on its market analysis, each NRA must determine whether the market
in question is effectively competitive. If the market is not effectively competitive,
the NRA must identify the operators with SMP on that market and impose appro-
priate, specific, regulatory obligations, or maintain or amend obligations already
in place (Article 16(4)). The decision to designate or not designate an operator as
having SMP is subject to a consultation procedure under the Framework Directive.
The Commission ultimately has the power to require that a market definition or
market analysis decision be withdrawn (Article 7).
Where a finding of SMP is made, NRA must impose at least one of the rem-
edies detailed in Articles 9 to 13a of the Access Directive. These wholesale
remedies, graduated in nature from mild behavioural obligations to signifi-
cant structural intervention, are given preference over the imposition of retail
remedies under Article 17 of the Universal Services Directive. The intent of the
Commission is that these remedies should be exhaustive; therefore, no other
obligations in respect of access and interconnection can be imposed unless the
Ibid.
55
Commission Guidelines on market analysis and the assessment of significant market power under the
56
Community regulatory framework for electronic communications networks and services, OJ C165/6, 11
July 2002.
57
Commission Recommendation (2014/710/E U) of 9 October 2014, OJ L 295/79, 11 October 2014.
58
Framework Directive, Arts 6, 7, and 15(3). See further Chapter 4, at 4.6.
548
Remedies The most important obligation for the present discussion is that
which can be imposed under Article 12 of the Access Directive and which re-
lates to the provision of ‘access to, and use of, specific network facilities’,
particularly in situations where it is considered that denial of access or unrea-
sonable terms and conditions would hinder the emergence of a sustainable
competitive market at the retail level, or would not be in the interest of end-u sers
(Article 12(1)).
The following types of access obligations are specifically envisaged in Article 12:
59
‘Dark fibre’ is optical fibre transmission capacity which has not been connected to a laser system.
60
Access Directive, Art 8(4). See also C-2 8/15, Koninklijke KPN BV v ACM, 15 September 2016.
549
Where an access obligation is imposed under Article 12, additional factors must be
taken into account:
Apart from the access obligations in Article 12, as described, the other regula-
tory obligations that may be imposed on SMP operators are as follows:
Article 9 Obligation of transparency—Operators may be required to make public
a range of information, such as accounting information, technical specifications,
network characteristics, and terms, conditions and prices, or to publish a ‘refer-
ence offer’. The reference offer should be sufficiently unbundled to enable a re-
questing operator to only receive what is necessary for the requested service.
Where an operator is required to give access to the local loop, the operator must
be required to publish a reference offer which includes the provisions specified in
Annex II of the Access Directive;
Article 10 Obligation of non- discrimination— Operators may be required
to provide equivalent conditions in equivalent circumstances, and provide
services and information of the same quality as it provides to its own down-
stream businesses;
Article 11 Obligation of accounting separation—Operators may be required
to keep separate accounts in respect of interconnection or access services,
including its internal transfer pricing. The regulator shall be able to require the
disclosure of accounting data and may publish such information as it considers
necessary to contribute to a competitive market, while respecting commercial
confidentiality; and
Article 13 Price control and cost accounting obligations—I n specific circum-
stances, operators may be subjected to price caps, including controls requiring
that prices are ‘cost orientated’. The meaning of ‘cost orientation’ is not further
defined, although the CJEU has held that it does not mean the ability to recover
all costs incurred by an operator, which meant that a NRA was able to ‘set the
prices of the services covered by such an obligation below the level of the costs
640
incurred by that operator to provide them, if those costs are higher than the costs
of an efficient operator’.61 The burden of proving compliance with any obliga-
tion of cost-orientation will reside with the operator (Article 13(3)). Reflecting
the desire to encourage the deployment of NGNs by SMP operators, NRAs are
required to allow ‘a reasonable rate of return on adequate capital employed,
taking into account any risks specific to a particular new investment network
project’ (Article 13(1)). Introduced under the 2009 Reforms, this amendment
was designed to respond to the threat that Member States could be tempted
to grant regulatory ‘holidays’ to SMP operators, generally the national incum-
bent, in order to promote investment in NGNs. Such an approach was adopted
in Germany in respect of so-c alled ‘new markets’, although it was subsequently
struck down by the CJEU.62
Article 13a Functional separation—Introduced under the 2009 Reforms, this is
an ‘exceptional’ structural remedy, under which an operator would be required ‘to
place its activities related to the wholesale provision of relevant access products
in an independently operating business entity’. The remedy is modelled on the
UK’s experience with BT and Openreach. In addition, the separated undertaking
may be subject to any of the behavioural obligations specified in Articles 9–13. As a
measure of last resort, imposition of this remedy is subject to a distinct assessment
and notification procedure.
While Articles 9–13a are ex ante regulatory measures, an additional procedure
has been inserted into the Access Directive as an ex post response to an under-
taking deciding to voluntarily separate its ‘local access network assets’ into an-
other legal entity (Article 13b). The operator is required to notify the NRA ‘in
advance and in a timely manner’, to enable the NRA to carry out an assessment of
the intended separation and impose, maintain, amend, or withdraw any obliga-
tions in respect of the operator.
61
Case 277/16, Polkomtel v Prezes Urzędu Komunikacji Elektronicznej, 20 December 2017, at para 40.
62
Case 424/07, Commission v Germany, 3 December 2009, [2009] ECR I-11431.
63
C-2 8/15, Koninklijke KPN BV v ACM, 15 September 2016, at para 38.
416
but was limited in scope to cost accounting systems and accounting separation
(Article 7(5)).
In 2009, the Commission issued a recommendation on ‘the regulatory treat-
ment of fixed and mobile termination rates in the EU’, designed to further har-
monize NRA approaches to cost-orientation under Article 13.64 The Commission
recommends that ‘efficient costs’ should be based on current costs, the use of a
bottom-up modelling approach and long-r un incremental costs (LRIC) as the cost
methodology.65
Concerns about divergent Member State and NRA approaches to encouraging
investment in Next Generation Networks lead to the issuance of a recommen-
dation ‘on regulated access to Next Generation Access Networks (NGA)’66. The
recommendation calls upon NRAs to mandate access to a range of facilities
controlled by an SMP operator, including its ‘civil engineering infrastructure’,
particularly ducts; as well as the terminating segments of fibre-to-the-home
(FTTH), which are replacing the traditional copper segments accessed under
LLU. A further recommendation was adopted in 2013 to ensure more consistent
approaches to the imposition of non-d iscrimination obligations and the use of
costing methodologies.67
66
Recommendation 2010/572/E U, OJ L 251/35, 25 September 2010.
67
See Commission Recommendation on consistent non-d iscrimination obligations and costing method-
ologies to promote competition and enhance the broadband investment environment, COM(2013) 5761 final,
11 September 2013.
462
68
Proposal for a Directive establishing the European Electronic Communications Code, COM(2016) 590
final, 14 September 2016 (‘2016 Proposal’).
69
Ibid, at 15. 70
Ibid.
71
eg BT has announced its intention to close its legacy PSTN core network by 2025.
463
72
Regulation No 717/2007 was amended by No 544/2009, and was then replaced by No 531/2012, which has
been amended by No 2120/2015 (‘Roaming Regulations’).
73
eg Commission Implementing Regulation (EU) 2017/2 311 setting the weighted average of maximum mo-
bile termination rates across the Union, OJ L 331/39, 14 December 2017.
74
Regulation No 717/2007, at recital 4. 75
Ibid, at recital 13. 76
Ibid, at recitals 6–8 and 10.
77
Regulation No 531/2012, at recital 81 and Art 16(5). 78
Ibid, at Art 6(c).
79
Commission Implementing Regulation (EU) 2016/2286, OJ L 344/4 6, 17 December 2016, at recital 27.
80
See n 76.
64
digital economy.81 To lower the cost of such deployment, improved access to ex-
isting physical infrastructure, such as pipes, ducts, and masts, is seen as a key
facilitator. While such access could be mandated through the Access Directive
(Article 5) or the Framework Directive (Article 12), both are restricted in scope by
being limited in application to providers of electronic communication services,
preventing measures being imposed across other utility sectors with similar net-
work infrastructures.
To address this lacuna, a Directive was adopted applicable to a broad range of
‘network operators’, including gas, electricity, water, and transport services.82 The
measure requires that all network operators should have a right to offer access to
its physical infrastructure for electronic communication networks, as well as an
obligation ‘to meet all reasonable requests for access . . . under fair and reasonable
terms and conditions’ (Article 3). To facilitate such access, network operators must
provide certain information to providers of ‘public communication networks’,
either upon request or via a ‘single information point’, subject to any limitations
required for reasons of network security and integrity, national security, public
health and safety, or commercial confidentiality (Article 4).
Where new physical infrastructure needs to be built, network operators shall
have a right to negotiate agreements for the coordination of any ‘civil works’83 re-
quired to build the infrastructure. Such coordination is also seen as a means of re-
ducing the social and environmental costs associated with such works, including
pollution and traffic congestion (Recital 13). However, an obligation to meet any
reasonable request for coordination only arises where the works are being fi-
nanced in whole or part through public funds (Article 5). As with access to existing
physical infrastructure, network operators have transparency obligations in re-
spect of ‘on-going or planned’ civil works (Article 6).
Another perceived obstacle to network deployment is the number and var-
iety of different permissions that may be required to carry out any works, such as
building, planning, and environmental permits; as well as the lengthy procedures
associated with their issuance. Member States are therefore required to establish
‘single information point’ systems to make available information on the proced-
ures and conditions applicable to such permits and require that the competent
authorities grant or refuse any such permits within four months from receipt of a
completed request (Article 7).
81
Commission Communication, ‘A Digital Agenda for Europe’, COM(2010) 245 final/2, 26 August 2010.
82
Directive 2014/61/E U on measures to reduce the cost of deploying high-speed electronic communica-
tions networks, OJ L 155/1, 23 May 2014 (‘Deployment Directive’).
83
This ‘means every outcome of building or civil engineering works taken as a whole which is sufficient of
itself to fulfil an economic or technical function and entails one or more elements of a physical infrastructure’
(Art 2(4)).
456
84
This ‘means physical infrastructure or installations at the end-u ser’s location, including elements under
joint ownership, intended to host wired and/or wireless access networks, where such access networks are
capable of delivering electronic communications services and connecting the building access point with the
network termination point’ (Art 2(7)).
85
The Communications (Access to Infrastructure) Regulations 2016, SI 2016/700, at Pt 3 (‘Infrastructure
Access Regulations’).
64
Directive.86 Finally, consideration is given to the role of state aid in regulating ‘ac-
cess’ in the UK.
86
Some of the determinations and guidelines discussed in this section were published by the DGT, others
were published by Oftel, whilst those published since 29 December 2003 were published by Ofcom. To avoid
confusion, this section will refer to Ofcom in the main text, whilst referencing the actual publishing body in
the footnotes.
87
The Electronic Communications and Wireless Telegraphy Regulations 2011, SI 2011/ 1210 (‘2011
Regulations’).
88
DCMS, Implementing the revised EU Electronic Communications Framework: HMG Response, April 2011,
at para 48. Such guidance does not appear to have been forthcoming.
89
Ibid, para 59. 90
Ofcom, Infrastructure Report, 1 November 2011, at para 3.19.
91
Infrastructure Access Regulations, at reg 4 (in respect of physical infrastructure) and reg 8 (in respect of
civil works).
92
Inserted by the Digital Economy Act 2010, s 1. Ofcom is obliged to publish a report every three years (s
134A(4)), but has since committed to providing updates on an annual basis.
467
2011. 93 Addressing network sharing, Ofcom noted that it would expect any com-
pany seeking shared access to infrastructure to attempt to negotiate a com-
mercial agreement in the first instance, before it would consider regulatory
intervention;94 which is consistent with its approach to handling disputes in
general. 95
Ofcom reported in 2011 and 2014 that there had been limited sharing of passive
infrastructure in respect of access networks, which is expected to change with the
implementation of the Deployment Directive.
93
Ofcom, ‘Infrastructure Report’, 1 November 2011. It was renamed the ‘Connected Nations Report’ in 2015.
94
Ibid, at para 3.18. 95 See Chapter 3, at 3.3.7.3. 96 See further Chapter 6.
97
See Part 1 of the General Conditions and General Condition 1.4.
98
Oftel, Guidelines for the interconnection of public electronic communications networks (2003).
99
Ibid, para 4.8.
648
Ibid, Chapter 6.
100
469
103
Oftel, Market review guidelines: criteria for the assessment of significant market power (2002).
104
Oftel, ‘Imposing access obligations under the new EU Directives’, 13 September 2002, <http://
webarchive.nationalarchives.gov.uk/20080714072725/http://w ww.ofcom.org.uk/static/a rchive/oftel/publi-
cations/i nd_ g uidelines/acce0902.htm>.
740
operator to ‘meet all reasonable requests for access’. The Access Guidelines state
that Ofcom is likely to consider that a request which is technically feasible is
‘reasonable’ if the SMP operator can reasonably expect to receive at least a rea-
sonable rate of return on any necessary investments when the access product
is supplied at a price the requesting operator is willing to pay. Only in ‘extreme
examples’ should a request for access be denied on the basis that the request is
unreasonable.105
New products and innovative products The Access Guidelines also provide guid-
ance on the situation arising when a competing operator demands a new whole-
sale product or where products become available because of innovation on the
part of the SMP operator.
In the case of a demand on an SMP operator to make a new or untested whole-
sale access product available to a competitor, it can be difficult to determine
whether demand for the product will materialize. It is therefore difficult to de-
termine whether the SMP operator can expect a reasonable rate of return on the
investment that they will make, which is Ofcom’s test for whether a request is ‘rea-
sonable’. The Access Guidelines state that if the SMP operator will incur signifi-
cant development costs in supplying a product for which demand is uncertain, the
requesting operator may be required to take on an appropriate level of risk. This
could involve:
Ibid, at 13.
105
Ibid.
106
741
The risk with this approach, obviously, is that SMP operators will be
disincentivized from investing in the development of new services, because
they will be required to share the results of their innovation with their com-
petitors, rather than being able to gain a competitive advantage and increased
market share by being ‘first to market’. The Access Guidelines propose that
this problem should be dealt with by allowing SMP operators to impose suf-
ficiently generous terms in the supply of innovative wholesale products to
other operators. Where a new or innovative product involves a high level of
risk, cost-based price controls will normally be avoided, even if the SMP op-
erator has very high market share. In such markets, either no charge control,
or a retail minus form of regulation may be more appropriate. A retail minus
pricing model would in this case allow an element of supernormal profit to be
built into the retail price to be retained by the SMP operator. Setting any kind
of cost-based charge control risks distortion of commercial and investment de-
cisions and discouragement of innovative market offerings, ultimately to the
detriment of consumers.107
Access to information protected by intellectual property rights The Access
Guidelines state that if information which is protected by intellectual property
rights is essential to allow competitors to the SMP operator to offer a competing
product, the SMP operator would be expected to make the information available.
The operator requesting the information would be expected to demonstrate that it
is indeed essential.108
Ibid, 14, 33–35.
107
Ibid, 35. Note that this rule does not apply to standard network interfaces, which must be made available
108
Imposition of transparency obligations The Access Guidelines state that any
new wholesale product offered by an SMP operator will normally need to be pub-
lished in the form of a reference offer. Initial reference offers for new products,
and changes and updates to a reference offer for an existing product, should be
released in a timely manner, allowing enough time for a reasonably efficient op-
erator to make necessary preparations. Information (including terms, conditions,
and prices) must be supplied to any downstream business at the same time that it
is released to the market. Sufficient information should be given at the time of or
before the launch of a product to enable competitors to make full and effective use
of the product supplied, although the disclosure of such information can be made
subject to a confidentiality agreement. A list of the minimum information that
must be included in a reference offer, as well as its applicability and availability, is
now set out by Ofcom in respect of each market review.114
111
Ibid, 24. 112
Ofcom, ‘Undue discrimination by SMP providers’, 15 November 2005.
113
Access Guidelines, at 16–17 and 30–31.
114
eg Ofcom statement, ‘Business Connectivity Market Review— Volume 1’, 28 April 2016, at paras
8.110–8.124.
743
before the Competition Appeals Tribunal (CAT), which has quashed its decisions
on both geographic and product market definitions,119 and further appeals from
the CAT.120
The size of this task should be appreciated. The UK market reviews total many
thousands of pages of detailed analysis. A detailed analysis of each of the market
reviews is beyond the scope of this chapter.121
119
eg British Telecommunications plc v Ofcom [2017] CAT 25.
120
eg Hutchison 3G (UK) Limited v Office of Communications [2009] EWCA Civ 683.
121
See further Chapter 2, at Section 2.15.1.6 et seq.
122
DGT, National Roaming Condition, A consultation on proposals to set a national roaming condition after
25 July 2003 (2003), 4.
123
DGT, End-to-end connectivity; Guidance issued by the Director General of Telecommunications (2003).
745
operators whenever one of their customers wants to reach a user or service on that
other network, and positively required to ensure that they terminate any call re-
ceived onto their network. These obligations would have gone beyond the obliga-
tion to negotiate interconnection on request, which all operators are required to
do under Condition 1 of the General Conditions of Entitlement.
Ofcom concluded that the imposition of obligations to ensure end-to-end con-
nectivity was not appropriate, for several reasons. In considering imposing an ob-
ligation to purchase call termination from other operators, Ofcom considered that
the imposition of such an obligation on the universal service providers (that is,
BT and KCom) would be disproportionate. This is because those operators must
in any case meet reasonable requests for access to publicly available telephone
services, which, it is implied, includes being able to contact other customers and
services, irrespective of the terminating network. For operators other than BT and
KCom, Ofcom considered that the commercial incentives to provide end-to-end
connectivity were sufficiently strong to ensure that they seek to purchase call ter-
mination without any additional obligation to ensure that they do. This is clearly
correct, as it is almost unthinkable that an operator would seek to set up a new ser-
vice that did not allow customers to contact users and services on other networks.
Ofcom considered that it was not necessary to impose any additional obligation
on any operator to provide call termination services to other operators because
almost all public electronic communications networks are already under an SMP
condition requiring them to provide call termination to all other public electronic
communications networks on fair and reasonable terms.
124
Mercury, trading as One-2-One, appealed the decision to impose the condition and was initially suc-
cessful (Moses J, QBD, 6 August 1999); although it was overturned in Mercury Personal Communications Ltd v
Secretary of State for Trade & Industry [2000] UKCLR 143.
476
out their 3G network, whereas a new entrant would not have this advantage, and
would not be able to compete.
With the abolition of telecommunications licences in July 2003, Ofcom had
to consider whether to re-impose national roaming obligations on 2G operators
under the new regulatory regime. The initial conclusion of its predecessor, Oftel,
was that all four of the UK’s 2G operators should be subject to a new ‘access-related
condition’, under the Communications Act 2003, ss 73–74, requiring them to pro-
vide national roaming on fair, reasonable, and not unduly discriminatory terms.125
Instead, Ofcom decided to issue continuation notices to O2 and Vodafone, pending
a further consultation in July 2004. In this consultation, Ofcom proposed that any
access condition be removed in favour of ‘less intrusive regulation’, on the grounds
that there was sufficient commercial interest in the offering of national roaming to
3 and Ofcom’s ability to intervene to resolve any disputes if they arose.126 However,
3 asked Ofcom to delay the withdrawal of the condition until it had re-tendered for
the roaming contract, which it successfully completed in 2006, signing a contract
with Orange for national roaming outside the 88 per cent coverage it had already
built. The condition has since been withdrawn.
formal notification pursuant to section 48(1) of the Communications Act 2003 (2003).
74
access services. Ofcom carried out a market review of ‘wholesale digital platform
services’ in 2006, as well as issuing guidance on how it interprets the obligations
on SSSL.128 In 2015, Ofcom let the access control obligations lapse, on the grounds
that SSSL had made voluntary commitments that achieve the same outcomes,
while Ofcom retained the necessary powers to intervene if required.129
8.5.7 Broadband UK
As noted previously, much of the focus on ‘access’ issues is currently driven by a
policy concern to promote the roll-out of NGNs, in the UK as well as other jurisdic-
tions. In addition to providing market participants and new entrants with regu-
lated access incentives, however, governments are also directly intervening in the
market through the public funding or subsidization of NGN roll-out by network
operators. Such schemes impact on access and interconnection through the con-
ditionality imposed on the receipt of such state funding. State aid may, in itself,
create competition problems and is therefore regulated at an EU level;133 but in
terms of access, governments will generally impose pro-access contractual obli-
gations upon any undertaking that receives public monies.
128
Ofcom, ‘Review of Wholesale Digital Platform Services’, 10 October 2006 and ‘Provision of Technical
Platform Services—Guidelines and Explanatory Statement’, 21 September 2006.
129
Ofcom, ‘Review of Sky’s Access Control Services Regulation’, 17 March 2015.
130
Section 185A was inserted in 2003, enabling Ofcom to invite parties to refer a dispute.
131
Communications Act 2003, s 104(4). 132
Ofcom, Dispute Resolution Guidelines (7 June 2011).
133
Commission Communication, EU Guidelines for the application of State aid rules in relation to rapid
deployment of broadband networks, OJ C 25/1, 26 January 2013.
748
In the UK, the Department for Culture Media and Sport has established a team,
Broadband Delivery UK, to allocate and distribute public funds designated for
broadband roll-out in rural areas.134 Local authorities can submit plans and apply
for a share of the monies; which, if allocated, the work would then be put out to
tender to potential suppliers.
Every time that access arrangements are entered into there should, of course, be
a contract in place setting out the parties’ rights and responsibilities. This section
will examine some of the practical and contractual issues that are likely to arise
when negotiating arrangements for access to parts of the public switched network.
IP interconnection agreements are discussed in Section 8.7.
Access agreements are by no means a generic set. For example, a complex agree-
ment to establish a mobile virtual network135 will have little in common with a
basic agreement to interconnect two networks. One factor that does act to distin-
guish between different kinds of access agreements, however, is whether either
party has SMP in the relevant market and, in particular, whether an SMP party is
required to publish a reference offer for the access that is sought. This section will
analyse first some aspects of those access arrangements not subject to a reference
offer, with a particular focus on interconnection agreements. Some special con-
siderations relevant to reference offers will then be examined.
134
See <https://w ww.gov.uk/g uidance/broadband-delivery-u k>. See also Commission Press Release,
Commission endorses UK National Broadband Scheme for 2016–2020, IP/16/1904, 26 May 2016.
135
A mobile virtual network agreement gives one operator, usually with limited infrastructure of its own,
and without a licence for radio spectrum, the right of access to parts of the network of a mobile operator in
order to provide mobile telephony services to its own customers. See further Chapter 11, at Section 11.2.5.
749
8.6.2. The new entrant may then seek direct interconnection agreements with
other operators.
Interconnection agreements are usually bilateral; that is, they govern the terms
on which each party will terminate traffic onto the other party’s network. Each
party is usually subject to almost identical obligations, including identical war-
ranties, and the same exclusions and limits on liability. For this reason, bilateral
interconnection negotiations are often fairly harmonious. The parties should still
ensure, however, that the contract gives them the legal protection they need. If
it is unclear or poorly drafted, it will be of no help to the parties in the event of a
later contractual dispute that the initial contract negotiations were not difficult.
Interconnection agreements will, of course, contain many of the terms that you
would expect to see in any commercial agreement, including provisions setting
out payment arrangements, confidentiality, limitations and exclusions of liability,
and provisions relating to whether the agreement can be assigned or transferred.
The following sections describe some provisions that are particular to intercon-
nection agreements.
determined as having SMP in the market for the termination of calls onto their re-
spective networks.136 Ofcom’s approach in imposing SMP conditions in respect of
call termination services has varied according to the structure of related markets,
and whether each operator possesses market power in the related market for call
origination. As a result, in the mobile market Vodafone, O2, T-Mobile, and Orange
were required to comply with charge controls in respect of their 2G call termin-
ation services. In the fixed market, BT is required to base its call termination
charges on efficiently incurred long-r un incremental costs, reducing each year
in line with charge controls, and each other operator providing call termination
services is required to provide such services on terms, conditions, and charges
that are fair and reasonable.
New-entrant fixed telephony operators must, therefore, only levy fair and rea-
sonable termination charges. However, as a dispute between BT and Telewest
has demonstrated, in practice Ofcom requires that charges for fixed geographic
call termination are calculated on the basis of ‘reciprocal charging’.137 This means
that fixed geographic call charges will be calculated according to a formula based
on BT’s regulated charges. There is some room for the charges to vary if there are
relevant differences between the terminating network and BT’s network, but in
practice reciprocal charging usually means that each party’s termination rates are
identical.138
136
See further Chapter 2, at Section 2.15.2.
137
Ofcom, Resolution of a dispute between BT and Telewest about reciprocal charging arrangements for call
termination rates (16 April 2004).
138
See Ofcom, ‘Determination to resolve dispute between Opal Telecom and BT about Opal’s fixed geo-
graphic termination rates’, 29 October 2009.
418
8.6.2 Reference offers
Reference offers are standard contracts setting out the terms on which an operator
will enter into access arrangements. As noted above, NRAs are empowered by the
Access Directive to require operators with SMP in a given market to publish a ref-
erence offer for network access.
As the largest SMP operator in the UK, BT is required to publish reference offers
for a large number of different services. Rather than publishing a separate agree-
ment for each different type of access, BT historically published a small number
of agreements with schedules for each different service. However, with the oper-
ational separation of BT and the establishment of Openreach, each SMP product
now has a distinct reference offer.139
Regulators like reference offers because they can see exactly what terms an SMP
operator is offering. The other principal advantage is that they eliminate the pos-
sibility of the SMP operator unduly discriminating in the terms it offers different
operators, because the terms are identical. For this reason, the terms set out in ref-
erence offers are usually not open to negotiation.
139
Available from <www.openreach.co.uk>.
482
and paying transit arrangements. A new-entrant ISP is likely to start with one or
more paying transit arrangement to achieve worldwide connectivity in one step,
and then to pursue peering arrangements with local ISPs once it has established
its business, in order to reduce costly transit charges.
Although some descriptions of internet industries give an impression that they
are unregulated (and unregulatable!), IP interconnection falls within the defin-
ition of interconnection under the Access Directive, and, as such, is regulated in
the same manner as other forms of interconnection.140 As a consequence, those
providing a public electronic communications network (which would catch all
publicly available ISPs) must generally negotiate IP interconnection with each
other on request. IP interconnection agreements are also subject to the same dis-
pute resolution mechanisms as other interconnection agreements, as set above.
140
However, the Commission has not identified ‘wholesale internet connectivity’ as a market for the pur-
pose of ex ante market analysis. See Commission Staff Working Document, accompanying Recommendation
2007/879/EC, SEC(2007) 1483/2 rev1, at 37.
141
In circuit- s witched interconnection, similar such arrangements are generally referred to as ‘bill
and keep’.
142
Norton, WB, Internet Service Providers and Peering (2003), 3. Available at <http://c seweb.ucsd.edu/
classes/w i01/c se222/papers/norton-isp-d raft00.pdf>.
438
potentially be faster and more reliable, resulting in lower ‘latency’, meaning that
less packets of data are lost.
It is obviously in ISPs’ interest to ensure faster traffic consumption, particularly
if they are billing their users based on the amount of data downloaded!
Peering is not always the right solution, however. The most common reason why
parties will not peer is that the traffic flow between them is asymmetrical, and one
party will therefore bear a greater proportion of the cost of peering.143 This is not
only a question of the respective size of the networks, but also of whether the net-
works are content-r ich. A network that is content-r ich will have a small amount of
inbound traffic (such as in the form of requests for data on the websites it hosts),
but will generate a large amount of outbound traffic (in the form of the content
from those websites being sent to the network from which the request was gen-
erated). The relative bargaining position of the parties will in this case influence
whether a peering arrangement or a paying transit arrangement is established.
In considering a potential peering partner, an ISP is therefore likely to examine
how much incoming traffic on its network originates from the potential peer, and
how much outgoing traffic is addressed to the potential peer.144 Calculations are
then made to assess whether peering is likely to reduce the cost of the transit be-
tween the two networks. Peering will require investment in router capacity and
interconnection circuits to carry traffic to the point of interconnection, so will only
be justified if significant savings in transit costs will follow.
Some ISPs, particularly large ones, have peering policies which are freely avail-
able.145 Any ISP that meets the criteria can apply to become a peering partner
of that ISP. Backbone ISPs’ peering policies can include requirements that the
peering partner has presence in four or more regions where both parties have a
presence, along with sufficient transport bandwidth and traffic volume to warrant
direct interconnection.146
Once the parties have decided to establish peering arrangements, they will
enter into negotiations on the contractual terms that will govern the relation-
ship. Whilst some peering agreements run to hundreds of pages, most are very
short and informal documents compared with typical switched interconnection
For further discussion on the advantages and disadvantages with peering, see further ibid, 3–5.
143
Internet traffic carries in it data that indicates which ISP the traffic originated from (‘originating autono-
144
mous system’ or ‘originating AS’) and is destined for (‘terminating autonomous system’ or ‘terminating AS’).
An ISP may therefore sample their inbound and outbound traffic and determine approximately how much of it
originated from another ISPs’ network (in the case of inbound traffic) or is bound for routers on the other ISP’s
network (in the case of outbound traffic).
145
For example, ‘Verizon Business Interconnection Policy for Internet Networks’, available at < http://w ww.
verizonenterprise.com/terms/peering/>.
146
See n 142, 3–4.
84
147
BEREC Report, ‘An assessment of IP interconnection in the context of Net Neutrality’, BoR (12) 130, 6
December 2012, notes, ‘99% of interconnection arrangements are concluded on a handshake basis’!
458
interconnection circuit between its network and the exchange in order to peer
with many other networks. Some internet exchanges also have standard bilateral
agreements on which their members contract, which can significantly simplify
negotiations.148
Peering at a public internet exchange will not always be viable or desirable,
however. For remote networks, the closest exchange may be far from any of the
operator’s network nodes,149 or an operator may anticipate having few regional
peering partners, meaning that peering at the exchange does not result in bene-
fits from economies of scale. In these cases operators will enter into arrangements
to peer at private peering points. Private peering points give the parties greater
control over the interconnection, and, accordingly, much greater control over the
quality of the service that can be expected.
Private peering is typically arranged by each party obtaining co-location services
at a telecommunications exchange, and then establishing interconnection be-
tween the networks. The parties may find that they have points of presence150 in
common exchanges already, in which case establishing the physical interconnec-
tion can be achieved very quickly. If peering points are needed at further exchanges
where the parties do not have points of presence, then they will need to agree
which exchange(s) best suit their needs, and approach those exchange(s) to obtain
co-location.
Many cost and operational issues will influence the decision when choosing an
exchange, including whether competitively priced interconnection circuits are
available between the parties’ respective networks and a particular exchange.
It is not uncommon for large networks to establish peering at a variety of public
internet exchanges and private peering points. Some peering agreements pro-
vide that the parties are required to investigate moving the location of a point of
interconnection from a public internet exchange to a private peering point if and
when the volume of traffic exchanged over the point of interconnection exceeds a
certain level.
This is intended to give operators greater control over the quality of service at
those points of interconnection which carry the heaviest traffic.
148
See, for example, the LINX bilateral interconnect agreement at <https://w ww.linx.net/about/good-of-
the-i nternet/bcps/i nter-peer-technical-resolution>.
149
A network node describes a point in the network at which interconnection can be established.
150
A point of presence is a point in the network from which users are connected.
486
into a peering relationship and the peering partner subsequently fails to meet the
criteria? For example, many operators will only be looking to peer with networks
where the traffic flow between the two networks is relatively equal. However, the
ratio of traffic transmitted from one network to the other may change over time,
for example if one network operator develops its hosting business and becomes a
net exporter of content.
Peering agreements sometimes deal with this by setting a traffic ratio (eg 4:1
outbound traffic to inbound traffic). The parties agree to peer (without settle-
ment) up to the ratio, beyond which they must pay the other party, usually for
each Megabyte of outbound traffic. This arrangement is a form of paid transit
arrangement, discussed further. The issue of traffic ratios was brought into
focus in 2003 in a dispute in the US between America Online (AOL) and Cogent
Communications Group. AOL offered peering when the traffic ratio was no
more than 2:1 and when the ratio with Cogent reached 3:1, AOL terminated
the peering connection. In enforcing its traffic ratio criteria, America Online
began to charge Cogent for traffic that had previously been exchanged free of
charge.151
Another example where peering criteria may be enforced is where one operator
requires the other’s network to have certain minimum characteristics. If the char-
acteristics are not met, then sometimes transit charges are payable, such as if a
minimum threshold for packet loss is exceeded.
Any provisions in a peering agreement that can potentially change the nature
of the relationship to a paying transit relationship should be closely reviewed by
legal advisers.
151
See Noguchi, Y, ‘Peering dispute with AOL slows Cogent customer access’, Washington Post, 27 December
2003, available at <http://legalminds.lp.findlaw.com/l ist/c yberia-l/m sg42080.html>.
487
152
A detailed discussion of the series of events in the US in this period is included in Cukier, KN, ‘Peering
and fearing: ISP Interconnection and Regulatory Issues’, <http://w ww.cukier.com/w ritings/peering-c ukier-
dec97.html>.
153
Both third party CDN service providers, such as Akamai, and self-provision CDNs, such as Netflix Open
Connect.
84
address latency concerns for sensitive traffic, such as streaming video services, as
well as greater regional peering, enabling circumvention of transit provided by the
tier 1 backbone providers.154
Many ISPs find that they must pay their upstream internet access provider or back-
bone provider a fee, often called a ‘download fee’ for inbound traffic received over a
point of interconnection, whether the traffic has originated on a third party network
or the internet access provider/backbone provider’s own network. The download fee
will be charged, for example, when a user on the downstream ISP downloads a web
page hosted on the backbone provider’s network or on any network from which the
backbone provider has agreed to provide transit.
Some internet access providers/backbone providers also charge a fee, some-
times called a ‘backchannel fee’, for data received onto their network from the
downstream network. A backchannel fee will be charged to the downstream net-
work, for example, when a user on the backbone provider’s network, or on any net-
work to which the backbone provider has agreed to provide transit, downloads a
web page hosted on the downstream network.155
ISPs can find, therefore, that they are paying for both inbound and outbound
traffic carried by their upstream access provider. Although some content-r ich net-
works are able to negotiate a more favourable position, many ISPs cannot. These
arrangements have caused some disquiet amongst small and medium sized ISPs.
Unsurprisingly, in paying transit arrangements customers look for more detailed
and more onerous contractual terms from their transit providers than they do
from their peering partners. Some of these terms will be similar to those described
for peering agreements, above. Some of the terms that are likely to differ are con-
sidered in the following sections.
See n 152, at 58.
154
A good explanation of the payment arrangements can be found in the ACCC discussion paper ‘Internet
155
8.7.2.3 Service levels
Detailed schedules are likely to specify service level guarantees and may spe-
cify service credits or liquidated damages in the event that the service levels are
not met.
8.7.2.4 Charges
Much detail is likely to be dedicated to how the charges payable are to be cal-
culated, invoiced, and paid. Numerous different models may be used to calcu-
late transit charges, including on a per byte basis or on a port basis (ie a flat-rate
charge). Discounts may be applied based on volume or the perceived value of the
content hosted on the customer network.
The interconnection and access regime does not have a significant impact on inter-
connection and access arrangements between two operators who do not possess
SMP. Although the regime requires them to negotiate interconnection with each
other, there are commercial incentives for them to do so in any case. In this respect
the regime may be said to merely reinforce rational market behaviour.
However, it is apparent that competing operators in the EU continue to rely
heavily on the existence of sector-specific rules, particularly in the form of ex ante
conditions and directions on network access, to obtain access rights from oper-
ators with SMP. This appears to be as much the case under the Access Directive
as under the Interconnection Directive before that. Although there is no doubt
that general competition law would prohibit the refusal to supply access in many
cases,156 it seems unlikely, for example, that general competition law would have
resulted in competing operators obtaining rights access to FRIACO interconnec-
tion, ATM interconnection, and to wholesale line rental.
However, the gradual erosion of the market shares held by incumbents, and the
emergence of new technologies in which they do not have a stranglehold, such as
voice-over IP, is likely to prompt incumbents to argue for the retreat of the sector-
specific rules. This is an ongoing battleground between incumbent operators and
their competitors.
Since the last edition of this book, there has arguably been a shift in focus in favour
of consumer protection in the telecommunications industry both at EU and at UK
level, which is illustrated through two developments. First, in 2016, the European
Commission published its proposal for an Electronic Communications Code to co-
dify and amend the four main telecommunications directives, which for the first
time explicitly takes full account of the Charter of Fundamental Rights of the EU.1
It states that the proposed measures aim at achieving higher levels of connect-
ivity ‘with a modernised set of end-user protection rules’. It mandates end-user
rules where existing regulation has only set out minimum requirements. Second,
in the UK in May 2017, the consumer association Which? awarded Ofcom Chief
Executive, Sharon White, who took up the position in March 2015, the Positive
Change Award for her work on putting consumers at the heart of Ofcom’s agenda.2
1
Proposal for a Directive establishing the European Electronic Communications Code, COM(2016) 590
final, 14 September 2016 (‘2016 Proposals’).
2
Which? Press Release, 17 May 2017.
492
This chapter looks at why special consumer protection measures exist for
telecommunications services, over and above the consumer protection meas-
ures that apply to other products and services more generally. The answer is
twofold.
9.1.1 Utility
First, some telecommunications services have become analogous to public util-
ities: like gas, electricity, water, and sewage, they are considered to be so important
to people’s lives that measures have been put in place to ensure that people have
access to them and are not prevented from using them. In the EU, the Universal
Service Directive3 ensures provision of a ‘universal service’: an affordable basic
telephony service available to everyone.4 In the UK, the Universal Service Order5
sets out the minimum requirements for the universal service, and has been im-
plemented through the General Conditions of Entitlement (GCs) and through
specific conditions on BT and KCOM, who are designated as the universal service
providers in the geographic areas they cover.6 Across the EU, the universal service
obligation has historically been limited to fixed-l ine voice services.7 However, the
European Commission’s 2016 proposals to amend the telecommunications regu-
latory framework included an extension of the universal service obligation to basic
broadband (defined on the basis of a minimum list of online services needed to
enable end-users to participate in civil society 8). The Commission has also pro-
posed removing the obligation in relation to some older technologies (payphone
provision, comprehensive directories, and directory enquiry services9). Under the
proposals, Member States will have flexibility to extend affordability measures to
mobile as well as fixed line.10
3
Directive 2002/22/EC of the European Parliament and of the Council of 7 March 2002 on universal ser-
vice and users’ rights relating to electronic communications networks and services, OJ L 108, 24 April 2002.
4
Ibid, Art 1(2). See further Chapter 4, at Section 4.8.
5
Electronic Communications (Universal Service) Order 2003, SI 2003/1904.
6
Designation of BT and Kingston as universal service providers, and the specific universal service condi-
tions: Statement and Notification issued by the Director General of Telecommunications on the implementa-
tion of the Universal Service Directive, 22 July 2003.
7
The European Commission concluded in 2011 that it was not appropriate to extend the obligation to
broadband. See, European Commission Communication, COM(2011) 795.
8
Defining functional broadband by way of a list was criticized by the European Economic and Social
Committee (EESC) in its opinion on the proposals (2017/C 125/56), as it created an arbitrary list of accessible
internet services, as opposed to a neutral minimum quality link and might give rise to discriminatory prac-
tices detrimental to end-u sers.
9
Removing the obligation in respect of older technologies was also criticized by the EESC (ibid).
10
2016 Proposals, at Arts 79–86.
493
9.1.2 Competition
The second reason why special consumer protection measures exist for telecom-
munications is that, since the end of monopolies in the sector, operators have
been regulated to provide for a competitive sector. The introduction of competi-
tion is seen as being generally beneficial to consumers in terms of choice, cost,
and quality. In a fully competitive market, it could be argued that there would be
no need for sector-specific consumer protection rules, because the availability of
choice means that in theory a disgruntled subscriber could simply switch to a pro-
vider offering a better service, and the availability of alternative services means
that each communications provider has to have an eye to its competitors and en-
sure that its subscribers remain happy enough with their service that they do not
want to switch provider.
So, concerns about consumers could be seen as being an issue only during the
process of liberalization, before markets are fully competitive and while oper-
ators with ‘significant market power’ continue to be prevalent. Indeed, when the
European Commission first proposed the Universal Service Directive in 2000, it
considered that as competition continued to develop it was likely that the con-
sumer protection environment would become more homogenous among the ex-
isting EU Member States and that regulation of consumer protection would not
be so necessary. It based its inclusion of consumer protection measures on the
prospect of EU enlargement, which was expected to introduce a wide range of na-
tional differences, and make it necessary to ensure regulatory intervention to up-
hold a minimum set of consumers’ rights throughout the Community.11
The 2002 Universal Service Directive included consumer protection measures
to increase the ability of consumers to optimize their choices and so benefit fully
from competition.12 In fact, consumer protection issues remain a central con-
stituent of the EU regime now, and further measures to strengthen and improve
such protections have been included in subsequent amendments and in the latest
proposed amendments in 2016.
The continued inclusion of consumer protection measures in the light of com-
petition law is justified in two ways. First, competition alone may not be enough to
satisfy the needs of all citizens and protect users’ rights. Additional protections are
needed, both in the form of the universal service (for more on the universal ser-
vice, see Chapter 4 European Union Communications Law, Section 4.8), and in the
form of consumer protection laws that help to balance the respective bargaining
11
European Commission Proposal for a Directive of the European Parliament and of the Council on uni-
versal service and users’ rights relating to electronic communications networks and services, COM(2000) 392
final 2000/0183(COD).
12
Universal Service Directive, Recital 30.
94
positions of consumers and the companies with whom they contract. Secondly,
consumer protection law also plays a role in stimulating competition. Most EU
measures are focused on stimulating competition from the supply-side, but con-
sumer protection measures help to stimulate competition from the demand-side.
In order to create demand, consumers need to be educated about the services on
offer. Even in a competitive environment, consumers who purchase telecommu-
nications services are likely to know less about the product or service than the sup-
plier, be required to contract on the supplier’s terms, and possibly vulnerable to
pressure to buy.
13
See definition of ‘consumer’ in the GCs. In the Enterprise Act 2002, ss 129 and 183, a consumer is a person
to whom goods and services are supplied (whether by sale or otherwise).
14
See, for example, the definitions of ‘subscriber’ in art 2 of the Electronic Communications (Universal
Service) Order 2003/1904, in Regulation 2 of the Privacy and Electronic Communications (EC Directive)
Regulations 2003/2 426 and in the GCs.
15
See definition in GCs. 16
See also the definition in the GCs in force from 1 October 2018.
17
GC C1.3 (GC9.3(a) in the version of the GCs that applies up until 1 October 2018).
18
GC C4 (GC14 in the version of the GCs that applies up until 1 October 2018).
19
See definition in GCs.
495
9. 2 EU PROV ISIONS
Directive 2002/22/EC on universal service and users’ rights relating to electronic communications networks
and services, Directive 2002/58/EC concerning the processing of personal data and the protection of privacy
in the electronic communications sector, and Regulation (EC) No 2006/2004 on cooperation between national
authorities responsible for the enforcement of consumer protection laws, OJ L 337, 18 December 2009.
23
Regulation 2015/2120 of the European Parliament and the Council laying down measures concerning
open internet access and amending Directive 2002/22 on universal service and users’ rights relating to elec-
tronic communications networks and services and Regulation 531/2012 on roaming on public mobile com-
munications networks within the Union, OJ L 310/1, 26 November 2015 (‘2015 Regulation’).
24
Respectively, Directive 88/301, Art 7 (minimum notice one year) and Directive 90/388, Art 8 (minimum
notice six months).
946
However, the European Court of Justice annulled these provisions on the basis
that such private contractual arrangements were not ‘State measures’ to which
Article 86(3) of the then EC Treaty was applicable.25 Subsequently, provisions
governing subscriber contracts and QoS issues were introduced under the ONP
framework.26
In 2002, Article 20 of the Universal Service Directive introduced a requirement
for a clear set of minimum contract terms to be included in contracts for connec-
tion to the public telephone network and other services, because contracts are an
important tool for consumers to ensure a minimum level of transparency of infor-
mation and legal security.27 These minimum terms include:
Subscribers also have a right to withdraw from contracts without penalty on no-
tice of proposed modifications in the contractual conditions, and must be given at
least one month’s notice.
The amendments made by the Citizens’ Rights Directive strengthened these re-
quirements.28 In particular, amended Article 20 sets out more detailed require-
ments of what should be included in the description of the services.
It must be clear whether or not the service allows access to emergency services
and whether caller-location information is available, and whether there are any
limitations on the provision of emergency services. This is particularly relevant for
VoIP services, as mobility and location independence mean that, unlike during a
PSTN call, it may be difficult to locate the user when an emergency call is made,
25
Case C-202/8 8: France v Commission [1992] 5 CMLR 552; and Case C-271/9 0 Spain v Commission [1992]
ECR I-5833. Article 86 is now Article 106 of the Treaty on the Functioning of the European Union.
26
See Chapter 4, at Section 4.5. 27
Universal Service Directive, Recital 30.
28
Citizens’ Rights Directive, Art 1(14).
497
meaning that calls may not be directed to the nearest emergency service call
centre. With some VoIP services, emergency calls cannot be made at all.29
Service description must also include information on any other conditions
limiting access to or use of services and applications. This would include any caps
on bandwidth or connection speed.
Minimum service quality levels must also be given. This includes the time for
the initial connection, and any other QoS parameters required by the national
regulatory authority (NRA). This provision backs up NRA powers to impose min-
imum QoS requirements on communications providers. For more on this, see
Sections 9.2.3 and 9.3.1.1.
Under Article 4(1) of the 2015 Regulation, end-user contracts with IAS providers
must include information on any traffic management measures that could impact
on the quality of the IAS, the privacy of end-users, and on the protection of their
personal data. Explanations of the following must also be included:
• How any volume limitation, speed, and other QoS parameters may impact on
IAS and, in particular, on the use of content, applications, and services.
• How other services to which the end-user subscribes might impact on the IAS.
• Minimum and maximum upload and download speeds, and how significant de-
viations from advertised speeds could impact the end-user.
• Remedies available in the event of any continuous or regulatory recurring dis-
crepancy between the contracted and actual performance of the IAS.
29
However, most VoIP services would not consider themselves to be ECSs. The linking of interper-
sonal communications services (as VoIP services are defined under the 2016 Proposals for an Electronic
Communications Code) with the use of public numbering resources has been contested by service providers.
30
See further Chapter 15. 31
2016 Proposals, at Art 95(4).
948
the data concerned.32 Details of prices must include details of payment methods
offered and any differences in costs due to payment method.
Contracts must also include details of:
Some of these issues are discussed in more detail in the section on the UK im-
plementation of these provisions (Section 9.3 below).
Lastly, contracts must also include the type of action that might be taken by
the undertaking in reaction to security or integrity incidents or threats and
vulnerabilities.
The European Commission’s 2016 proposals for an Electronic Communications
Code have shifted the emphasis for end-user rules. Rather than providing a min-
imum harmonization approach, they mandate end-user rules, making these sub-
ject to full harmonization to the extent possible. Member States must not introduce
more or less stringent provisions for end-user protection, except where specified.33
This approach is in line with the aims of the Commission’s Digital Single Market
Strategy,34 which aims to increase the ability of individuals and businesses to ac-
cess services seamlessly across national borders within the EU by removing the op-
portunity for different consumer protection rules to evolve in each Member State.
However, the European Economic and Social Committee has expressed doubts
about the maximum harmonization approach in the context of end-user rights.35
In the light of the increasing number of software- based communications
services and uncertainty about whether the rules apply to those services, the
Commission is proposing to redefine ‘electronic communications service’ so that
it applies to:
• IASs.
• Interpersonal communications services. This term is intended to catch
over-t he-top (OTT) software-based applications where the service enables a
32
See further Chapter 13.
33
eg the proposals on contract duration (in Art 98 in the first draft) are not subject to maximum harmon-
ization, so providers can in certain circumstances agree contract periods longer than the general maximum
of two years: 2016 Proposals, at Art 94.
34
Communication from the Commission to the European Parliament, the Council, the European Economic
and Social Committee and the Committee of the Regions: A Digital Single Market Strategy for Europe, COM
(2015) 192 final.
35
Opinion of the European Economic and Social Committee on the proposal for a directive of the
European Parliament and of the Council establishing the European electronic communications code (Recast)
(COM(2016) 590 final, 2016/0288 (COD)) (2017/C 125/56).
94
36
See further Chapter 4, at Section 4.2. 37
2016 Proposals, at Art 2(4).
38
2016 Proposals, at Art 95.
39
Directive 2011/8 3/E U on consumer rights, OJ L 304/6 4, 22 November 2011.
50
9.2.2 Transparency
Provisions that require communications providers to publish information about
their services enable end-users and consumers to make informed choices about
the services they plan to purchase, and back up the provisions on what must be
included in contracts.43 The transparency provisions also provide some protection
for those end-users who are not also subscribers and so do not benefit from the
requirements for certain information to be made available in contracts. Article 21
of the Universal Service Directive introduced a number of transparency require-
ments, which reflect the requirements for information to be provided in contracts
under Article 20.
Article 21 of the Directive, as originally drafted, required member states to en-
sure that transparent and up-to-date information on applicable prices and tariffs,
40
Directive 2002/58/EC (the European Commission proposed in January 2017 replacing this Directive with
an E-P rivacy Regulation). See further Chapter 13.
41
Micro and small enterprises are a category of small and medium enterprises as defined in Commission
Recommendation of 6 May 2003 concerning the definition of micro, small and medium-sized enterprises
(2003/361/EC, L124/36). A microenterprise is defined as an enterprise that employs fewer than ten people and
whose annual turnover and or annual balance sheet total does not exceed EUR 2 million.
42
Regulation 531/2012, Art 15(3) and Regulation 531/2012, Art 15(2a) as amended by Regulation 2015/2120.
43
Universal Service Directive, Recital 30.
510
A particularly important change here was the requirement that obliges oper-
ators to inform consumers, before contracting, of any service restrictions, which
would include caps on bandwidth or connection speed.45
Also, price comparison continued to be a concern in 2007 so, under the amend-
ments, where comparator guides are not made available by the market, NRAs are
now obliged to make them available, either themselves, or through third parties;
and, to support this, third parties have a right to use published information free of
charge to provide such guides.
Member States can also require undertakings to distribute public interest infor-
mation on how ECSs can be used for unlawful activities or to disseminate harmful
content, including infringements of copyright and related rights and their legal
consequences, and on means of protecting personal data when using ECSs.
Under the European Commission’s 2016 proposals, Article 21 is to be replaced.
Under the Code, national regulatory authorities are obliged to ensure that end-
users have access free of charge to at least one independent comparison tool, and
the Code sets out requirements for the comparison tool itself.46
The latter reference is clearly applicable to the growth of the internet as a commu-
nications environment.
Article 22 of the Universal Service Directive as originally drafted enabled NRAs
to require providers of publicly available ECSs to publish comparable, adequate,
and up-to-date information for end-users on the quality of their services, and set
out parameters, in Annex III, that NRAs may use.
The amendments to Article 22 made by the Citizens’ Rights Directive49 extended
this to providers of publicly available ECNs. It also introduced a new right for NRAs
to set minimum QoS requirements on an undertaking providing public commu-
nications networks, in order to prevent the degradation of the service and the
hindering or slowing down of traffic over the networks. This had become relevant
in the context of the internet, particularly as higher volumes of data in the form of
moving images, for example, via the BBC’s iPlayer, which launched in December
2007, were being transmitted, causing traffic to slow where the bandwidth was not
large enough to cope.
These are reserve powers, which enable NRAs to introduce regulation where
they consider that market players are not using their commercial freedom in an ef-
fective way to satisfy users’ and consumers’ demands, which could be detrimental
to consumer choice and, by extension, competition in the market.
Under the European Commission’s 2016 proposals, stricter requirements are
placed on NRAs to enable more comparability between service providers including
across the whole EU.50 Under the proposals, NRAs are under a requirement to
specify QoS parameters. In doing so, they have to take account of guidelines to
be produced by BEREC. NRAs would also have to specify the applicable measure-
ment methods and the way the information should be published including pos-
sible quality certification mechanisms. NRAs can, where appropriate, use the
parameters, definitions, and measurement methods set out in the Annex.
Directive 97/33/EC with regard to operator number portability and carrier pre-selection, OJ L 268/37, 3
October 1998.
54
Universal Service Directive, Recital 40.
504
numbering plans, for example, mobile services, or fixed-line services with the
same geographic area code.
Article 30 of the Universal Service Directive sets out a right to number portability
for all subscribers of PATS, including mobile services. The right is restricted in the
case of geographic numbers to numbers within the same location, or exchange.
Following the amendments made by the Citizens’ Rights Directive, numbers must
be ported within the shortest possible time and, once there is an agreement with
a subscriber to port a number, the number must be activated within one working
day. NRAs can also order compensation in cases of abuse or delay in porting a
number.
The 2016 amendments proposed by the European Commission include a re-
quirement that the switching and porting process should be led by the receiving
provider. There is also a requirement that in the event of a failure of the porting
process, the transferring provider must reactivate the number until the porting is
successful.55
9.2.6 Bundled offers
The European Commission’s 2016 proposals add a new provision on bundled offers
intended to avoid unwanted lock-i n effects, so that adding on additional services
to a bundle cannot restart the overall contract period unless a special promotional
price is available only on conditions that the existing contract period is restarted.
Key provisions, such as the information requirements for contracts, maximum
contract duration and termination rights, and switching rights, would apply to the
entire bundle.56
9.2.7 Non-discrimination
For the first time, the 2016 proposals overtly take account of the fundamental
rights and principles recognized by the Charter of Fundamental Rights of the
European Union. This may have impacted the new requirement that providers of
electronic communications networks and services must not apply any discrimin-
atory requirements or conditions of access or use on end-users based on nation-
ality or place of residence unless such differences are objectively justified and the
introduction of a fundamental rights safeguard.57
55
2016 Proposals, at Art 99. Ibid, at Art 100.
56 57
2016 Proposals, at Arts 92 and 93.
50
9.3.1 Marketing
When consumers are considering signing-up for a new communications service,
or switching providers, the main information they need to know is what service
they should expect to get and how much it will cost them. With an increasing range
of telecommunications services available via different technologies and packaged
in different bundles, getting the right information about service options can be a
challenge for consumers. Competition, which creates more supply-side options,
may have an adverse effect on demand, because it can create confusion among
consumers who find themselves faced with such a myriad of options, it may be
hard to understand which service would best suit their needs. The Citizens’ Rights
Directive sought to deal with this, by giving national regulatory authorities the
right to publish their own price comparison guides, where the market has not pro-
vided them.60 For the UK’s approach, see Section 9.3.1.5.
Ofcom, ‘Review of the General Conditions of Entitlement, Statement and Consultation’, 19 September
59
61
Ofcom Direction under General Condition 21.1 on quality-of-service, 27 January 2005.
62
Carrier pre-selection. 63
Wholesale line rental.
64
See further Chapter 8, at Section 8.3.4.3.
65
Ofcom statement about research report by GfK, 13 July 2010.
66
Ofcom research document: Provision of quality of service information, 30 January 2009.
67
Communications Act 2003, s 51(2), as amended by the Electronic Communications and Wireless
Telegraphy Regulations 2011, SI 2011/1210, Sch 1, para 27.
570
However, Ofcom’s recent approach has been to regulate by using existing compe-
tition tools and consumer transparency options, rather than imposing minimum
QoS requirements,68 an approach which is supported by the government.69
To improve transparency for consumers, since 2011, Ofcom has published com-
plaints data, periodically,70 and has taken action to improve the way in which con-
sumer complaints are handled by the communications industry, for example by
ensuring that consumers have an increased awareness of their rights to use ADR.71
In July 2010, Ofcom introduced new rules to require communications providers to
improve awareness of dispute resolution services.72
Ofcom has the power to require communications providers to provide Ofcom
with information to enable it to carry out comparative overviews of services.73 In
order to improve the availability of comparative information on quality and prices,
the government has recently granted Ofcom a new express power to carry out and
publish wide-ranging comparative overviews.74 Further new provisions set out the
scope of Ofcom’s powers to require communications providers to collect, generate,
and retain information for publication and these represent stronger powers than
Ofcom had previously to require information from communications providers.75
In revising the GCs, Ofcom has removed the power in GC21 to make directions re-
lating to quality of service, as this is now redundant.76
In April 2017, Ofcom launched an online interactive tool to allow consumers to
compare QoS between providers.77 At the same time, Ofcom published its first an-
nual report comparing QoS between providers, which addressed in particular call
waiting, complaints handling, and reliability.78
68
See Ofcom discussion document on traffic management and ‘net neutrality’, 24 June 2010.
69
See speech by Ed Vaizey, ‘The Open Internet’, FT World Telecoms conference 2010, 17 November 2010.
70
See Ofcom Telecoms Complaints papers, 21 April 2011, 22 September 2011, and so on.
71
Ofcom quality of service research report, 13 July 2010.
72
See Ofcom statement: A review of Consumer Complaints Procedures, 22 July 2010, and GC C4 and Annex
to C4 (GC14.4 and Annex 4 to GC14 in the GCs that apply up to 1 October 2018).
73
Communications Act 2003, s 136. 74
Ibid, s 134D, inserted by Digital Economy Act 2017, s 83.
75
Ibid, ss 137A and 137B, inserted by Digital Economy Act 2017, s 86.
76
Ofcom, ‘Review of the General Conditions of Entitlement: Statement and Consultation’, 19 September
2017, at paras 8.13–8.14.
77
<https://w ww.ofcom.org.uk/phones-telecoms-a nd-i nternet/advice-for-c onsumers/quality-of-service/
report/i nteractive-report>.
78
Ofcom report, ‘Comparing service quality’, 12 April 2017. 79
<http://a sa.org.uk>.
508
CAP and BCAP have published guidance on the use of ‘unlimited’ claims in
telecommunications and broadband advertising.80
The guidance says that when describing services as ‘unlimited’, advertisers must
only use the term where the user incurs no additional charge or suspension of ser-
vice as a consequence of exceeding a usage threshold associated with a fair-usage
policy, a traffic management policy or similar, and where limitations that do affect
the speed or usage of the service are moderate only and are clearly explained in the
advertisement.81 Following a review in November 2017, CAP published new guid-
ance on broadband speed advertising applicable to residential broadband services,
which came into effect on 23 May 2018.82 Claims about broadband speeds now
have to be based on the speed available to at least 50 per cent of customers at peak
time (8pm–10pm), which must be described as ‘average’. The previous position was
that advertised ‘up to’ speeds were acceptable if they were available to at least 10
per cent of customers. The guidance also contains a recommendation that speed-
checking facilities should be promoted in advertisements wherever possible. The
review followed publication of a report by the Advertising Standards Authority that
indicates that consumers do not correctly understand claims made in advertising
about broadband speeds. In particular, the review found that most consumers be-
lieve they are likely to receive a speed at or close to the headline speed claim, al-
though for many people this is unlikely to be the case.83 CAP believes that the new
standard will help consumers better understand what is available when deciding
to switch providers and to appreciate that there are a range of factors that will af-
fect the broadband speed they receive (location, technology, number in their
household using broadband). Most of the major fixed-line ISPs have signed up to
a voluntary code of practice on broadband speeds published on the Ofcom web-
site, but initiated by the Broadband Stakeholders Group, under which they are
required, among other things, to give specific information on broadband speeds
at the point of sale and on their websites. An update in 2015 gave consumers im-
proved rights to leave their broadband contract if speeds fell below an acceptable
level.84 A similar code was published for business customers in 2016. Amendments
to both codes that come into force on 1 March 2019 require speed estimates given
at point of sale to reflect the speeds customers are likely to experience at peak times
80
Guidance on making ‘unlimited’ claims in advertising for telecommunications services. See also
Chapter 14.
81
See, for example, ASA Adjudication on UK2 Group (29 February 2012), in which the ASA upheld a com-
plaint in respect of an advert stating that a Business Cloud package offered ‘unlimited’ storage space.
82
<https://w ww.asa.org.uk/uploads/a ssets/uploaded/dbf3043b-02b4-4134-9ba50f2ad0be4d06.pdf>.
83
ASA News item: ‘ASA calls for a change in the advertising of broadband speed claims’, and research re-
port by GfK.
84
Voluntary code of practice: broadband speeds, version 3.0, June 2015.
509
(8–10pm for residential services and 12–2pm for business services). A minimum
guaranteed speed and the right to exit connected to this speed must be given at
point of sale. The right to exit will also apply to bundled products. There will be a
thirty-calendar day limit to the time providers have to improve speeds before they
must offer the right to exit, and providers will be required to make after-sale infor-
mation about the right to exit more prominent and clearly link it to the minimum
guaranteed speed. Because the codes will measure customer speeds at peak time,
they can apply to all types of access technologies.85
In 2016, the Broadband Stakeholder Group established an ‘Open Internet Code’
to clarify the rules on internet traffic management. It brought together the BSG’s
early Traffic Management Transparency Code, its 2012 Open Internet Code, and
the requirements of the 2015 Regulation. The new Code is built round four ISP
commitments:
All the major ISPs have signed up to the Code, representing 90 per cent of UK
subscribers on fixed and mobile contracts.
Ofcom consultation, ‘Review of the General Conditions of entitlement: Consultation on the general con-
86
9.3.1.4 Price
In its March 2016 budget, the government noted that broadband pricing can be
opaque and asked for industry to act to improve clarity. Accordingly, since 31
October 2016, the ASA has required that, in order to comply with ASA rules, broad-
band adverts that include price claims must convey a consumer’s full commit-
ment required to purchase the service. The ASA has determined that, if followed,
the following three ‘key principles’ should produce advertisements that are in line
with the advertising code:
• The advertisement presents all compulsory elements of the total financial com-
mitment (up-f ront costs, ongoing costs, and contract length) together, avoiding
undue emphasis on any one element.
• The advertisement presents one inclusive price for compulsory up-front costs
and an inclusive price for a consumer’s ongoing monthly cost (combining the
line rental and broadband cost where line rental is offered by the provider).
• The advertisement makes clear if an introductory discounted price for one/some
of the elements applies and, if so, for how long.89
87
Electronic Communications and Wireless Telegraphy Regulations 2011, SI 2011/1210, Sch 1, para 27(b).
88
Ibid, Sch 1, para 88.
89
ASA advice online: Compulsory charges: Telecommunications, 7 July 2016.
90
Universal Service Directive, Art 21 (as amended).
51
91
Ofcom statement, ‘Accreditation scheme for price calculators: Decision on changes to the scope and op-
eration of the Scheme’, 6 November 2013.
92
In the version of the GCs that applies up until 1 October 2018, this information is set out in GC9.
512
The 2015 Regulation introduced new provisions affecting contracts that include
IASs and that are concluded or renewed from 29 November 2015. Such contracts
must also include the following information:
In addition, contracts for IASs should take into account provisions in Article 3 of
the Regulation on safeguarding open internet access: end-users have the right to
access and distribute information and content, use and provide applications and
services, and use terminal equipment of their choice, irrespective of the end-user’s
93
Art 4(1).
531
94
C1 in the revised GCs and GC9 in the version in force until 1 October 2018. 95
GC C1.6.
96
GC C1.7. 97
GC C1.8.
514
March 2018, Ofcom published new guidance on the procedures for terminating
contracts.98
98
Ofcom statement on emergency planning direction, number withdrawal and guidance on contract ter-
mination, 26 March 2018.
99
Ofcom statement on automatically renewable contracts, 13 September 2011. 100
GC C1.3.
101
GC B3 in the revised GCs and GC18 in the version that applies up to 1 October 2018.
102
See further Chapter 3, at Section 3.4.1.
51
There were also problems with cashback schemes. A cashback scheme is a form
of promotion offered by independent retailers promising a payment or, for ex-
ample, a mobile handset in exchange for signing up to a service contract. Ofcom
received complaints on the following issues:
• The independent retailer refused to honour the cashback promise on the basis
that the contractual terms of the offer had not been met, although these were
often designed so that it was difficult to claim successfully.
Ofcom statement and consultation, Protecting citizen-consumers from misselling of fixed-l ine telecoms
104
services, 22 November 2004.
516
• The independent retailer was unable to honour the cashback offer because it had
gone out of business.
Ofcom started taking action against misselling in April 2004, when it published
a consultation on misselling in the fixed-line market. At that time, there were a
number of applicable consumer- protection measures in place. For example,
Section 52(2)(e) of the Communications Act 2003 gave Ofcom the power to set GCs
on any matter appearing to Ofcom to be necessary for securing effective protec-
tion for the domestic and small business customers of public communications
providers, but at the time Ofcom said that it would only use this power if it was per-
suaded that there was evidence that there was a serious problem and that current
safeguards were not effective. There was also an industry-agreed CPS and WLR
customer transfer process in place between the fixed-l ine telecommunications in-
dustry and consumer representatives aimed at preventing misselling and, in par-
ticular, slamming.105 It required a switchover period of ten working days before a
customer order could be fully processed during which time both the transferring
and the receiving provider would send the customer a ‘notification of transfer’
letter, to ensure that the customer was not being transferred without its knowledge
and consent. Also, at this time, voluntary guidelines on sales and marketing had
been agreed between the industry and consumer representatives, and Ofcom had
published a consumer guide to using alternative phone companies.106
There were also a number of applicable non- telecommunications- specific
consumer protection laws in place at the time, which variously prohibited false,
inaccurate, or misleading descriptions about goods and services;107 made it an of-
fence for a person in the course of business to give a consumer a misleading indica-
tion of the price of services;108 protected consumers against unfair standard terms
in contracts they made with traders;109 prohibited unfair (misleading) commercial
practices;110 allowed consumers to cancel contracts made on their doorstep;111 and
required consumer rights information to be provided when orders were made on-
line or over the phone.112 Ofcom is also a designated ‘enforcer’ under Part 8 of the
Enterprise Act, which means that it can get an enforcement order from the courts
against anyone who breaks consumer protection legislation.113
105
Ibid. 106
Ibid. 107
The Trade Descriptions Act 1968.
108
Consumer Protection Act 1987.
109
Unfair Terms in Consumer Contracts Regulations (SI 1999/2083).
110
Control of Misleading Advertisements Regulations (SI 1988/915).
111
Consumer Protection (Cancellation of Contracts Concluded Away from Business Premises) Regulations
(SI 1987/2117).
112
Consumer Protection (Distance Selling) Regulations (SI 2000/2 334).
113
Enterprise Act 2002, s 213(5A).
571
The requirement for communications providers to establish codes of practice on sales and marketing was
115
contained in GC14, Annex 3, which, until December 2009, was Annex 4. The extension to LLU was achieved by
amending the definition of ‘fixed-l ine telecommunications service’ in GC14.9(h) so that instead of covering
a specific list of fixed-l ine services, it covered all types of fixed-l ined services, so capturing LLU (see Ofcom
statement, Protecting consumers from misselling of telecommunications services, 21 May 2007, para 3.19).
518
In 2007, the mobile network operators introduced their own voluntary code to
tackle misselling.116
9.3.2.7 Cancel Other
At the same time as Ofcom introduced the requirement for communications pro-
viders to establish codes of practice on selling and marketing, it also published a
draft proposal for resolving a dispute between BT and a number of communica-
tions providers relating to BT’s use of the ‘Cancel Other’ function. BT used this to
cancel orders for CPS and WLR in certain circumstances, including where a cus-
tomer had been slammed.
At the time the dispute arose, BT was subject to a Direction issued in 2003 speci-
fying the circumstances in which BT was permitted to use Cancel Other. However,
it only applied to CPS because it was made before WLR was introduced. The al-
ternative providers were concerned that under the 2003 Direction BT could in-
appropriately cancel transfers and that the system limited their ability to address
allegations of slamming, because a customer who believed he had been slammed
was not required to contact the service provider that placed the transfer request.
The alternative providers wanted customers who wanted to cancel a transfer to be
required to contact the gaining provider (the one that placed the transfer request)
first, while still allowing the losing provider to cancel in certain circumstances.
They also argued that the system should apply to all providers, not just BT, as cus-
tomers might want to transfer between alternative providers.
In January 2005, Ofcom published a new Cancel Other Direction to BT,117 the
detail of which was further amended in July 2005.118 The Direction set out the cir-
cumstances in which BT could use Cancel Other. BT could continue to use the
Cancel Other function in cases of slamming but was required to provide more in-
formation to alternative providers on its use of Cancel Other. Ofcom also provided
further guidance on the definition of slamming, and expected that the Direction
would lead to a reduction in the number of cases in which BT used Cancel Other.
Ofcom resolved to reconsider BT’s use of Cancel Other before the anticipated two
year period for the sales and marketing codes of practice under GC14 ended. If
slamming was no longer a problem at that stage, the role of Cancel Other as a con-
sumer protection mechanism would also be reduced, and Ofcom thought it might
remove BT’s ability to use Cancel Other at this point.
116
Code of practice for the sales and marketing of subscriptions to mobile networks, 31 July 2007.
117
Ofcom Direction under section 49 of the Communications Act 2003 and Condition AA1(a) imposed on
British Telecommunications plc as a result of the market power determinations made by the Director General
of Telecommunications that BT has significant market power, 20 January 2005.
118
Ofcom Direction modifying a Direction, 28 July 2005.
519
GC C8.2.
119
The phrase ‘durable medium’ originally came from the Consumer Protection (Distance Selling)
120
Regulations 2000, SI 2000/2 334. These were replaced by the Consumer Contracts (Information, Cancellation
and Additional Charges) Regulations 2013/3134 in relation to contracts entered into on or after 13 June 2014.
520
121
Sales incentives are described in GC C8.11 (GC23.10 in the version of the GCs that applies until 1 October
2018), and are incentives from which the customer does not benefit until he has entered into the contract for
the mobile service. The terms and conditions of such offers must not be unduly restrictive and the customer
has to be provided with certain information about the deal.
122
GC C7.3 in the version in force from 1 October 2018 and GC22.3 in the old version.
521
123
GC C7.9 and C7.10 in the version in force from 1 October 2018 and GC22.10 and GC22.11 in the old version.
124
GC A1.3 in the version that applies from 1 October 2018. GC1.2 in the old version.
125
Ofcom consultation, ‘Making switching easier and more reliable for consumers’, 29 July 2016.
25
The gaining provider must have procedures in place to enable the customer to ter-
minate their contract without unreasonable effort, including by telephone, email,
and post. Gaining providers also have to keep sales records for at least six months
and records of consent for not less than twelve months.126 The requirement to keep
records of consent aims to reduce the incidence of slamming by enhancing Ofcom’s
enforcement capabilities.127
Gaining providers are under an obligation to ensure that where they have elected
to coordinate a migration of broadband and fixed-line over the same line, they
submit an order to Openreach or KCOM as applicable for the simultaneous transfer
with minimal loss of service of both communications services.128
Where the switch involves a company not using the Openreach or KCOM access
network, such as a cable company, consumers will sometimes have to contact the
losing provider: a process known as ‘cease and re-provide’.
126
GC C7.6 and C7.7 in the version in force from 1 October 2018 and GC22.7 and GC22.8 in the old version.
127
Ofcom statement, ‘Consumer switching, A statement on the GPL NoT+ elements’, 20 December 2013.
128
GC 7.13 in the version in force from 1 October 2018 and GC22.14 in the old version.
129
Digital Economy Act 2017, s 2.
130
Queen’s Speech 2016 background briefing notes, 18 May 2016.
253
Both changes are implemented through amendments to GC7 and take effect
from 1 July 2019.131
Ofcom also looked at introducing rules on switching between providers who
operate on different platforms, for example, switching landline, broadband, and
pay TV between the Openreach, KCOM, Virgin cable, and Sky satellite platforms.
It consulted on two options. The first was an enhanced cease and re-provide pro-
cess that would give consumers flexibility in how they contact their old provider
to cancel their existing services. The new provider would be required to offer to
organize the switch on the consumer’s behalf. The second was a gaining provider-
led process.132 However, ultimately, Ofcom decided not to take any regulatory
action.133 Instead, it considered that it could better further consumer interests by
increasing its focus on helping consumers navigate the communications market;
subsequently, it published a call for inputs to inform a new project on customer
engagement.134
131
Ofcom consultation, ‘Consumer switching: Further proposals to reform mobile switching’, 29 July
2016, Ofcom consultation, ‘Consumer switching: Proposals to reform switching of mobile communications
services’, 19 May 2017 and Ofcom decision on reforming the switching of mobile communications services,
19 December 2017.
132
Ofcom consultation, ‘Making switching easier and more reliable for consumers’, 29 July 2016.
133
Ofcom statement, ‘Decision on switching landline, broadband and/or pay TV between different plat-
forms’, 14 July 2017.
134
Ofcom call for inputs, ‘Helping consumers to engage in communications markets’, 14 July 2017.
135
Section 102. 136
Section 124S.
542
137
Communications Act 2003, s 52.
138
Annex 4 to GC 14 in the version that applies until 1 October 2018 and Annex to GC C4 in the version that
applies from 1 October 2018.
139
Ibid.
52
140
Ofcom review of the General Conditions of Entitlement, ‘Consultation on the general conditions relating
to consumer protection’, 20 December 2016.
141
2015 Regulation, Art 3. 142
Ibid, Art 4(1). 143
Reg 7.
144
Ofcom, ‘Monitoring compliance with the EU Neutrality regulation: A report to the European
Commission’, 23 June 2017.
562
145
Oftel consultation, ‘Developing a telecommunications ombudsman’, March 2001.
146
Communications Act 2003, s 52(2)(b). 147
Ibid, s 52(3). 148
Ibid, s 52(5).
149
Ibid, s 54.
150
GC C4.3 in the version in force from 1 October 2018 and GC14.5 in the old version.
151
<http://w ww.ombudsman-services.org/communications.html>.
152
<http://w ww.cisas.org.uk/>.
527
the ADR scheme for the postal sector. The fact that one is an ombudsman and the
other is a consumer adjudication service makes the two schemes fundamentally
different in their approach to dispute resolution, although the fact that providers
can choose between schemes also effectively creates a competitive market in dis-
pute resolution schemes.
The ombudsman scheme offers a high degree of customer support. This includes
helping consumers to fill out their dispute resolution application form and pro-
viding advice on any evidence that a consumer may wish to consider submitting.
An investigations team examines the allegations and submissions from the
communications provider and will contact either party to seek further informa-
tion on any points. The process is an iterative one and each party has the oppor-
tunity to make submissions on the provisional conclusion before it is passed to the
ombudsman for a final decision (if one or other party does not accept the provi-
sional conclusion).
As the operator of an adjudication scheme, CISAS places great weight on treating
consumers and communications providers equally. It will not help either party to
put their case together and will not advise either of them on what evidence they
would need to support their case, although it does publish guidance including evi-
dence checklists for enquirers and consumers. Consumers can complete applica-
tions online or by post. On request, CISAS staff will guide and assist applicants on
completing their application. CISAS does not investigate consumer complaints,
but consumers are provided with an opportunity to comment on the communi-
cation provider’s response to their claim. Adjudicators have the ability to request
further information from either party in order to help them to make a fair deter-
mination of the claim. Adjudicators apply legal principles to determine whether
the consumer has proven, on the balance of probabilities, that their communi-
cations provider has breached the contract or their code of practice. Neither con-
sumers nor communications providers have a right to challenge an adjudication,
although the complainant can choose to accept or reject the decision.
Ofcom has to secure consistency in standards between the schemes,153 and
undertakes reviews periodically. Some differences are an inevitable by-product of
having two schemes, and some are a direct result of differences in scale. However,
where those differences mean that consumers will receive a lower standard of
treatment depending on which ADR scheme their communications provider be-
longs to then Ofcom has to take steps to ensure an appropriate degree of alignment.
Any significant discrepancies between the two schemes could potentially create
concern about whether the ADR schemes are meeting the needs of consumers and
could also create incentives for the communications providers to choose to belong
to a particular ADR scheme if there is a perception that dispute resolution under
one scheme is more likely to favour them than under another.
The Alternative Dispute Resolution for Consumer Disputes (Competent
Authorities and Information) Regulations 2015 introduced requirements for busi-
nesses to provide information about certified ADR providers to consumers. Ofcom
has certified four schemes: Ombudsman Services, Centre for Effective Dispute
Resolution, Promediate, and The Retail Ombudsman.
9.3.3.5 Compensation
In March 2017, Ofcom proposed the introduction of a system of automatic com-
pensation for consumers and smaller businesses when things go wrong with their
communications services.154 However, a voluntary initiative was proposed by in-
dustry, which Ofcom decided to accept instead of pursuing regulatory action.155
The scheme contains automatic compensation for the following failures:
• Delayed repair following loss of service. £8 per day if the service is not repaired
after two working days.
• Delayed provision. £5 per day.
• Missed appointments. £25 per missed appointment if the engineer does not turn
up or the appointment is cancelled with less than twenty-four hours’ notice.156
154
Ofcom Consultation, ‘Automatic Compensation: Protecting consumers from service quality problems’,
24 March 2017.
155
Ofcom Statement, ‘Automatic Compensation: Protecting consumers from service quality problems’, 10
November 2017.
156
These amounts were less than originally proposed by Ofcom, which provided for £10, £6, and £30
respectively.
157
Section 3.
529
158
GC14.2. 159
GC C2. 160
See further Chapter 14, at Section 14.7.
161
Non-geographic numbers starting 084, 087, 090, 091, 098, and 118. 162
In Annex 2 to GC14.
163
GC C2.4–2 .9. 164
See further Chapter 14, at Section 14.7.
165
Consumer Protection Partnership: fourth report, October 2017.
530
166
This is also a problem for FTTP (fibre to the premises) networks, as optical fibre networks do not con-
tinue to operate during a power failure. Ofcom published Guidelines on the use of battery back-up to protect
lifeline services delivered using fibre optic technology on 19 December 2011 but announced in its statement
to its Strategic Review of Digital Communications (25 February 2016) that it was withdrawing this and would
proceed by assessing what operators were doing on a case-by-c ase basis and keeping under review the resili-
ence of operators’ networks.
167
GC A3.3 and A3.6(c).
513
10
Telecommunications regulation was born from the need to police the removal of
state-owned telecommunications monopolies in Europe from the 1980s onwards.
Indeed, the telecommunications sector is not a market which has developed nat-
urally. Instead it was a response to governmental choices, originally to create a
state monopoly and, in the late twentieth century, to move to market provision,
ultimately to be controlled by general competition law. This chapter introduces
these competition rules and illustrates how they have been applied in the tele-
communications context. Since competition law applies across all sectors, the
law and practice developed in competition cases from other areas also plays an
important role in deciding how competition law applies to telecommunications
operators.
354
The move to a market driven approach has not been entirely successful, with
former monopolists continuing to hold substantial market power in some mar-
kets, often reflecting how the market was originally formed (whether on the basis
of substantial state intervention or not), as well as a trend towards horizontal con-
solidation. It is too simplistic, however, to suggest a one-way trend, as technological
developments (eg development of mobile as a substitute for fixed line services), as
well as policy changes (eg the removal of the prohibition on BT operating in the
content market) provide opportunities for market entrants who then change the
structure or nature of the market.
Whilst competition law may now be the preferred tool to control the behaviour
of private actors acting, alone or in concert, to distort competition, regulation re-
mains relevant and our discussion takes place against the background of the regu-
latory framework for electronic communications.
In contrast to previous editions, this chapter includes material on content pro-
vision and competition law. In a text dealing with telecommunications, this may,
at first glance, need explanation. Convergence (according to which transmission
technology is service neutral, allowing the same service to use different transmis-
sion technologies and the same transmission technology to distribute different
types of service) may be seen to entrench the divide between transmission (tele-
communications) and electronic content (including broadcasting and interactive
services such as gaming). This divide, however, was never as clear as policy docu-
ments viewed it, and market changes have led to:
a race towards building up gatekeeper positions at the different levels of trade,
both platforms and content, with the danger of the monopolization of large parts
of the sector.1
Content providers were always aware of the importance of having access to distribu
tion systems: access determines the possibility of access to the audience. Content
is similarly important for telecommunications companies. There has been signifi
cant vertical integration between content and transmission markets. To take one
example,2 in 2014 Liberty Global (a TV, broadband, and mobile service pro-
vider) acquired shares in production company All3media (the UK’s largest in-
dependent producer), free-to-air broadcaster ITV, De Vijver Media (a Belgian
production and free-to-air television company),3 and Ziggo (a Dutch cable TV
1
Ungerer, H, ‘The Reasons for Intervention through Competition Policy’ in Donders, K, Pauwels, C, and
Loisen, J, (eds), The Palgrave Handbook of European Media Policy (Basingstoke: Palgrave Macmillan, 2014), 405.
2
Other examples include Telefonica/DTS in Spain, and Vivendi/Telecom Italia.
3
European Commission, Mergers: Commission clears Liberty Global’s acquisition of controlling stake in
De Vijver Media, subject to commitments, Press Release, 24 February 2015.
53
operator),4 the year after it acquired Virgin Media.5 Content now is not just about
television, but includes a range of services provided across the internet. The rela-
tionship between actors in each of these fields is becoming increasingly intertwined.
The chapter is structured as follows. After identifying the distinction between
competition and regulation (Section 10.2), we begin with consideration of the pro-
hibition on restrictive agreements and concerted practices in Article 101 TFEU (and
s2 Competition Act 1998 in the UK) (Section 10.3) before considering Article 102
TFEU—unilateral abuse of market dominance (Section 10.4). As well as looking in
more detail at how different competition law tools have been applied in telecommu-
nications markets, we will consider how competition rules have been used to con-
trol anti-competitive outcomes in the interface between transmission and content
provision. Particular competition issues may arise at the interface where regulated
transmission services and the provision of (‘unregulated’) content meet, and these
are discussed at Section 10.5. Our merger control section (Section 10.6) focuses on
the EU and UK approach to international telecommunications mergers, to avoid
infrastructure monopolies being created. We also briefly consider media plurality
rules in merger control. Market investigations (Section 10.7), both at EU and UK
level, have played an important role in UK telecommunications over the last decade.
We conclude with a short section on enforcement and appeals in the ‘concurrent’
UK competition law enforcement architecture, which sits alongside the regulatory
structure for electronic communications (Section 10.8). Although the role of the
state has been important in shaping the telecommunications market, this chapter
does not cover Article 106 TFEU nor the rules on State aid.6
4
European Commission, Case COMP/ M.7000, Liberty Global/Ziggo, OJ [2015] C 145/ 7; European
Commission, Mergers: Commission clears acquisition of Dutch cable TV operator Ziggo by Liberty Global,
subject to conditions, Press Release 10 October 2014. The Commission decision was the subject of successful
appeal to the General Court for failure to state adequate reasons: Case T-394/15, KPN v Commission, judgment
26 October 2017, ECLI:EU:T:2017:756.
5
European Commission, Case COMP/M.688, Liberty Global/Virgin; Mergers: Commission approves acqui-
sition of UK cable operator Virgin Media by Liberty Global of the US, Press Release 15 April 2013.
6
See further Chapter 4, at Section 4.4, and Chapter 8, at Section 8.5.7. See also Rose, V and Bailey, D, (eds),
Bellamy & Child, European Union Law of Competition (7th edn, Oxford University Press, 2013), chapters 11
and 17.
536
behaviour had not occurred? As a general rule, competition law analysis is retro-
spective, regulatory analysis is prospective. Competition law addresses business
deals or practices which misuse (or threaten to allow the misuse of) market power
to restrict or distort competition, ultimately harming consumers of services. The
consumer harm can be felt through higher prices than necessary, or lower quality
or choice of service—or often a combination of these.
There are four ways in which competition law looks to prevent (or at least min-
imize) distortion of competition:
The two competition law controls which do not fit quite so neatly into a framework
which sees competition law as a retrospective and behavioural matter—merger
control and market investigations—have been used by both EU and UK author-
ities to prompt desired structure changes in electronic communications markets
where regulatory and behavioural competition powers alone may not be sufficient.
These structural controls have been particularly used to prevent the formation of
‘converged’ firms which would be able to leverage market power from content to
transmission or vice versa (see Sections 10.5 and 10.6).
Case C-2 80/0 8P, Deutsche Telekom v Commission [2010] ECR I-9555.
7
Art 101(1) Treaty on the Functioning of the EU (TFEU); Regulation EC 1/2003 [2003] OJEU 1/1, Art 4;
8
10.3.1 Overview
Competition infringements carried out by two or more businesses in collusion
with each other are prohibited under both EU law (Article 101 TFEU) and UK
statute (Competition Act 1998, s 2). The prohibitions are (deliberately) identically
worded. EU law will apply where the agreement or collusion may have an appre-
ciable effect on trade between Member States.11 Where there is no appreciable
effect on inter-state trade, but trade within the UK is affected, the Competition Act
1998 will apply to any restrictions in the agreement.12 UK authorities and courts
are obliged to apply EU competition law if inter-state trade is appreciably af-
fected13—we will look at how the EU and national competition law systems interact
in Section 10.7.
Certain ‘agreements’ are prohibited. To be prohibited, the agreement
11
Art 101 Treaty on the Functioning of the EU [2016] OJEU C202/47.
12
Competition Act 1998, s 2(1)(a). 13
Regulation 1/2003, n 8, Art 3(1).
14
C-2 and 3 /01P, BAI and Commission v Bayer [2004] ECR I-2 3.
15
Case 107/82, AEG Telefunken v Commission [1983] ECR 3151; C-74/0 4P, Commission v VW [2006] ECR
I- 6585.
16
Case 48/69, ICI v Commission (Dyestuffs) [1972] ECR 619, para 64.
17
See Bellamy & Child, n 6, paras 2-0 81–2 .084.
504
‘Agreements’ between legal persons forming part of the same economic group
(undertaking) will not be caught by competition law—they are considered to be
part of the internal workings of a single undertaking.18 Where, however, ‘functional’
separation of different parts of the same undertaking is required by an NRA to en-
sure competition in network markets, it is possible that the agreements between the
parent and the functionally separated business could be subject to competition law
scrutiny.
An ‘undertaking’ is any form of business—any ‘entity’ engaged in an ‘economic
activity’.19 This can include companies (groups of companies under common con-
trol can be treated as a single undertaking)20 and partnerships, as well as individuals
carrying on business.21 An economic activity is the offering of goods or services onto
a market—it is not necessary that the activity is profit making.22 But the offering must
be on an economic basis, so some kinds of state activities fall wholly or partly out-
side the scope of the prohibition.23 In modern electronic communications markets,
all network operators will now be undertakings.
The ‘object or effect’ of the agreement must be to restrict competition. An agree-
ment will have the ‘object’ of restricting competition if it would reduce the inde-
pendent action of the parties on the market.24 A cartel or other resale price fixing
agreement is a clear example. The test is objective: it is not necessary to show that the
parties actually intended to restrict competition in the particular case.25 Nor does the
anti-competitive object necessarily need to be the only reason for the agreement.26
Many agreements having legitimate overall aims (eg network interconnection) can
nevertheless be infringements if their terms are restrictive (eg they attempt to fix
downstream pricing).
If an agreement is restrictive by object, there is no need to demonstrate an ac-
tual effect on competition. In all other cases, there must, with a reasonable degree
of probability, be an appreciable effect either on actual or on potential competition
arising from the agreement. The CJEU has insisted that there is, in each ‘effect’
case, some definition of the relevant market in which the effects are said to occur.27
18
Case C73/95P, Viho v Commission [1996] ECR I-5 457.
19
Case C-41/9 0, Hofner and Elser v Macrotron [1991] ECR I-1979. 20 Viho, n 18.
21
Case C-309/99, Wouters [2002] ECR I-1577—i ndividual members of the Dutch bar were undertakings.
22
Case 209/78, Van Landewyck v Commission [1980] ECR 3125.
23
Bellamy & Child, n 6, paras 2.014–2 .016.
24
Case 56/65, Société Technique Minière v Maschinenbau Ulm [1966] ECR 235.
25
Joined cases 56 and 58/6 4, Consten and Grundig v Commission [1966] ECR 299, at 342.
26
eg Case C-2 35/97P, Montecatini v Commission [1999] ECR I-935, at para 122.
27
Case C-2 34/89, Delimitis v Henninger Brau [1991] ECR I-935, at paras 15–16.
514
The European Commission has published guidelines describing how this exercise
should be carried out.28
An agreement will have an anti-competitive effect in the market if it causes the
pattern of competition to develop differently from undistorted competition in that
market.29 Where related parties enter into a network of similar agreements, the cu-
mulative effect of the restrictions operated by the whole network will need to be con-
sidered together—even if a single one of the agreements alone would not have an
appreciable market effect.30
‘Exemption’ from the prohibition, for restrictive agreements which nevertheless
have countervailing economic benefits, is available (under TFEU, Article 101(3))
where the agreements:
28
Commission Notice on the definition of the relevant market, 9 December 1997 [1997] OJEU C 372/5. We
consider this in more detail in the discussion of Article 102 (abuse of dominance) at Section 10.4 below.
29
Société Technique Minière, n 24. 30
Delimitis, n 27.
31
The Commission has produced guidance on how the exemption will be applied in practice: Commission
Guidelines on the application of Article 81(3) [now Article 101(3) TFEU], 27 April 2004, [2004] OJEU C101/97.
524
be justified solely on the basis of the EBU members’ public service broadcasting
mission to provide cultural, scientific, and minority programmes.32
The term ‘consumers’ has been widely cast to include all users of the products
in question.33 In general, where the restrictive effect on competition is not large,
and the parties to the agreement or their distributors do not have a high degree
of market power, there will be an assumption that customers and consumers will
ultimately benefit from the efficiency gains.
Restrictions will normally only be considered ‘indispensable’ where, in the
absence of the restriction, the efficiency gain would not occur. If there are feas-
ible alternative arrangements which are less restrictive, the agreement will not
benefit from exemption. 34 This requirement applies not only to the strength
of the impact on competition but also on its duration. 35 For example, a non-
competition covenant between the parents and a new joint venture may be in-
dispensable for the period they each remain shareholders, but not indefinitely. 36
Similarly, if the parents’ covenants extend to products which the joint venture is
not producing, it is unlikely that the restrictions they accept will be considered
indispensable. 37
Finally, competition should not be ‘eliminated’ for a substantial part of the
market for products supplied under the agreement.38 This is only likely to be an
issue where the market power of the parties to the agreement (their market shares
are usually a good indicator) are high. Competition must be eliminated in a sub-
stantial part of the market: it is not enough for an isolated third party to complain
that he cannot get supply.39 Even where the parties have market power, if there
is the potential for competition from third parties, for example in markets where
technological change can ‘tip’ the competitive balance fairly rapidly, a finding of
‘elimination’ of competition may be unlikely.40
In the remainder of this section we look at how the framework we have just
described applies to some common agreements in electronic communications
markets.
32
Case T-528/93, Metropole Television v Commission [1996] ECR II-6 49, paras 116–123. The General Court
(formerly the Court of First Instance) is the first-t ier court hearing appeals against European Commission
competition decisions.
33
Exemption Guidelines, n 31, para 84. 34 Ibid, para 75. 35
Ibid, para 81.
36
eg Commission decision M.852 BASF/Shell, para 49—a two-year post term restrictive covenant was not
justified.
37
eg Commission decision COMP/39736, Areva/Siemens, 18 June 2012: commitments accepted permitting
a non-compete clause with a joint venture on condition that it applied only to specified core products made
by the joint venture.
38
Exemption Guidelines, n 31, paras 105–115. 39 Case C-75/8 4, Metro II [1986] ECR 3201, para 64.
40
See the analysis in the Exemption Notice, n 31, paras 114–115.
534
• sector regulation and competition law should be applied consistently with each
other, but they pursue different aims. Mainstream competition law looks at what
has happened, whereas regulation looks towards shaping what might happen in
the future;44
• EU competition laws apply ‘in the normal way’ to access agreements which have
been approved or authorized by NRAs or to terms which have been approved
after inclusion by agreement between the parties.45 In Deutsche Telekom (an
abuse of dominance case, see below Section 10.4.3) the CJEU confirmed that
the fact that a general ‘wholesale’ interconnection tariff had been approved by
the German NRA did not absolve DT from complying with competition rules to
avoid a ‘margin squeeze’ on competing downstream operators;46
• non-exclusive access agreements are ‘in principle’ unlikely to be restrictive pro-
vided that there are proper safeguards (as also required under the EU regulatory
Commission guidelines on market analysis and the assessment of significant market power under the
42
Community regulatory framework for electronic communications networks and services, 11 July 2002 [2002]
OJEU C165/6 (‘SMP guidelines’), paras 22–32.
43
‘Access guidelines’, 22 August 1998, [1998] OJEU C265/02. 44
SMP guidelines, n 42, para 22.
45
Access guidelines, n 43, para 60. 46
Deutsche Telekom v Commission, n 7.
54
47
Access guidelines, n 43, paras 132 and 139. 48
Ibid, paras 136 and 141.
49
eg O2 UK/T-Mobile UK (3G) [2003] OJ L200/59. 50
Access guidelines, n 43, paras 140 and 143.
51
European Conference of Postal and Telecommunications Administrations: Commission Press Release
IP/9 0/188 6 March 1990.
52
See SMP guidelines, n 42.
54
53
eg the dispute on BT’s SIA on charge change terms: CW/01083/01/12.
54
Of the 23 open cases at the CAT as of 15 August 2017, four were related to electronic communications.
55
Commission Decision, O2 UK/T-Mobile UK (UK Network sharing) [2003] OJEU L200/59.
56
See eg T-Mobile Deutschland (network sharing Germany) [2004] OJEU L75/32.
57
Framework Directive [2002] OJEU L108/33 Art 12; and Access Directive, [2002] OJEU L108/7, Art 5.
564
58
eg case T-328/03, O2 Germany v Commission [2006] ECR II-1231, at para 109.
59
Council Regulation 531/2012, 13 June 20102, roaming on public mobile communications networks, [2012]
OJEU L172/10; Commission Regulation implementing Regulation 531/2012, 14 December 2012, [2012] OJEU
L347/1.
60
n 55. 61
n 56. 62
O2 Germany (n 58).
574
63
Ibid, paras 68–69, 109. 64
Ibid, para 72.
Case 258/78, Nungesser and Eisele v Commission [1982] ECR 2015.
65
584
Even if a licence could fall within the ‘restrictive agreements’ prohibition, there
are two relevant EU ‘block’ exemptions—for ‘technology transfer’ licences66 and
for ‘vertical’ agreements67 (which can apply—in particular—to software copyright
licences).
The ‘technology transfer’ block exemption, and Commission guidelines ac-
companying it, apply to IP licences whose primary purpose is for the licensee to
‘produce’ a product (including services).68 IP here also includes recorded confi-
dential technical ‘know-how’, provided the know-how is substantially valuable in
producing the services licensed.69 In contrast, straightforward licences for the re-
sale of a service—a lready produced including the rights—fall within the ‘vertical’
exemption.
Drawing a bright line between these two scenarios in many electronic commu-
nications agreements is not straightforward. An agreement where an originating
company licences a manufacturer to produce mobile devices for sale to consumers
involves ‘technology transfer’. An agreement between the same originating com-
pany and a retail chain allowing the retailer to sell the same devices under the
originator’s trademark would be a ‘vertical’ agreement. But the electronic commu-
nications sector has a large number of licences in the ‘grey’ area between the two
exemption regulations. For example, a licence of IP (eg brand rights) for an MVNO
to run a new service over a mobile network could be seen as ancillary to the resale
of capacity on the network owner’s system (a ‘vertical’ agreement). Or, it could be
characterized as necessary to allow the MVNO to produce (new) services for con-
sumers incorporating the software, know-how, branding etc. needed to supply the
innovative service (likely to be a ‘technology transfer’). In these borderline cases,
much will depend on a closer analysis of the agreement, the importance and type
of IP included, and other surrounding circumstances.
Both block exemptions only apply to agreements where the parties have a mod-
erate share of relevant product and territorial markets. For the vertical agreements
block exemption and those technology transfer agreements where the parties are
not already competitors, neither party should have a share of more than 30 per cent
in any market.70 For technology transfers where the parties are competitors, their
combined share of any market should not exceed 20 per cent. Beyond these market
share thresholds, an individual assessment of the technology licence against the
competition rules will be required.
These market share requirements need to be read alongside the Commission’s
guidance on when agreements are presumed not to restrict competition
66
Regulation 316/2014, [2014] OJEU L93/17 (technology transfer).
67
Regulation 330/2010, [2010] OJEU L102/1 (vertical agreements). 68
Art 1(1)(c).
69
Art 1(1)(i). 70
Regulation 316/2014, n 66, Art 3; Regulation 330/2010, n 67, Art 3.
594
74
Regulation 316/2014, n 66, Art 4(1)(c)(ii); Guidelines on the application of Article 101 to technology
transfer agreements (‘TT guidelines’), [2014] OJEU C89/3, paras 105–114.
75
Commission guidelines on vertical restraints (‘Vertical guidelines’), [2010] OJEU C130/1, paras 52–5 4.
76
TT guidelines, n 74, paras 113–114, 208–215. 77
Ibid, para 115. 78
Ibid, paras 184–188.
50
If the main purpose of the licence is not the ‘production’ of services, so that the
IP is a secondary part of a wider resale arrangement, the ‘vertical’ agreements ex-
emption may be available. This is most likely to be the case for agreements relating
to trademarks and some forms of copyright. This exemption only applies to agree-
ments between non-competitors in the ‘resale’ market, and—as noted above—
only where the reseller has less than a 30 per cent share of that market.
The prohibitions in the ‘vertical’ exemption Regulation on resale price main-
tenance and on absolute territorial protection for distributors—m irroring the
technology transfer exemption—apply to products incorporating the IP used for
marketing them.80 Again the Commission has produced detailed guidelines on
how the ‘vertical’ block exemption should be applied which also guides firms
whose distribution arrangements may be a close, but not exact, match for the
terms of the exemption.81
Although the main purpose of a vertical agreement is the distribution of prod-
ucts, the use of IP rights in connection with the distribution is also exempted on
the same terms—for example, where a franchisee uses the trademarks of his sup-
plier in a retail context.82 The IP licence must be directly related to the resale of the
goods or services—if it is not necessary for this purpose then it will not be auto-
matically exempted and an individual assessment of competition compliance will
be required.83
Licences of software—i n Europe, protected by copyright—may pose particular
competition compliance issues. If the software licence is a secondary aspect of an
overall ‘vertical’ agreement or technology transfer licence, the relevant ‘block ex-
emption’ will apply. However, ‘pure’ or self-standing software licences do not have
their own competition block exemption and may not come within either the ‘ver-
tical’ or ‘technology transfer’ block exemptions—t he technology transfer block ex-
emption focuses on licences of patents and know-how. Many software agreements
may need individual assessment for compliance with the competition rules.84
EU legislation goes some way to harmonizing the scope of protection for IP in
software so as to underpin competitive markets. In particular, the Directive on
copyright protection for software—g iven on the same basis as a literary work—
does not apply to the part of the program which interfaces with other programs
79
Ibid, paras 133–134, 242–2 43. 80
Vertical guidelines, n 75, paras 31–38.
81
Ibid, esp. at paras 60–6 4. 82
Ibid, paras 43–45. 83
Ibid, para 31(d).
84
Commission Guidelines on technology transfer agreements, 28 March 2014 [2014] OJEU C89/3, paras 62–
63. However, the Commission here appears to imply that most software licensing can fit within one of the two
block exemptions.
51
• there are no restrictions on who may take part (in some capacity) in the standard
setting process;
85
Directive 2009/2 4 legal protection of computer programs, [2009] OJEU L111/16, Art 1(1).
86
Ibid, Art 6(1).
87
On the competition law consequences of State compulsion, see Bellamy & Child, n 6, paras 11.004–11.008.
88
Ibid, paras 6.084–6.086.
89
Commission guidelines on horizontal co-operation agreements, [2011] OJEU C11/1, paras 280–2 83.
52
Where a standards body’s procedures do not meet these criteria, exemption may
nevertheless be available. However, where the standard can in fact only be used by
a closed group of major industry operators, or where alternative standards cannot
be developed (ie there is a degree of exclusivity in the standards process), exemp-
tion is not likely. This is particularly true if a standard is adopted in these circum-
stances which becomes the ‘norm’, so that effective market entry cannot take place
without it. FRAND access to standards is an important principle of competition
compliance.90
This insistence on the FRAND requirement causes particular issues where IP
rights (often patents) overlap with the specification of a standard. This can mean
that the standard cannot (continue to) be used by third parties unless royalties are
paid to the IP owner. If the existence of the IP right is not known to the standard
setting body at the time the standard is made, and only becomes apparent after the
standard has been widely adopted, the assertion by the IP owner of his right can
seriously impede competition in the (new) market(s) which depend on the use of
the standard.
The Commission has insisted in the past that standard setting bodies take steps
to reduce this risk of ‘patent ambush’. ETSI agreed to amend its procedures to
strengthen the requirement on ETSI members to disclose as early as possible any
IP rights which might read onto a telecommunications standard being considered
by ETSI. This could then mean, in particular, that ETSI could decide whether to
adapt the proposed standard to avoid the infringement or to negotiate FRAND
royalty terms in advance of adoption with the right holder(s).91 Although the pos-
ition of the Commission would appear to cover all kinds of standard setting pro-
cedures, it is worth noting that ETSI is designated as the institute responsible for
telecommunications standards harmonization in the EU, which perhaps explains
the Commission’s particular concern that it should take steps to avoid ‘patent
ambushes’.92
Where a standard is adopted which incorporates one or more IP rights and use of
the standard then becomes essential for competitors on one or more downstream
markets, the Commission has taken the view that the owner of such ‘standard
90
Bellamy & Child, n 6, para 6.087. 91
Commission Press Release IP/05/1565.
92
Directive 2002/21 Electronic Communications Framework Directive, n 57, Art 17(1).
53
An agreement or concerted practice is not the only means through which compe-
tition may be restricted and competition law infringed. Unilateral behaviour can
also restrict competition—either by excluding competitors from the market94 or
by exploiting market power to raise prices or reduce service levels to customers.95
This kind of behaviour by one operator is only likely to be successful in re-
stricting competition where the undertaking carrying it out has a significant
degree of market power. This is essentially why the EU’s e-communications regu-
latory regime—in large part—now only applies to operators having ‘significant
market power’ (SMP). EU competition law similarly only prohibits unilateral be-
haviour where it is carried on by a dominant undertaking in a market.96
Although there is a great deal of similarity between the two concepts of SMP
for regulatory purposes and for competition law dominance—t he wording in the
Framework Directive adopts the language of Article 10297—t hey may be applied
differently. Telecommunications regulation seeks to prevent the most serious re-
strictions in electronic communications markets before they happen, whereas the
prohibition on abuse of dominance only applies (at the earliest) where an indi-
vidual undertaking behaves in a way which may fairly imminently restrict compe-
tition.98 We will consider the enforcement issues this creates below.
The prohibition in Article 102 TFEU on abuse of dominance requires a three-
step analysis:
93
Commission Communication, Setting out the EU approach to Standard Essential Patents, COM(2017) 712
final, 29 November 2017.
94
Art 102 TFEU, paras (c) and (d) in particular. 95
Ibid, para (a). 96
Ibid, first para.
97
Framework Directive, n 57, Art 14(2)— ‘. . . if . . . it [the operator] enjoys a position equivalent to
dominance . . .’
98
See, eg, case C-2 80/0 8P Deutsche Telekom [2010] ECR I-9555; case C-52/0 9 Telia Sonera Sverige [2011] ECR
I-527; see also Bellamy & Child, n 6, at 10.059.
54
Just as possessing SMP is not unlawful, simply being dominant does not infringe
competition law: for infringement of the prohibition, the dominant undertaking
must abuse that dominance.99 We will look later in this section at abusive practices
which are particularly relevant in telecommunications but first we consider the
definitions of the relevant market and of dominance (contrasting them with SMP).
The European Commission has issued two important Notices relevant to abuse
of dominance: its Market Definition Notice100 and the Notice on its Enforcement
Priorities for Article 102 cases,101 both of which give substantial guidance on how
the Article 102 prohibition will apply in practice.
99
Case C-322/81, Michelin v Commission [1983] ECR 3461, para 57. 100
[1997] OJEU C372/5.
101
[2009] OJEU C45/7. 102
Market Definition Notice, n 28, para 9. 103
Ibid, para 1.
104
[2002] OJEU C165/6. The European Commission is currently (mid 2017) consulting on revising these
guidelines.
105
Market Definition Notice, n 28, para 11.
5
sustained price increase, even where it is small. This is likely to be a particular fea-
ture of telecommunications markets where innovation by competitors regularly
changes the type, price, or quality of services available to consumers, as can be seen
in the entry of mobile data services (discussed in Section 10.4.2). Potential market
entry (‘supply-side substitution’) will tend to reduce the market power of incumbents.
But the possibility of potential (future) competition is not normally used as a factor
in defining the relevant market.106 Rather, it can be an important factor in assessing
market power once the market has been defined.
This market definition process is the same as that used for identifying markets for
sector regulation of telecommunications networks by national regulatory author-
ities.107 However, because the purpose of the market definition analysis is different
under the two regimes, any market definition for regulatory purposes is ‘without
prejudice’ to a different view being taken in any individual competition case.108 In
addition, the Commission has recommended that certain markets be particularly
carefully considered by NRAs.109 NRAs are required to review at regular intervals
which operators in their countries might possess SMP—at least in the markets iden-
tified by the Commission.110
10.4.2 Dominance
The definition of the relevant market sets the background for the next step in the
analysis—whether the undertaking under investigation is dominant. A dominant
position will exist where the undertaking has:
. . . a position of economic strength . . . which enables it to hinder the maintenance
of effective competition on the relevant market by allowing it to behave to an ap-
preciable extent independently of its competitors, customers and ultimately of
consumers.111
As a proxy for this test, the (sustained) market share of the undertaking in the rele-
vant market is often used. The CJEU has held that there is a presumption of domin-
ance where the undertaking has a persistent market share of 50 per cent or more.112
Persistent market shares of 40 per cent or above are generally taken as being a strong
indicator of dominance.113
Ibid, para 24.
106 107
SMP guidelines, n 42, paras 40–43.
Framework Directive, n 57, Art 15(1); SMP guidelines, n 42, paras 26–27.
108
109
Commission Recommendation 2014/710/E U of 9 October 2014 on relevant product and service markets
within the electronic communications sector, OJ L 295/79, 11 October 2014.
110
See Chapter 4, at Section 4.6. 111
Michelin, n 99, para 30.
112
Case C-62/86, AKZO v Commission [1991] ECR I-3359, para 61.
113
Although not conclusively; see Case 27/76, United Brands v Commission [1978] ECR 207, paras 108–112.
56
If the market, defined using the SSNIP test, is a narrow one, it is quite likely
that there will be high market shares. The question of supply substitution will
then become very relevant—w ho else can enter the market in the short to
medium term?
For telecommunications markets, where technology often rapidly changes the
method of supplying a service to a market, a strong position in the market can
be quickly eroded by such innovation. ‘Voice-over internet’ services have, for ex-
ample, become substitutes—for many consumer services at least—to using fixed
voice telephony infrastructure. Previous regulatory monopolies—reserving cer-
tain telecommunications services to state-controlled enterprises—have also now
been removed as a consequence of national and EU regulation.114
Although only sustained high market power leads to a finding of domin-
ance, there is no firm rule as to how long the allegedly dominant undertaking
must have held that position. But, where a high market share has been held
for less than about three years—e ven in the electronic communications sector,
characterized by a high degree of dynamic innovation—t hat is unlikely alone
to demonstrate dominance. In contrast, a high and stable market share sus-
tained over a longer period (eg five years) would normally be sufficient to prove
dominance.115
Market shares and other ‘traditional’ methods of measuring market power may
be a poor method of assessing dominance in markets characterized by bidding
for a limited number of large value contracts for inputs needed to supply down-
stream services.116 This issue arises particularly in content markets where rights
owners—for example to sporting events—regularly invite tenders for longer term
licences of rights. We consider the particular issues raised by ‘convergence’ be-
tween telecommunications (transmission) markets and content provision in
Section 10.5.
Some telecommunications markets show little prospect of competitive entry
even after more than two decades of liberalization. A prominent example is in
supplying fixed ‘local loop’ infrastructure connecting domestic premises to the
core fixed communications network. The former incumbent telecommunica-
tions operators—who by and large own this infrastructure element—may not face
any realistic short-term threat of competition in supplying local loops. The cost
of installing a parallel fixed loop from the local exchange to a consumer’s home
is high and unlikely to be recouped in the short to medium term from the rev-
enues for services using it. For this reason there was a concerted effort in the 2000s
Deutsche Telekom, n 7.
117
Originally provided in Regulation 2887/2000, [2000] OJEU L336/4, now replaced by Directive 2002/
118
124
Case C-202/07P, France Télécom v Commission [2009] ECR I-2 369, para 105.
125
Case T-111/96, ITT Promedia [1998] ECR II-2937, esp at para 26.
126
Commission Press Release, 11 April 1997, IP/97/292.
59
an input price to the competitor which (implicitly) advantages its own business
by not allowing the competitor a reasonable margin on sales (‘margin squeeze’).
In France Télécom/Wanadoo,127 the European Commission found that Wanadoo
(a subsidiary of France Télécom—now Orange) had priced its ADSL services at
below cost from 1999 to 2002. From 1999 to 2001, the Commission found that the
service had been provided at below Wanadoo’s variable cost of supplying the ADSL
service. From 2001 to 2002, the price covered Wanadoo’s variable costs, but not
its total cost of supplying the ADSL service. On appeal—a nd relying on earlier
case law128—the CJEU upheld the General Court decision,129 which found that
supplying a service below the variable cost of producing it—as Wanadoo had done
in the earlier period—is automatically an abuse of dominance.130 Where the price
is above the variable cost but below the total (unit) cost, the price can be abusive if
it is shown to be part of a plan to exclude competitors from the market (predation).
Controversially, the CJEU upheld previous case law that it is not necessary for
the Commission to show a reasonable prospect that the dominant undertaking
could recoup the losses incurred during the period of predation by increasing its
prices afterwards.131 EU case law is thus out of step both with US anti-t rust case
law (which has such a requirement) and with general economic thinking in this
area. Also, the use of variable cost measures for the predation test may be diffi-
cult in telecommunications markets where the variable cost of producing many
services is very low. For this reason, the use of ‘long-r un incremental cost’ (LRIC)
is preferred by NRAs for measuring unfair pricing132 and should also be considered
when applying competition law to allegations of abusive predation in the telecom-
munications sector.
In Wanadoo, the Commission found that, for the later period (2001–2002),
Wanadoo had priced below average total cost—not including an appropriately al-
located amount for fixed costs—as part of a strategy to exclude competing sup-
pliers of ADSL services from the market. The CJEU confirmed that this was also an
exclusionary abuse of dominance.
A margin squeeze will arise where a dominant firm ‘leverages’ its dominance
into a neighbouring market—i n contrast to predation, where the pricing practice
relates to the market in which the undertaking is already dominant. For margin
squeeze, where the difference in price charged for the input services and the
(downstream) retail price for the consumer services supplied using the input are
127
Commission Decision, Case COMP/38.233—Wanadoo Interactive [2005] 5 CMLR 5.
128
Set out in AKZO, n 112.
129
Case T-3 40/03, France Télécom SA v Commission of the European Communities [2007] ECR I-117.
130
[2009] 4 CMLR 25. 131
Case C-202/07P, para 37.
132
eg Ofcom, ‘Consultation on LLU and wholesale line rental charge controls’, 20 April 2013.
560
either negative or not sufficient to cover the costs which the downstream com-
petitor has to incur to provide a competing consumer service (its variable or LRIC
costs for that service), an abuse will occur.133 This may in particular happen where
the undertaking which is dominant in the upstream market also competes in the
downstream (consumer) market.
Clearly the issues around pricing abuses may be closely linked to the regula-
tory regime in place for price controls. As regulatory price control has been drawn
back, use of competition law to prevent distortions of competition in telecom-
munications markets becomes more likely. The interplay between pricing abuses
under competition law and regulation is well illustrated by the Deutsche Telekom
margin squeeze case, which we consider in the next section.
133
Telia Sonera Sverige, n 98. 134
n 7.
135
Commission Decision relating to a proceeding under Art 82 of the EC Treaty, case COMP/C-1/37.451,
37.578, 37.579—Deutsche Telecom AG, OJ L 263/9, 14 October 2003.
136
Case T-27/03, Deutsche Telekom AG v Commission [2008] ECR-I I 477. 137
n 7.
516
fact that RegTP had set the wholesale price for unbundled local loops at or near
DT’s retail prices for telecommunications services did not relieve DT of the obliga-
tion to ensure that the margin between the wholesale price and its retail prices was
sufficient to allow an equally efficient competitor to enter and remain in down-
stream retail markets. Although in practice this might mean that DT’s retail prices
to its own end consumers would increase in the short term, the CJEU decided
that, in the longer term, consumers’ interests were best protected by a competitive
market for unbundled services provided by a number of retail suppliers.
138
Case C-7/97, Bronner v Mediaprint [1998] ECR I-7791.
139
eg copyright in Case C-4 81/01, IMS Health v Commission [2004] ECR I-5039.
140
Hoffmann-La Roche, n 122.
562
the downstream service. Although the CJEU has insisted that there is only one
standard for assessing dominance, in practice a type of ‘super-dominance’ has
been developed.141 Businesses which control an essential facility—a nd so can be
said to be ‘superdominant’—may be required to allow access to that facility under
Article 102 TFEU under certain circumstances. If the access is not ‘essential’—we
will consider what this means in a moment—t hen a refusal to grant access can
never be an abuse. Effectively this limited ‘essential facilities’ theory should stop
market entrants taking a free ride on the innovative (but dominant) firm’s IP rights
simply to introduce ‘me too’ services.
The issue of ‘essential facilities’ can arise in a number of ways. Among the most
common is the requirement for access to operating system software interfaces for
potential entrants to make sure that their own new services can work together with
the dominant firm’s operating system—t he issue in the leading case Microsoft, dis-
cussed below.142
As noted at 10.3.6, standards play an important role in telecommunications
markets. As well as ensuring that the agreements forming the standard setting
body do not infringe the Article 101 prohibition,143 unilateral behaviour by mem-
bers of the body may also be abusive contrary to Article 102. Where a standard set-
ting body is creating a new standard, it may not be fully aware of all of the IP in the
field—particularly if it is not put in the public domain. Thus the new standard may
inadvertently require users of the standard to adopt infrastructure or use software
which infringes the IP rights of the dominant undertaking—and they will have
to pay a royalty for this (a ‘patent ambush’).144 Access to such ‘standard essential
patents’ (SEPs) may thus be required for a particular set of services: and refusal to
allow access to them on FRAND terms may be an abuse.
The decisions of both the European Commission and the General Court in
Microsoft145 illustrate how the ‘essential facilities’ doctrine applies in innovative
technology markets. Among other behaviour, Microsoft—which at the time had
over a 90 per cent share of all installed PC operating system software (when PCs
were the main method of accessing online information etc)—refused to allow ac-
cess to the full set of interface information needed for rival programmers to create
new products for use on PCs using Microsoft operating software.
The General Court upheld the Commission’s infringement decision, finding that
this was an abuse, and confirmed that—a lthough refusal to license an IP right is
not normally abusive for a dominant undertaking—u nlawful abuse will neverthe-
less arise from a refusal to license if the following conditions are met:
141
eg in Bronner, n 131. 142
Case T-201/0 4, Microsoft v Commission [2007] ECR II-3601.
143
See Section 10.3. 144
As in Commission decision COMP/386.36, Rambus, 9 December 2009.
145
n 142; Commission decision of 24 March 2004 COMP/C3/37.792, [2007] OJEU L 32/2 3 (summary).
563
• the use of the right must be indispensable to market entry in the neighbouring
(downstream) market;
• the refusal to license excludes any effective competition in the downstream
market;
• the refusal prevents the emergence of a new product or service; and
• there is a clear consumer demand for that product or service.
Given Microsoft’s market position in operating systems, and the fact that it was it-
self offering downstream competing programs and services, the Court had no dif-
ficulty in upholding the Commission’s findings that these criteria had been met.
The Court rejected Microsoft’s argument that forcing it to license its IP—even in
these limited circumstances—would effectively remove all value from its IP by al-
lowing any entrant a ‘free ride’ on its rights.
The Court also noted that there was no need for the Commission to show that
competitors had been already excluded from the market—potential exclusion of
competitors wishing to supply a new product was sufficient.
Microsoft’s market position at the time was unusual—but not unique in tele-
communications markets—as its PC operating system was in effect the industry
standard. A similar set of issues arises in the ‘traditional’ standard setting context
if IP rights are indispensable for the use of a standard and (if the standard is widely
used) ‘essential’ for potential competitors.
The Rambus case illustrates the way in which IP right owners can use the
standard setting process to ‘ambush’ competitors and thus to charge abusively
high (exclusionary) royalties on their SEPs.146 The Commission reached a provi-
sional finding that Rambus had intentionally concealed the existence of various
patents or patent applications during the process for setting the standard for
DRAM micro-processors. The case did not proceed to an infringement decision
since Rambus gave binding legal commitments to the Commission to offer a bun-
dled worldwide licence for each of its SEPs to potential users on FRAND terms and
setting a maximum royalty rate for these licences.
146
Case COMP/38.636, Rambus, 9 December 2009 discussed Schellingerhout, R and Cavicchi, P, ‘Patent
Ambush in Standard-setting: the Commission Accepts Commitments from Rambus to Lower Memory Chip
Royalty Rates’ (2010) 1 Competition Policy Newsletter 32.
654
case, the Commission imposed a fine of €1 billion on Intel for abuse in the CPU
market. It had used concealed payments to customers to exclude competitors.147
The Commission will also normally require abusive conduct to cease if this has
not already happened.
However, other remedies are also used—particularly in cases of non-pricing
abuse. Remedies can be imposed both following an administrative competition
investigation by the Commission (or a national competition authority) or by a na-
tional court in litigation of a competition dispute involving an abuse of dominance.
147
Commission Decision 13 May 2009, Intel [2009] OJEU C 227/13; Case T-2 86/0 9, Intel Corp v Commission
[2014] 5 CMLR 9; Case C-413/14 P, [2017] 5 CMLR 18, which set aside the judgment in T-2 86/0 9 and remitted the
case to the General Court.
148
Commission Memo 09/516, 24 November 2009. However, Qualcomm was again notified in 2015 of a
Commission investigation into its pricing practices for chipsets made using the SEPs—Commission Press
Release, 8 December 2015, IP/15/6271.
149
Case C-170-13, Huawei Technologies v Commission [2015] 5 CMLR 14.
56
In practice, national courts may be better placed to apply these principles than a
competition authority. They are often called on to resolve IP disputes, including as
to royalty and other terms, and generally have greater expertise in doing so.
The CJEU Huawei principles were applied in the English courts in the parallel
case of Unwired Planet v Huawei.150 The Patent Court gave some guidance on how
FRAND terms should be reached. It noted that the FRAND undertaking is an in-
dependent obligation to offer terms under French law (the governing law of ETSI)
and does not depend solely on the possible existence of a competition law in-
fringement. This obligation does not mean that the court can compel a contract on
FRAND terms. If no agreement is reached, IP remedies are available: if the patent
owner refuses to agree FRAND terms, the court should refuse to enforce his patent
against the market entrant. If the entrant refuses to accept FRAND terms offered,
and operates without a licence, then normal patent infringement remedies are
available to the patent holder—assuming of course that the patent in question is in
fact infringed by the new entrant.
As to the method of reaching FRAND terms, the court held that the starting
point should be a licence which would be negotiated between a willing licensor
and licensee in the absence of the FRAND undertaking. In order to reach that
point, benchmarking against comparable licences or use of decisions of other
courts in setting terms would be useful. In a later hearing, the court granted a final
injunction requiring Huawei to enter a FRAND agreement, although the injunc-
tion could (unusually) be varied if circumstances changed significantly.151
This position appears sensible on its face. It may, nevertheless, put an SEP owner in
a difficult position in deciding when the negotiations on FRAND terms for access
to the ‘standard essential’ IP right have broken down, so that an application for an
injunction by the right owner becomes permissible.
10.5.1 Introduction
We have noted the impact of convergence in Section 10.1, which illustrates the dif-
ficulty of seeing telecommunications as entirely separate from content. Actors in
one sector may affect the other sector. Bottlenecks may also occur where telecom-
munications providers offer content. Internet intermediaries may act as bottle-
necks too; they are also now providing content, again blurring the boundaries
between content provider and intermediary. The boundary between telecommu-
nications operator and intermediary is also unclear: some social media platforms
offer messaging services and ‘voice chat’ (eg Steam). As noted, there is consider-
able vertical integration. A consequent concern relates to triple play bundles—
usually comprising fixed telephony, TV and (broadband) internet156 which have
153
Decision, 29 April 2014, press release IP/14/4 89, [2014] OJEU C344/6. 154
n 149.
155
Ibid, para 60.
156
OECD, ‘Broadband bundling: Trends and Policy Implications’ Digital Economy Paper No. 175, (OECD
Publishing: Paris, 2011), <http://d x.doi.org/10.1787/5kghtc8znnbx-en>, at 5.
567
almost entirely replaced the retail offers for individual retail services in electronic
communications—a nd now quad play (usually triple play plus mobile).157
157
Generally see, OECD, ‘Triple and quadruple-play bundles of services’, 18 June 2015, <http://w ww.oecd-
ilibrary.org/content/workingpaper/5js04dp2q1jc-en>.
158
The SSNIP test is very similar to that used in relation to defining markets within the Telecommunications
Framework—see Section 10.4.
159
Torsten Körber, ‘Common Errors Regarding Search Engine Regulation–a nd How to Avoid them’, (2015)
ECL Rev 239, 241.
160
Commission, Notice on the definition of relevant market, n 28.
161
Harrison, J, and Woods, L, European Broadcasting Law and Policy (Cambridge: Cambridge University
Press, 2007), 147–151.
162
See further Chapters 14 and 15.
163
See eg Cases COMP/J V 37, BSkyB/Kirch Pay TV, decision of 21 March 2000, OJ C 110, 15 April 2000, at
45; IV/M.993, Bertelsmann/Kirch/P remière, decision of 27 May 1998, OJ L 53, 27.2.1999, at 1; COMP/M.2211
Universal Studio Networks/De Facto 829 (NTL) Studio Channel Ltd, decision of 20 December 2000, OJ C 363, 19.
December 2001, at 31; COMP/J V 57 TPS, decision of 30 April 2002, OJ C 137, 8 June 2001, at 23.
658
164
Commission Decision, Case IV/M.469, MSG Media Service, 94/922/EC OJ [1994] L364/1.
165
Commission Decision COMP/M.2876, Newscorp/Telepiu.
166
Commission Decision 18 July 2007 Case M.4504, SFR/Télé 2 France, para 46; Commission Decision 21
December 2010 Case M.5932, News Corp/BSkyB, paras 103–105; and Commission Decision 15 April 2013 Case
M.6880, Liberty Global/Virgin Media, para 44.
167
Liberty Global/Virgin Media, ibid, para 47.
168
See eg Commission Decision COMP/M.4066, CVC/SLEC 20 March 2006.
169
European Commission Case COMP/37.398, UEFA Champions League, Decision 2003/778/EC [2003] OJ
L291/25; Canal+/RTL/GICD/J V.
596
Liberty Global/Ziggo, the Commission left the question open. A similar reluctance
to engage with this issue can be seen in the UK.170 As noted at Section 10.4, the def-
inition of a market may have significance for the issue of dominance (for Article
102 TFEU purposes, or even in the context of a merger).
170
Anticipated acquisition by BSkyB Broadband Services Limited of Easynet group plc, decision 30
December 2005 (published on 13 January 2006); OFT, Anticipated Merger of NTL Incorporated and Telewest
Global Inc, decision 30 December 2005 (published 10 January 2006), para 17. For an analysis on quad play, see
CMA report on the proposed BT/E E merger, 15 January 2016.
171
Inquiry into e- Commerce, at <http://ec.europa.eu/competition/a ntitrust/sector_ i nquiries.html>,
para 17.
172
See eg Commission Decision IV/36.888, Football World Cup [2000] OJ L5/55.
173
European Commission, Guidelines on the effect on trade concept in Articles 81 and 82 of the Treaty,
[2004] OJ C101/81, para 19; see also European Commission ABC/G énéral des Eaux/Canal+/W HSmith (Case IV/
M.110) [1991] OJ C244 on the transnational nature of sports broadcasting.
570
174
For earlier approaches see Commission Decision, Case IV/32.524, Screensport/E BU [1991] OJ L63;
Commission Decision Case IV/32.150, EBU/Eurovision [1993] OJ L 179, overturned on appeal in Case T-528,
542-3 and 546/93, Metropole television SA (M6) v Commission [2002] ECR II-3805.
175
European Commission Case COMP/37.398, UEFA Champions League, Decision 2003/778/EC, [2003] OJ
L291/25.
176
On the feasibility of clubs selling separately, see Toft, T, Sport and Competition Law (Comp/C .2/T T/hvds
D(2005)), 5.
177
Film Distribution on Pay TV (Case AT.40023, Cross border access to pay TV) (C(2016) 4740 final), 26
July 2016.
178
Inquiry into e-Commerce, n 171, para 69. 179
Ibid, paras 70–71.
180
eg Intel’s work on virtual reality to give a more ‘immersive’ experience.
571
181
Art 102(d) TFEU.
182
Case C-95/0 4P, British Airways EU:C:2006:133, paras 70–71. See also Section 10.4.
183
Microsoft, n 142.
752
of new apps and services. Here, the underlying concern is the threat to innovation.
The Commission reached a preliminary conclusion that Google abused its dominant
position. This reflects its approach to tying and pre-installation in Microsoft where
both the Commission and the Court noted the impact of the need to use WO/S stand-
ards on innovation and consumer choice.184
There are more subtle forms of exclusion, exemplified by the bundles offered to
consumers: triple play and quad play. While such deals may benefit the consumer in
the short-term, particularly through lower prices, there are concerns that the ten-
dency to bundling favours the larger players—which can offer the range of services—
rather than smaller operators and new entrants to the market. Within this sector, as
with Microsoft, there is the question about how sophisticated we expect a consumer
to be. Is it reasonable for consumers to be expected to know about and to access alter-
native products which compete with individual elements of the bundle?185
Refusal to supply (as opposed to conditional dealing or tying) is another con-
cern arising from a vertically integrated market, especially where certain types of
content have been seen as a significant factor in successful market entry, as noted
in Section 10.5.3. Can the essential facilities doctrine (discussed in Section 10.4.3)
apply here, either as regards a telecommunications operator’s access to content or
a content provider’s right to a distribution network? As regards access to content,
the resulting new ‘product’ (as required by the CJEU jurisprudence) need only be
something that is not a duplicate.186 As regards the requirement for customer de-
mand, only potential consumer demand needs to be shown to satisfy the test.187
The answer to both of these questions might depend on whether the Commission
and the Court would look at the specific content (ie the match, film, or series) or
rather the genre or type of content (romantic comedies, documentaries) to as-
sess the question of duplication. It seems likely that the latter approach would
be taken, meaning that a content offering with the same type of content would
not be new. It is more likely that showing the content in a new way—for example
a novel presentation of mobile clips of highlights of sporting matches—would
be considered ‘new’.188 It seems hard to argue, however, that premium content is
184
See generally, Ibáñez Colomo, P, ‘Restrictions on Innovation in EU Competition Law’, (2016) EL Rev
201. As at 1 January 2018, the Commission had yet to reach a final decision on these objections: see Case
COM40099, Google Android at <ec.europa.eu/competition/a ntitrust/c ases/i ndex.html>.
185
For discussion of consumers see eg Tušek, I, ‘EU Competition Law Policy versus Intellectual Property
Rights: A Study of the Microsoft Case’, [2010] CYELP 103, 121.
186
Case C-418/01, IMS Health v NDC Health [2004] ECR I-5039.
187
Joined Cases C-241/91 P and C-242/91 P, Radio Telefis Eireann (RTE) and Independent Television
Publications Ltd (ITP) v Commission (‘Magill’) [1995] ECR I-743.
188
Magill ibid—t he listings were already available within newspapers—w hat was new was the weekly guide.
753
189
See IMS Health, n 186, para 28, though note that Microsoft accepts indispensability for competition
within the market—see para 377 and note discussion by Ibáñez Colomo, n 184, at 213.
190
Bronner, n 138.
191
European Commission, Antitrust: Commission fines Google €2.42 billion for abusing dominance as
search engine by giving illegal advantage to own comparison shopping service—Factsheet (MEMO/17/1785),
27 June 2017, <http://europa.eu/rapid/press-release_ M EMO-17-1785_en.htm>.
192
Contrast Vesterdorf, B, ‘Theories of Self-P referencing and Duty to Deal—Two Sides of the Same Coin?’,
(2015) 1(1) Competition Law & Policy Debate 4, <https://papers.ssrn.com/sol3/papers.cfm?abstract_
id=2561355>; and Petit, N, ‘Theories of Self-P referencing Under Article 102 TFEU: A Reply to Bo Vesterdorf’ (29
April 2015), <http://d x.doi.org/10.2139/ssrn.2592253>.
547
the focus has shifted to a concern about the impact on innovation and the harm to
competition as a consequence.193
This reasoning may be contrasted with the approach of the English courts in a
case handed down before the Google decision: Streetmap v Google.194 Streetmap
argued that Google bundled Googlemaps with Google Search. By using a display
at the top of its search results of a clickable image from Google Maps, and no other
mapping provider (including Streetmap), following geographic queries, Google
was abusing its dominant position in the market for online search and online
search advertising. In essence, this is a discrimination argument that Google was
using its dominant position with the intent or effect of undermining competitors’
ability effectively to compete,195 by leveraging Google’s dominance in ‘search’ into
a market (geographical services) other than that in which Google was dominant.196
In making its claim, Streetmap relied heavily on the reasoning in Microsoft.
The English High Court disagreed with this approach. It held this case did not in-
volve bundling in the sense of Microsoft because, although Google would provide
the links to Google Maps, the user was under no obligation to click on those links.
In Microsoft, although users could obtain alternative Windows Players, there were
several barriers in their way to doing so. The Court also suggested that, following
Microsoft, there should be a reasonable likelihood that harm would ensue.197 In the
view of the Court, since the concern related to a neighbouring market rather than
the market on which Google was dominant and where the market structure was
already undermined, harm could not be assumed.198 The High Court accepted that
it was Google’s intention to improve its search engine and increase user conveni-
ence rather than to damage competition. The impact on Streetmap was therefore a
‘by-product’ rather than attributable to Google.199
193
Ibáñez Colomo, n 184, 212. 194
Streetmap EU v Google Inc & Ors [2016] EWHC 253 (Ch).
195
Ibid, para 63. 196
Ibid, paras 59–6 0. 197
Ibid, para 88.
198
In coming to this conclusion, the Court referred to Whish and Bailey, Competition Law (8th edn, Oxford
University Press, 2015), 212; Faull and Nikpay, The EU Law of Competition (3rd edn, Oxford University Press,
2014), para 4.929.
199
Case C-2 3/14, Post-Danmark II, ECLI:EU:C:2015:651, para 47.
57
effects of future changes to the structure of markets. Where a merger (or the cre-
ation of a new ‘independent’ joint venture) may lessen competition by weakening
the number or relative strength of remaining competitors, national or EU merger
controls may apply.
As well as being different from behavioural competition law by looking (pri-
marily) at market structure, merger control is necessarily also forward looking:
in most cases the merger will not yet have taken place. It shares this characteristic
with most of the electronic communications sector regulation applied by NRAs.
Competition principles are applied to the authorities’ prediction of the likely
market changes the proposed merger will cause. These two significant differences
in approach between merger control and behavioural competition enforcement
mean that merger control in electronic telecommunications markets has been ap-
plied rather differently from behavioural competition law examined previously.
‘Merger’ control (the ‘control of concentrations’ in EU law) operates both
under EU law and under national merger control regimes, with the European
Commission having the sole right to examine the largest mergers. The test
which the Commission applies in deciding whether to block or approve a merger
is whether it would lead to a significant impediment to (lessening of) effective
competition in the EU. 200 In practice this means that, wherever there is a poten-
tial overlap in the parties’ activities and one undertaking has a market share of
about 20 per cent or more, the parties should start looking carefully to see if the
merger proposed has an adverse effect in its own or in related markets.
Where national merger control applies, national competition authorities use
their own rules. In practice these are very similar—in most countries—to those
in the EU Merger Regulation. For example, in the UK the CMA also applies a sub-
stantive test modelled on that in the Regulation.201 Where there are differences
between national and EU practice, these tend to be primarily procedural—for
example, under UK merger control law, Ofcom may be required to report to the
Secretary of State on media plurality issues before the CMA makes a decision on
some ‘mixed’ mergers in the electronic communications sector.202 There is no EU
law requirement that national merger control procedures (including clearance
timetables) are harmonized with those of the European Commission. There is of
course significant cooperation between the Commission and national competi-
tion authorities in merger control matters, as there is with behavioural competi-
tion law enforcement.
A detailed explanation of the scope of the EU Merger Regulation, and its proced-
ures, is outside the scope of this work.203 For our purposes, it is sufficient to bear in
mind that:
• The Merger Regulation only applies to the largest mergers in the EU—where the
combined turnover of all the parties involved in the concentration, in their last
financial year, exceeds €5 billion.
• The Regulation does not apply to mergers whose economic effects are likely to be
felt in just one Member State. Even if the combined turnover of the undertakings
concerned exceeds €5 billion, the Regulation does not apply where at least two-
thirds of the combined turnover of the parties within the EU (so, disregarding
turnover elsewhere in the world) is earned in one and the same Member State. In
that case only the merger control procedures of that Member State apply.204 The
BT/EE merger was dealt with by the CMA rather than the Commission as a result
of the ‘two thirds’ rule.
• To avoid the need for multiple merger clearances in several Member States the
Regulation will also apply where the turnover in at least three Member States
of at least two of the undertakings concerned in the merger is above certain
thresholds, and the combined worldwide turnover of all the undertakings con-
cerned exceeds €2.5 billion.205
• Member States may not apply their national competition rules to a merger falling
under the Regulation—so including national behavioural rules.206 They may,
however, take measures under non-competition legislation to protect their other
‘legitimate interests’ as a result of a merger.207 These interests include ‘public
security’, ‘plurality of the media’, and ‘prudential rules’. The second interest is
particularly important for Member States wishing to ensure a wide variety of
choice in broadcasting and newspapers.208 This exception to the otherwise clear
division of responsibility between Brussels and the Member States has had a sig-
nificant impact on the approach to mergers in the broader electronic communi-
cations sector, with a number of mergers being subject to national intervention
on media plurality grounds.209
203
See Faull and Nikpay, and Whish, n 198. 204
Art 21.2 and 21.3 Merger Regulation, n 8.
205
Art 1.1 Merger Regulation, n 8. 206
Art 21.3 Merger Regulation, n 8.
207
Art 21.4 Merger Regulation, n 8.
208
As for example in the (subsequently abandoned) proposed acquisition of BSkyB by News Corporation—
see the Secretary of State’s statement in June 2011 at <http://w ww.culture.gov.uk/news/news_stories/8259.
aspx>.
209
eg Twenty-First Century Fox/Sky plc: European Commission clearance decision M8354, 7 April 2017,
[2017] OJEU C238; UK intervention notice, 16 March 2017 but contrast the Liberty Global/Z iggo (2014) merger
where the request for national jurisdiction was turned down. European Commission Press Release IP-16-271,
3 August 2016.
75
• So-called ‘concentrative’ or ‘full function’ joint ventures (where the parties give
up independent activity in the field of the joint venture) are normally dealt with
under the Merger Regulation. In contract, where the parties both remain as in-
dependent suppliers in the same or related markets as the new joint venture,
approval of the terms of the joint venture restricting the activities of the parents
is dealt with under the Article 101 criteria.210
Articles 4(4) and 9(2) of the Merger Regulation permit the transfer of the merger
investigation back to a national competition authority if the concentration
threatens to affect competition significantly in a market within that Member
State, which presents all the characteristics of a distinct market. In a number of
mergers involving the mobile telephony sector, the national competition author-
ities have requested jurisdiction on this basis, but the European Commission
has rejected the applications.211 This approach in the telecoms sector seems to go
against the trend in relation to requests under Article 9 more generally. It may be
that the political significance of the sector, as well as the likely complexity of the
cases, are factors. Moreover, this does not mean that the national competition au-
thority or regulators are excluded entirely: the Commission has noted in its final
decisions the close degree of co-operation between it and the relevant national
body.212
The Commission has generally approved mergers in the telecommunications
sector provided that access to network infrastructure for third parties is not unrea-
sonably restricted.213 The dynamic nature of most telecommunications markets,
and the regulatory regime requiring operators with SMP to allow infrastructure
access on regulated terms, will often mean that a transaction can be cleared—
possibly with some changes to address any specific European Commission or
regulator concerns.214 Many mergers which are not between competing communi-
cations services providers do not give rise to substantial competition concerns.215
210
eg Commission Decisions 96/5 46/EC [1996] OJEU L 239/2 3 (IV/35.337—Atlas), and 96/5 47 [1996] OJEU
L239/57 (IV/35.617—Phoenix/GlobalOne).
211
See eg Hutchison 3G Austria/Orange Austria (2012); Telefonica Deutschland/E-Plus (2014); Hutchison
3G UK/Telefonica UK (2016).
212
See eg letter from CMA to European Commission of 11 April 2016 in Hutchinson 3G UK/Telefonica
merger, <https://w ww.gov.uk/government/uploads/s ystem/uploads/attachment_d ata/fi le/515405/C MA_
letter_to_Commissioner_ Margrethe_Vestager.pdf>.
213
An early case where a ‘pure’ telecommunications merger was, nevertheless, prohibited, is Case M.1741,
MCI/Worldcom/Sprint, [2003] OJEU L300/1.
214
See commentary in Bellamy & Child, n 6, para 12.066.
215
See eg the BT/EE decision by the UK CMA, 15 January 2016, clearing the acquisition by BT (the UK’s
largest fixed communications services provider) and EE (the largest mobile provider), at <https://a ssets.
publishing.service.gov.uk/media/56992242ed915d4747000026/BT_E E_fi nal_report.pdf>.
578
For example, the merger of BT and EE was approved in 2016 despite concerns,
expressed by competitors and others, that the combination would increase BT’s
market power. BT and EE operated in largely separate segments of the telecom-
munications market—fi xed line and mobile respectively—a nd there was very little
overlap in their respective businesses at the time of merger. Other concerns about
the increase in BT’s power in telecommunications generally were not thought to
be specific to the merger: the CMA noted that they could be addressed in Ofcom’s
wider review of the UK telecommunications market.216
In contrast, the European Commission blocked the merger of two of the four main
UK mobile networks—Three and O2—in 2016. It found that the merger would leave
only three MNOs in the UK, which significantly reduced competition in the market
and which would likely have resulted in higher prices for mobile services in the UK
and less choice for consumers than without the deal.217 This merger illustrates that
the Commission seems to have a preference for structural (divestment) remedies in
MNO-to-MNO merger cases, so as to introduce a replacement entrant in the market.
Thus in Wind/H3G, the Commission accepted the merger between two out of the
four MNOs operating in Italy because the parties offered remedies involving the di-
vestment of certain assets necessary to enable a new competitor, the French oper-
ator Iliad, to enter the Italian market as a fourth MNO.
The contrast between the approach in Three/O2 and BT/E E also reiterates
the point that, as yet, while triple or quad play issues may be raised in merger
procedures, they have yet to be a decisive factor. In BT/E E they were not con-
sidered separately. The market is changing rapidly and it has been suggested
that the Commission’s approach is developing to take into account the potential
foreclosure effect of convergent mergers. 218 Commissioner Vestager suggested
that, if anything, quad play is beneficial because it can operate to lower prices
to customers. 219 In the cases that have come before the Commission there has
been no evidence that quad play would squeeze standalone companies out of
the market.
Concerns have also been raised in other recent mobile-mobile mergers that the
number of competing MNOs in some EU Member States is falling below the level
216
15 January 2016, n 215. See CMA press release in particular.
217
M.7612, Hutchison 3G UK/Telefonica UK, 11 May 2015, [2015] OJEU C357/15.
218
Manigrassi, L, Ocello, E, and Staykov, V, ‘Recent Developments in Telecoms Mergers’, (2016) 3
Competition Merger Brief 1, 6–7.
219
Vestager, M, ‘Competition and investment in telecoms’, Speech to CERRE Dinner Debate, 28
November 2016, <https://ec.europa.eu/commission/commissioners/2014-2019/vestager/a nnouncements/
competition-a nd-i nvestment-telecoms_en>; see also Curwen, P and Whalley, J, Mobile Telecommunications
Networks: Restructuring as a Response to a Challenging Environment (Cheltenham: Edward Elgar, 2014),
208–209.
579
220
Mergers which create an SIEC must be prohibited under the Merger Regulation, n 8, Art 2.3.
M.6497, Hutchison 3G Austria, 12 December 2012, [2013] OJEU C224; M.6992, Hutchison/Telefonica
221
Ireland, 28 May 2014, [2014] OJEU C264; M7018, Telefonica Deutschland, 2 July 2014, [2015] OJEU C86.
222
See the discussion of these cases in European Commission Competition merger brief 1/2104, 10 at <http://
ec.europa.eu/competition/publications/c mb/2014/CMB2014-01.pdf>.
223
Art3(4) Merger Regulation, n 8, simply stipulates that a joint venture may be a concentration: it does not
define ‘joint venture’.
224
Commission Notice 16 April 2008, [2008] OJEU C95/1 (jurisdiction Notice), paras 91–109.
225
Ibid, paras 98–101. 226
Merger Regulation, n 8, Art 4.
580
227
Ibid, Art 2(4).
228
Commission Notice, 5 March 2005, [2005] OJEU C56/2 4 (‘ancillary restraints’), paras 31–41.
229
O2 UK/T-Mobile [2003] OJEU L200/59 and T-Mobile Deutschland [2004] OJEU L7532—on appeal Case
T-328/03, O2 (Germany) v Commission [2006] ECR II-1231.
230
See further Chapter 8. 231
T-328/03, n 229, para 109.
518
232
eg in Hutchison/V impelCom (Wind)—t he merger was only approved when commitments were offered
by the parties to set up effectively a new mobile network in Italy, run by the French operator Iliad, to ensure
sufficient competition on the Italian market: M7758, Hutchison Italy 3G/Wind, 1 September 2016, [2016] OJEU
C391/7.
233
M.4942, decision of 2 July 2008, [2009] OJEU C13/8. 234
Merger Regulation, n 8, Art 21(4).
528
Secretary of State on the advice of Ofcom.235 This regime does not target vertical
integration—mergers involving media companies and telecommunications com-
panies—but includes other kinds of media mergers.
Ofcom and the DTI (then responsible for media mergers, now DCMS) have pub-
lished guidance on the factors to be taken into account when the Secretary of State
is considering intervening in a merger on media plurality grounds.236 This applies
whether the merger falls within EU or purely UK merger control. There are three
measures for media plurality—availability, consumption, and impact.237 These are
assessed on a case-by-case basis. Separately from the merger control process—but
often linked closely to it—t he Secretary of State may also consider if the acquirer is
fit and proper to own the media outlet in question.238
The CMA reports both on any competition concerns arising from the merger
and on the public interest (media plurality) concerns identified.239 However, un-
like purely ‘economic’ mergers, it is for the Secretary of State to decide if any en-
forcement action will be taken in case of an adverse CMA report.240
10.7.1 General
There are powers for the competition authorities to carry out investigations or
inquiries into an economic sector at both national and EU level. While Article
101 and 102 (and the corresponding provisions of the Competition Act) look at
anti-competitive agreements or abuses of dominance by individual businesses,
a market investigation aims to determine whether the process of competition is
working effectively in markets as a whole. Market investigations tend to be used
when the provisions regarding anti-competitive agreements and those regarding
abuse would be (or have been) ineffective. They may be appropriate, for example, in
the case of non-coordinated parallel conduct. The European telecommunications
235
Enterprise Act 2002, s 42.
236
DTI, ‘Enterprise Act 2002: Public Interest intervention in Media Mergers’, May 2004, <http://webarchive.
nationalarchives.gov.uk/20100512170615/http://w ww.bis.gov.uk/fi les/fi le14331.pdf>. Ofcom, ‘Measuring
media plurality’, 19 June 2012, <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 031/57694/measuring-
media-plurality.pdf>.
237
Measuring media plurality, ibid, para 1.5.
238
eg 21st Century Fox’s acquisition of control over Sky News, referred to the CMA for public interest re-
view, 12 September 2017, <https://w ww.gov.uk/government/uploads/s ystem/uploads/attachment_data/fi le/
644186/DCMS_letter_to_Sky_Fox_12_Sep_2017_ _1_.pdf>.
239
Enterprise Act 2002, s 47. 240
Ibid, s 54.
538
sector inquiry (discussed at Section 10.7.2 below) was noteworthy because until
that point the Commission had not used its sector inquiry powers, perhaps be-
cause the powers under Regulation 17/62 were focused on individual firms’ behav-
iour rather than the state of the market. The inquiry marked an increase in interest
in use of this tool, even before Regulation 1/2003 was enacted (and EU sector en-
quiries given a proper legal basis241),in sectors where market structures were such
that effective competition would be hard to sustain. The overarching objective is
to allow the relevant body (national competition authority, telecoms regulator, or
the European Commission) to understand a sector in which the market does not
appear to be functioning effectively, and to identify the reasons why.
The outcomes of such inquiries and investigations may vary considerably.
The information gained may subsequently be used in competition enforcement
against individual undertakings. It may also prove useful in other contexts, such
as defining markets in the context of a merger. Therefore, market investigations
and inquiries should properly be considered as forming part of the EU competition
law enforcement tool-k it.
241
Regulation 1/2003, n 8, Art 17. 242
Formerly, Art 12 Regulation 17/62.
243
Case 60/81, IBM v Commission [1981] ECR 2639. 244
Regulation 1/2003, n 8, Art 20.
584
245
See eg Case 374/87, Orkem v Commission [1989] ECR 3283; Case T-112/98, Mannesmannröhrenwerke v
Commission [2001] ECR II-729; legally privileged information is subject to an exception here too.
246
<http://ec.europa.eu/competition/a ntitrust/sector_ i nquiries.html>.
247
Directive 92/4 4 on the application of open network provision to leased lines, [1992] OJ L 165/27, as
amended by Directive 97/51.
248
See further Chapter 8, at Section 8.3.4.4.
58
incumbent’s failure to rebalance line rental tariffs with their underlying costs
(both cases discussed at Section 10.3).
At the time of the 3G inquiry, the 2G mobile market was considered to be highly
competitive. The 3G market—which allowed the transfer of all kinds of data at
high speeds by mobile networks—was not equally so. The 3G market contained a
number of problematic features, notably:
• high sunk costs due to the cost of spectrum licences and costs associated with
network roll-out;
• barriers to entry because of the limited amount of spectrum available;
• strong incumbent operators including former 2G operators as well as fixed tel-
ephony operators;
• oligopolistic market structure (in most Member States, there were between
three and five network operators); and
• economies of scale and network externalities.
One of the key issues here was market definition. The Commission concluded
that mobile and fixed platforms for content were distinct markets, as there was
limited substitutability, especially given the physical characteristics of mobile
handsets, costs of usage, and inability to watch as a group. Further, a driver for the
uptake of 3G mobile was the availability of (audiovisual) content and the rights
to the most desirable content were in the hands of a few key (media) operators.249
The inquiry identified four areas of particular concern: cross-platform bundling of
rights, excessively restrictive conditions on exploiting rights (ie in terms of trans-
mission length and timing), joint selling, and exclusivity. As discussed above, the
Commission has taken action in relation to long, exclusive sports rights packages.
The inquiry noted that there were substantial costs involved in developing these
new services and so some security of return was needed.
The most recent inquiry was that into the e-Commerce sector, which has led to a
range of initiatives including some legislative proposals, as well as some enforce-
ment action under Articles 101 and 102 TFEU.
Enterprise Act 2002, s 131 (replacing the earlier provisions under the Fair Trading Act 1973 (FTA));
250
see generally OFT Market investigation references (OFT511) and Competition Commission, Guidelines
for market investigations (CC3 (revised)), both as amended by CMA Market Studies and Market
586
in place under the Enterprise Act, replacing it with a single body—t he CMA. Prior
to the ERRA, the initial consideration of the matter was carried out by the Office
of Fair Trading (OFT), which could choose to refer a matter to the Competition
Commission (CC) to carry out the market investigation independently. Despite
the change to a single body, the terminology from the old system remains—so the
CMA makes a reference to itself (which is carried out by a specially constituted
board drawn from a pool of possible members). The ERRA also amended the pro-
visions relating to public interest interventions and introduced a new category of
cross-market references.
The CMA may also carry out market studies251 under its general powers in
section 5 of the Enterprise Act. These are examinations into the causes of why par-
ticular markets may not be working well but are separate from a market investi-
gation. Where a market study gives rise to reasonable grounds for suspecting that
a feature restricts or distorts competition, and a market investigation reference
appears to be an appropriate and proportionate response, the CMA may make a
reference. Carrying out a market study is not a prerequisite to a market investiga-
tion, though a market study may allow the CMA to access information central to
carrying out its market investigation.
A recent example of a market study in the digital sector is that into digital com-
parison tools (DCT) (such as price comparison websites). The CMA published
a report at the end of the study. In its final report it put forward a number of re-
commendations (including the need for legislative action) and stated that it will
continue to keep some practices under review (eg non brand-bidding, negative
matching, and non-resolicitation agreements). It opened an investigation into the
behaviour of one comparison website under the UK Chapter 1 prohibition as well
as Article 101 TFEU.252
In addition to the CMA, other public sector enforcers may be involved in trig-
gering a market investigation. Ministers may, under section 132 of the Enterprise
Act, make references for a market investigation. In addition to the criteria to which
the CMA must have regard, a minister must either be ‘not satisfied’ with a CMA
decision not to make a reference or, having brought information to the attention
of the CMA, the CMA has not made a decision on whether to make a reference in a
period that the minister considers to be reasonable.
Investigations: Supplemental Guidance on the CMA’s Approach, (CMA3) January 2014 (revised July 2017);
for overview see eg Coscelli, A and Horrocks, A, ‘Making Markets Work Well: The U.K. Market Investigations
Regime’, (2014) 10 Competition Policy International 24.
251
See generally, OFT Market studies: Guidance on the OFT approach (OFT519), as amended by CMA
Market Studies and Market Investigations: Supplemental Guidance on the CMA’s Approach, (CMA3) January
2014 (revised July 2017).
252
<https://w ww.gov.uk/c ma-c ases/price-comparison-website-u se-of-most-favoured-nation-clauses>.
578
According to section 131(3), ‘conduct’ includes any failure to act (whether inten-
tional or not) and any other unintentional conduct. The boundary between ‘struc-
ture’ and ‘conduct’ may not always be clear-cut. In practice the approach has been
to identify the relevant ‘feature’, rather than worry about how to categorise that
255
<https://w ww.ofcom.org.uk/c onsultations-a nd-statements/c ategory-1/strengthening-openreachs-i n-
dependence> relying on ss 89A and 89B Communications Act which follows the requirements set out in Article
13a of Directive 2002/19/EC (‘Access Directive’), as amended; see notification, <https://w ww.ofcom.org.uk/_ _
data/a ssets/pdf_ fi le/0 026/94940/Final-signed-letter-to-t he-European-Commission-2 81116.pdf>.
256
< https://w ww.btplc.com/U KDigitalFuture/A greed/i ndex.htm>.
257
<https://w ww.ofcom.org.uk/about-ofcom/latest/features-a nd-news/openreach-statement>.
258
See Ofcom’s imposition of WMO obligation on Sky: Ofcom, ‘Review of the pay tv wholesale must-offer
obligation’, 19 November 2015, <https://w ww.ofcom.org.uk/_ _d ata/a ssets/p df_ fi le/0 022/76081/Review-
of-t he-pay-T V-w holesale-must-offer-obligation-.pdf>.
259
See discussion on essential facilities in Section 10.4. 260
Enterprise Act 2002, s 134(1).
261
Enterprise Act 2002, s 134.
589
• the nature and characteristics of the relevant products or services and of any
potential substitutes for these products;
• the nature of the customer base—whether customers are businesses or final
consumers, the extent of customer segmentation in a market, the demographic
profile of the customer base, or the extent to which customers are informed
about the products;
• any applicable legal and regulatory framework;
• industry practices;
• the history of the market, such as recent examples of entry, expansion or exit or
any anticipated significant changes; and263
• market outcomes—t hat is prices, profitability, and innovation—to understand
any resulting harms.
The CMA has a range of remedies which it can use if it finds an AEC as a result of
a market investigation. In this respect, market investigations are different from
both a market study and the Commission’s sector inquiry powers. They rely for
remedies on using competition law mechanisms or on a call for new legislation. A
market investigation may, however, also include recommendations to others, eg
to change existing legislation. In this comparative sense, the market investigation
provisions under the Enterprise Act can be said to have teeth. According to section
134 of the Enterprise Act, the CMA has to decide what action to take to remedy
an AEC: it should have regard to the need to achieve as comprehensive an out-
come as is practicable. In particular, the CMA can order divestment (structural
remedies) or behavioural remedies, such as price undertakings. Its preference is
to adopt structural remedies rather than behavioural remedies, so as to avoid the
difficulties of monitoring ongoing compliance with any undertakings given. It has,
however, no power to fine companies under its market investigation powers.
Prior to the ERRA, the CMA could make a market investigation reference under
the Enterprise Act in relation to a single market (‘ordinary references’). ERRA
added a second category which concern specific features or combinations of fea-
tures that exist in more than one market,264 though this is limited to ‘conduct’ not
‘structure’.265 The Secretary of State may also make a reference in cases that raise
defined public interest issues. Under the Enterprise Act, the CMA would inves-
tigate competition issues, while the Secretary of State would have responsibility
for investigating the public interest issues (now referred to as a ‘restricted public
interest reference’). Under ERRA, there is a new type of reference which requires
the CMA to investigate certain public interest issues together with the competition
issues (‘full public interest reference’).
National security is currently the only specified public interest consideration
in relation to the markets regime, but the Secretary of State may introduce new
public interest considerations. In contrast to the merger regime, media plurality
is not listed.
The most significant recent CMA market investigation relating to telecom-
munications is on the pay TV market. After receiving representations from a
number of competing market participants, including telecommunications oper-
ators as well as consumer groups, Ofcom started an investigation into the pay TV
market in 2007 (ie under the original Enterprise Act regime).266 Ofcom defined
the pay TV market broadly, to include subscription and video-on-demand tele-
vision services on all platforms. It took the initial view that distinct narrow eco-
nomic markets existed for pay TV subscription channels containing premium
sports content and movies, at both the wholesale and retail level. Sky was found
to have market power in these markets. Ofcom argued that bundling efficien-
cies (eg phone services with audiovisual content) would mean that these mar-
kets may be prone to ‘tipping’ towards one retailer, particularly where a retailer
on a particular platform has exclusive control over a core of premium content.
Competition from other platforms would only be a viable constraint if providers
on those platforms had access to comparable content. Ofcom referred the supply
and acquisition of subscription pay TV movie rights and the wholesale supply
and acquisition of packages which include core premium movies channels to
the Competition Commission.267 In the end, the Commission found that there
was no AEC.
The pay TV market investigation is significant for a number of reasons. It is an
example of competitors involving themselves in the regulatory and competition
law process to try to effect an outcome of benefit to them. Concerns raised were
partly related to the increasingly integrated nature of the telecommunications and
265
Ibid, s 131(1) and 131(2A).
266
Ofcom, ‘Pay TV market investigation’, <https://w ww.ofcom.org.uk/consultations-a nd-statements/
category-1/market_i nvest_paytv>.
267
Sports rights were dealt with under Ofcom’s sectoral powers in Communications Act 2003, s 316 to im-
pose a wholesale ‘must offer’ obligation on Sky, an obligation which was successfully challenged before the
courts, see Harrison and Woods, n 161.
591
content sectors, and the impact that access to premium content, as well as bund-
ling of other services, might have. It is also an example of an investigation where
the Competition Commission came to a different conclusion from the sector regu-
lator, albeit quite late in the process.
In its provisional report, the analysis of the Competition Commission had been
similar to that of Ofcom. The change in its view may have been in part because
of market changes, specifically the establishment of ‘over-the-top’ (OTT) services
(delivered over the internet, eg Lovefilm; Netflix) providing similar content to that
of Sky, as well as changes to Sky’s own service. Significantly, to use OTT services,
consumers did not have to buy new hardware—a fact facilitating rapid uptake of
these services. This highlights the difficulty of assessing market power and behav-
iours going forward—a nd consequently whether there is a need for intervention—
in a context where technology and the market are changing swiftly. Indeed, in its
pay TV Statement, Ofcom acknowledged that its investigation came at a time of
‘disruptive change’ in the way in which content was distributed, but while Ofcom
expressed concern about the future of these services, the CC’s report, coming
some time later, found a different landscape—d ifferent even from that which ex-
isted at the time of its provisional report.
10.8.1 Overview
From a public policy standpoint, telecommunications regulation and competition
law enforcement may fill the same purpose—ensuring that as far as possible markets
are open so that consumers get services at the best possible choice, quality, and price.
In most EU Member States, despite this overlap in purpose, competition enforce-
ment and communications regulation are separately enforced by different bodies.
This can lead to difficulties in coherent regulation of electronic communications
services. This was exemplified in the Deutsche Telekom case268 where the regulator
set a wholesale price above some of DT’s retail prices, allowing DT to claim—albeit
unsuccessfully—that it was forced into the anti-competitive margin squeeze.
In order to reduce the possibility of this kind of incoherence in UK markets, the
Competition Act 1998 introduced powers for some sector regulators, including the
telecommunications sector regulator, to apply mainstream competition rules in
268
n 7. The German telecommunications regulator, RegTP had set a regulated interconnection price which
moved at or above Deutsche Telekom’s own retail prices for some equivalent services.
592
their respective sectors alongside their regulatory powers. In practice, the ma-
jority of administrative competition enforcement in the electronic communica-
tions sector in the UK—w ith the exception of merger control—has been done by
Ofcom using this ‘concurrent’ competition power.
269
Competition Act 1998, s 54(2). 270
eg Three/O2, 1 February 2016.
271
Enterprise Act 2002, 131 and Communications Act 2003, s 370. 272
Enterprise Act 2002, s 131.
273
4 August 2010, at <https://www.ofcom.org.uk/__data/assets/pdf_file/0017/72008/pay-tv-movies-decision.
pdf>. The Competition Commission subsequently found that there was no appreciable adverse effect on com-
petition in any pay TV market: report of 2 August 2012 (at 15), <http://webarchive.nationalarchives.gov.uk/
20140402201316/http://www.competition-commission.org.uk/assets/competitioncommission/docs/2010/
movies-on-pay-tv/main_report.pdf>.
539
market issues in content provision (eg pay TV)274—or structural issues in a market
which arise outside a merger context.275
Where anti-competitive market behaviour is suspected in UK electronic com-
munications markets, three enforcement choices need to be addressed. First,
does EU competition law apply to the question? If so, then it must be used 276 and
any outcome from parallel use of domestic powers must be consistent with it.
Second, are regulatory requirements under the electronic communications le-
gislation relevant—for example, is the behaviour a breach of an interconnection
agreement? And third, can UK competition law also apply to the behaviour? These
choices overlap with the question of which authority should consider the issues.
For example, where EU competition law applies, the European Commission may
enforce the law directly—a nd if it does so, then no national authority may act.277
In the majority of cases where anti-competitive behaviour is suspected in UK
markets, either Ofcom or the CMA will be able to act. How is this choice made?
From 2013 the procedure was improved.278 There is now a statutory duty on Ofcom
to consider if using its competition powers would be more appropriate in each
case before it takes action under its regulatory (Communications Act) powers.279
In principle, Ofcom should use its concurrent competition powers before consid-
ering sector regulation. This new duty is backed up by a new power for the govern-
ment to remove all concurrent competition powers from Ofcom if the Competition
Act enforcement tools are not used sufficiently.280
Agreeing which of the CMA and Ofcom should act is likely to be straightfor-
ward. There are well developed communication channels to coordinate on these
issues and where the competition issue is wholly within the scope of Ofcom’s other
powers—electronic communications, broadcasting (TV and radio), and postal
services—it is probable that Ofcom will take the investigation. The CMA and
Ofcom have published a memorandum giving more detail on how they will allo-
cate competition cases between them.281 The factors taken into account when al-
locating a case include relevant sector knowledge, effect of the market behaviour
See Section 10.5.
274
See Section 10.6. In particular, Ofcom has used the possibility of a market investigation to prompt an
275
offer of undertakings in lieu of a reference for BT to separate its wholesale business (Openreach) from its re-
tail offerings, <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 023/47075/consolidated_u ndertakings24.
pdf>.
276
Regulation 1/2003, n 8, Art 3(1). 277
Ibid, Art 11(6).
278
Enterprise and Regulatory Reform Act (ERRA) 2013, ss 51–52 and Sch 14.
279
Communications Act 2003, ss 94 and 96A, as amended by ERRA 2013, Sch 14 paras 17–18; for the broad-
casting sector see s 317.
280
ERRA 2013, s 52.
281
17 June 2014, <https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/502645/
Ofcom_MoU.pdf>.
954
282
Ibid, para 30.
283
Regulated industries: Guidance on concurrent application of competition law to regulated industries,
CMA 10, March 2014; Ofcom’s guidelines for the handling of competition complains and complaints con-
cerning regulatory rules, July 2012.
284
United Kingdom Competition Network (UKCN) Statement of Intent (December 2013).
285
TFEU, n 11, Art 256.
286
Regulation (EC) 1/2003 [2003] OJEU L1/1, Art 31. For a more detailed description of the power of the
General Court, see Bellamy & Child, n 6; Faull and Nikpay, n 198, paras 5.1125–5.1200; Kerse, C and Kahn, N,
EU Anti-t rust Procedure (6th edn, Oxford University Press, 2012).
287
TFEU, n 11, Art 256(3).
59
to Ofcom’s decision, could make a reference to the CJEU). As Ofcom and the CMA
exercise the same powers in competition cases, the appeals regime for each
of their decisions under the Competition Act is the same. There is an appeal on
the merits (both fact and law) to the CAT and, with leave, a further appeal to the
Court of Appeal (or Court of Session for Scottish cases)288 and ultimately to the UK
Supreme Court.
Since the CAT also hears appeals from Ofcom regulatory decisions under the
Communications Act,289 as well as appeals against CMA and other concurrent
regulators’ decisions under the Competition Act, the CAT acts as the formal point
of final decision of fact for both the Competition Act and Communications Act en-
forcement systems in the UK. The CAT sits as a panel of three—a judge or legally
qualified chairman and two others—a nd the Tribunal’s members include several
with experience in electronic communications markets.
The procedure followed by the CAT when hearing appeals is different from the
civil procedure rules used in general litigation. The CAT Rules in particular re-
quire more information to be provided at the beginning of an appeal against an
Ofcom decision than is usual in judicial review proceedings in the High Court.290
However, although the CAT rules differ from the general English civil procedure
rules, the CAT has produced a Guide to Proceedings which gives detail on how the
Rules will be applied in practice, which have the same force as a Practice Direction
made under the general civil procedure rules.291
Since the CAT has the power to rehear a competition case on the merits, its order
making powers also go beyond those of the Administrative Division of the High
Court in England.292 As well as the power to remit the matter to Ofcom (or the CMA
as the case may be), it also has the wide power under the Competition Act to make
any decision which the CMA (and therefore Ofcom) could have made in the same
case—that is, it may overturn the regulator’s decision and substitute its own.293
Even if the CAT agrees with the operative parts of the decision, it may nevertheless
quash any of the findings of fact in it.294
CAT decisions are enforceable in the same way as a judgment of the High Court
in England.
Note that the standard of review for appeals under the Communications Act 2003 has recently been
289
amended by the Digital Economy Act 2017. See further Chapter 3, at Section 3.3.7.4.
290
Competition Appeal Tribunal Rules 2015, SI 2015/1648, esp. rules 4 and 6.
291
Competition Appeal Tribunal Guide to Proceedings (2015) at <http://w ww.catribunal.org.uk/fi les/
Guide_to_proceedings_2015.pdf>.
292
Competition Act 1998, Sch 8, para 3(2). 293 Ibid, para 3(2)(e). 294 Ibid, para 3(4).
956
10.9 CONC LUSIONS
Part IV
11
11.1 INTRODUC TION
1
The authors would like to thank Cathal Flynn, associate at Bird & Bird, for his assistance with updating
this chapter; Claire Brunel-Cohen for her assistance with updating the previous edition; and Richard Graham
for his assistance with the original version of this chapter.
60
11.1.2 History
Capacity agreements have developed since the liberalization of the communications
market and the introduction of competition. Originally, when there was only one in-
cumbent operator in each market, that operator built and operated its own network
and determined how much capacity each element of the network required.
With the liberalization of the communications sector and the fostering of com-
petition, however, it soon became apparent that operators were not going to build
complete competing infrastructures. Some competing networks were established, in
particular in the major metropolitan areas, but substantial areas of many countries
continued to be served by one or perhaps two networks. In addition, other operators
entered the market without owning any infrastructure at all, and these ‘reseller’ op-
erators instead purchased capacity from other operators at wholesale prices, then
on-sold that capacity to other service providers or used it for particular retail services
or applications.
As new services were developed, so demand for capacity grew, encouraging
the network operators to design, build, and deploy new networks with ever
greater capacity. Fibre-optic systems began to replace the traditional copper
610
wire systems, not least because of the advantages they offered over the old sys-
tems, such as speed and capacity, digital signalling functionality, and reduced
signal degradation. Fibre-optic systems are basically thin filaments of glass
through which light beams are transmitted. The light beams carry informa-
tion in digital form, sent through the fibre strand over a pre-determined wave-
length. Because light can only travel in one direction, fibre-optic systems are
often (but not always) deployed in fibre pairs to enable simultaneous two-w ay
communication.
As the communications industry developed into a global industry and capacity
requirements increased, so operators became more willing to purchase capacity
from their competitors. This was particularly true of capacity on undersea cables,
which were originally constructed by consortia of incumbents, who quickly real-
ized the opportunities for profit from the hordes of new entrants into liberalizing
markets in the late 1980s and early 1990s.2
By the late 1990s, at the peak of the first telecommunications/e-commerce boom,
a number of operators began using capacity swaps as a means of boosting their
sales figures. The basic idea of a capacity swap, whereby Company A provides cap-
acity to Company B in exchange for similar amounts of capacity in a different geo-
graphic location, was sensible and logical for operators having different capacity
requirements in different places and building out their own networks at differing
times. However, a number of capacity swaps were also recorded internally as sales,
apparently boosting revenues and/or profits for companies whose internal auditing
processes were perhaps not as robust as they should have been. The bursting of
the dotcom bubble that followed shortly afterwards resulted in a number of these
companies entering Chapter 11 bankruptcy proceedings, although not necessarily
directly as a result of having swapped capacity with each other. Capacity swaps still
take place today but there are fewer of them and the appropriate accounting treat-
ment is now clear.
As technology evolved, so communications companies developed new methods
of increasing capacity on optical fibres, for example through wavelength division
multiplexing (WDM). WDM increases the capacity of an optical fibre by transmit-
ting multiple signals simultaneously over the same optical fibre but at different
wavelengths. Lit optical fibres traditionally supported one light stream, using
one frequency of light. First introduced in the 1980s, WDM technology was ori-
ginally limited to two wholly discrete frequencies on the same fibre. Since the
2
See further Chapter 16, at Section 16.2.2.
620
11.2.1 Leased lines
Wholesale leased lines, also known as private lines, dedicated lines, or permanent
circuits, are perhaps the simplest form of capacity agreement. A leased line is ba-
sically a fixed circuit linking two locations, rented from another communications
provider for a specific period of time. The circuit or line is usually provided at a
number of different speeds up to and over 10 Gbps.
While leased lines are an important business retail service, they are also a key
enabler of competition at the wholesale level. This applies in respect of both the
market for fixed line services and mobile services. In the case of the latter, leased
630
lines are essential to the ‘back-hauling’ of mobile traffic over long distances (eg
where traffic is routed between mobile network cells that are not adjacent to one
another). The early EU communications regulatory regime has therefore required
ex-incumbent operators to make available leased lines to those seeking network
and service access.3
Traditional leased lines using analogue or digital circuits and SDH (synchronous
digital hierarchy) and PDH (plesiosynchronous digital hierarchy) transmission
are now being superseded in many instances by alternative methods of transmis-
sion, notably the Ethernet protocol.
In addition, partial private circuits (PPCs), which provide dedicated transmis-
sion capacity between an end-user’s premises and an operator’s point of handover,
with the remainder of the circuit being provided by the operator from whom the
PPC is rented, offer purchasers a further alternative solution. The key to the attrac-
tion of PPCs, indeed the key to the attraction of all capacity agreements, is owner-
ship of the end-user. The operator that rents the PPC retains billing control and the
relationship with its own customer.
PPCs, as a segment of the leased lines market, offer an economic solution to
entities not having a fully national network, enabling operators to obtain exten-
sive national coverage with minimal investment costs and the ability to configure
the circuits to provide required capacity (and security) requirements.
A further sub-set of leased lines comprises International Private Leased Circuits
(IPLCs or International Private Line Circuits in the USA). An IPLC is a point-to-
point circuit line offered by operators to end-users to provide a communications
link (eg in the form of a virtual private network or VPN) between two (or more)
offices of the same organization based in different parts of the world. The ending
of the BT/Mercury international facilities duopoly by Oftel in 1996, which en-
abled new entrants to provide International Facilities Based Services, including
facilitating the use of IPLCs to do so, played a key part in the development of inter-
national communications as other countries followed the UK’s lead.
Unlike a traditional leased line, however, an IPLC is not a single dedicated cir-
cuit. Rather, it is a service offering termination services between two (or more)
pre-defined points, although the operator may route traffic through a variety of
different networks between those points. This has the benefit of a lower cost, but
has the potential problems of different traffic engineering processes or unexpected
traffic volume surges impacting on transmission speeds.
3
For background on this, see Chapter 4, at Section 4.6. Leased lines are included in Market 4 in the Annex
to the European Commission’s Recommendation on Relevant Markets. This market is called the Market for
Wholesale high-quality access provided at a fixed location. In the UK, Ofcom refers to this as the market for
‘business connectivity services’. See also Chapter 8.
604
11.2.2 IRUs
Historically, indefeasible rights of use (IRUs) developed as a term of art to describe
certain long-term leases of part of the capacity of an international submarine cable.
The capacity is specified in numbers of channels of a given bandwidth. Many of the
original IRU arrangements were for very long terms (25 years was typical), which
led to a number of significant implications (which will be discussed further).
As noted earlier in this chapter, the original undersea cables were constructed
by consortia of incumbents. These have now largely been replaced by individual
companies or joint ventures, although it is not uncommon for new consortia to
be formed to construct systems. The membership of the consortia has changed,
however. Global liberalization, enabling operators to hold licences or be author-
ized to operate in multiple jurisdictions, has removed the original need for foreign
partners, and many operators have experienced the difficulties of managing by
committee, as a consortia requires.
The first ‘undersea’ cable was actually laid under the river Thames in 1840, and
by 1850 a cable connected England to France. An undersea telegram cable linked
Ireland and Newfoundland in 1858. The first transatlantic telephone cable system,
TAT-1, was constructed in 1956 and the first transatlantic fibre-optic system came
into service in 1988. The oceans of the world are now lined with submarine cables.
The success of subsea and long-d istance cables also drove the development
of bandwidth technology, in particular the invention of bi-d irectional analogue
technologies in the early 1960s, superseding the requirement for a separate cable
for each ‘direction’ of traffic.
In order to ensure resilience, IRUs can be purchased in pairs, on two separate
cables (operators may have other contingency arrangements in place—for ex-
ample a satellite link). The intention is that if one cable is accidentally cut, for ex-
ample by a ship’s anchor in shallow water, the built-i n redundancy of the design
will permit uninterrupted service because all traffic can be automatically routed
via the surviving cable. Usually, this works well: simultaneous cuts on two sep-
arate subsea cables are a relatively rare—i f still alarming—event. As a general rule,
once a subsea cable is laid, the only risks it faces are if it is in an earthquake zone
or a heavy fishing activity area.
IRUs can be purchased in pre-existing systems. More commonly, however, they
are purchased in systems that are being built or that have just been built. This ac-
knowledges the fact that technology continues to develop at a very fast rate, and
even relatively new subsea cables are quickly superseded by faster and cheaper
designs.
Early purchasers of IRUs in a planned subsea cable system are often granted
capacity bonuses on the basis that their committed participation makes initial
650
funding by the cable owner far simpler and may also encourage other operators
to become involved.
Unlike a traditional leased line, whereby the renting operator pays fixed peri-
odic charges (usually either monthly or quarterly), an operator purchasing an IRU
historically paid a substantial upfront sum on execution of the agreement, but
then is only liable for ongoing maintenance costs. The single upfront payment has
advantages for both the purchaser and the seller. The latter is able to book revenue
earlier than it would be able to for a normal leased line. The former can take ad-
vantage of specific capital allowances for IRUs, enabling them to obtain tax deduc-
tions for depreciation. This was not an issue when those providing the IRUs were
incumbent or ex-incumbent operators and credit risk was not even perceived as
an issue. However, after the significant expansion of communications operators
and the bursting of the ‘bubble’ in early 2001 many of these arrangements looked
far from sound with hindsight. Fundamentally, IRUs are usually, from a legal per-
spective, contractual rights. Having paid significant sums up front for a contrac-
tual right is not necessarily an argument that takes you very far in discussions with
an insolvency practitioner trying to maximize value from the assets of a failed op-
erator. A lot of time and effort was spent in attempts to characterize IRUs as prop-
erty rights, but to little or no avail.
The term ‘indefeasible’ is something of a misnomer, given that IRUs are not in-
defeasible and no legal title passes to the purchaser. Rather, an IRU is a long-term
‘lease’ that cannot be terminated by the cable owner (or other superior rights
holder) other than in particular specified and limited circumstances, for example
the insolvency of the purchaser, or the purchaser’s failure to contribute to main-
tenance costs of the cable.
It is interesting to note that the term ‘IRU’ now has a much wider usage than the
original IRU arrangements on undersea cables discussed above. Many capacity
arrangements are characterized as IRUs, including arrangements that might be
considered to be leased lines (as noted) or dark fibre arrangements (see Section
11.2.3). It is important to look below the label and understand the reality of the ar-
rangements that are being put in place and the associated risks and how these are
most appropriately addressed.
Act 2000 but this has now been rewritten into Parts 2 and 5 of the Income Tax
(Trading and Other Income) Act 2005 (‘ITTOIA’) with effect from 5 April 2005.
In ITTOIA, IRUs are defined as ‘indefeasible rights to use a telecommunications
cable system’ (or a right derived from such rights).4 Any IRU relating to a cable or
to a system a part of which comprises a cable will therefore be caught by ITTOIA.
IRUs in any medium other than a cable (eg a duct) should fall outside the re-
lief. For corporation tax purposes, the rules on the taxation of intangible fixed
assets 5 apply to IRUs with effect from April 2002 (subject to transitional rules),
so that profits and gains in respect of IRUs acquired by a corporate purchaser
are generally chargeable to corporation tax as income in accordance with their
accounting treatment. Under Part 8 CTA 2009, a deduction would be afforded
to a company in respect of expenditure on the IRU on an amortized basis, ie
over the life of the IRU. In either case, any income or proceeds derived from the
purchaser’s subsequent exploitation, disposal, or revaluation of the IRU will be
taxable income. If the purchaser of an IRU does not have a trading activity, it will
be subject to tax on income derived from the IRU but, in calculating the amount
of net income which is chargeable, it would still be able to make a deduction for
the acquisition costs of the IRU. In the case of a purchaser of an IRU, within the
charge to income tax (or corporation tax for pre-April 2002 assets), deductibility
of the revenue expense is subject to the usual conditions for trading deductions,
in particular, expenses must be incurred wholly and exclusively for the purposes
of the trade. If such purchaser is not a trader, equivalent conditions apply.6 From
the perspective of the owner of a cable system (other than a corporate owner
within the charge to corporation tax, which will be subject to tax on any profit
or loss, generally in accordance with its accounts under Part 8 CTA 2009), the
grant of an IRU by the cable-owner is not affected by the tax treatment of IRUs for
purchasers. For UK tax purposes, the grant of an IRU by a cable-owner might be
considered as either giving rise to (i) a trading receipt where the grant is part of
the owner’s normal exploitation of the cable system, or (ii) a capital receipt de-
rived from the use or exploitation of its cable system. Where the receipt is a cap-
ital sum, the cable-owner will not be regarded as having disposed of the cable
network itself, and neither will it be treated as having disposed of absolute own-
ership of the rights to use the cables. Instead, the cable-owner will only be taxed
on the capital sum (less any allowable costs) on the basis that it has received
consideration for granting a right of use of part of the cable system to the person
4
ITTOIA, s 146.
5
Corporation Tax Act 2009 (CTA 2009), Pt 8 (previously Schedule 29 to the Finance Act 2002).
6
The authors would like to thank Mathew Oliver, Partner Bird & Bird, for his assistance with updating this
chapter.
670
acquiring the IRU. Similarly, for UK capital allowances purposes, since the grant
of an IRU is not treated as a part ‘disposal’ of the cable system itself, the owner
continues to be entitled to capital allowances on its original cost of installing the
system.
11.2.3 Dark fibre
As a general rule, one of (if not) the largest expenses faced by any communi-
cations network provider is installing that operator’s cables in the ground. It
is commonly estimated that up to 80 per cent of the entire cost of a fibre optic
network can be spent on the civil engineering work necessary to design, con-
struct, and connect it.7 While certain regulatory initiatives had been taken to
reduce these costs by, for example, promoting the more efficient use of existing
infrastructure and mandating greater cooperation and information sharing
between infrastructure owners conducting civil works,8 the costs associated
with civil engineering as a proportion of the overall network deployment cost
remains significant.
The actual infrastructure itself is relatively cheap. As such, it makes obvious
commercial sense for operators to install more fibre than is actually currently re-
quired at the time of build in its ducts and channels, in order to provide for future
network development, resilience, and on-sale opportunities of leased lines and
dark fibre. (This of course is part of the reason for the bandwidth surplus noted
above.)
The term dark fibre has been defined in a variety of different ways over the years.
Some have argued that it is any optical fibre that is not attached to transmission
equipment at all. Others have suggested it is fibre that is not attached to trans-
mission at only one end, so is awaiting additional work before it can be utilized.
Certainly, there appears to be common consensus that it is optical fibre through
which no light and thus no signal is being transmitted. Fibre through which light
is being transmitted, and which is carrying a signal, is known as lit fibre. There
is also something of a halfway house known as ‘dim fibre’, which is a term used
in DWDM. DWDM supports as many as 160 wavelengths, as mentioned, each of
which is a different frequency of light. Thus, when some wavelengths are left ‘dark’
and some are ‘lit’, the resulting fibre is ‘dim fibre’.
7
Proposal for a Regulation of the European Parliament and of the Council on measures to reduce the
cost of deploying high-speed electronic communications networks, COM(2013) 147 final—2013/0 080 (COD),
Explanatory Memorandum, at 2.
8
See eg Directive 2014/61/EC of the European Parliament and of the Council of 15 May 2014 on measures
to reduce the cost of deploying high-speed electronic communications networks, OJ L 155/1, 23 May 2014.
608
9
See Section 16.2.1.
609
typically has a limited mobile network or no mobile network itself and has not
been assigned spectrum resources.10 Different types of MVNO models exist. These
range from a pure ‘reseller model’, where the MVNO owns/operates limited or no
wireless network of its own and simply resells to its end-users the service provided
by the ‘host’ MNO, to ‘full MVNOs’ using their own network infrastructure and
capability to handle much of the transmission and back office functionality. This
can include, for example, ownership and activation of SIM cards and numbering,
billing management, account authentication, ownership and management of the
subscriber database, and control of customer service. Each type of MVNO model
will, for obvious reasons, demand a specific type of capacity arrangement with the
capacity provider or host MNO.
The classic MVNO model is a business with a well-k nown customer brand, such
as Virgin or Tesco, leveraging its brand into mobile communications. A number
of MVNOs have also positioned themselves to appeal to specific target customer
groups, for example cost-sensitive customers who do not require any value-added
service elements. Some have used the MVNO model as a way of expanding their
product portfolio from a pure fixed telephony offering to also include mobile tel-
ephony.11 Others have adopted innovative business models by relying on adver-
tising revenues.
There are various attractions for commercial entities wishing to set up such ven-
tures. The most obvious attraction is that MVNOs will not be required to invest in
infrastructure assets (including spectrum) to enter the mobile service market. This
lowers barriers to entry by reducing the capital expenditure required to enter the
mobile marketplace and provides a potentially shorter route to profitability. The
lowering of barriers to entry in this way promotes greater competition for mobile
services, which is why MVNO access has been used in some countries as a regu-
latory remedy on national markets for mobile access and call origination services
where a competitive failure has been identified. The downside of MVNO access,
however, is that margins can be significantly more exposed when retail prices fall.
Another possible disadvantage is that MVNO access, if competitively priced, could
discourage new entrants from investing in mobile network infrastructure which
can have a long-term negative impact on (infrastructure-based) competition.
There are also benefits for the mobile network operator providing the services to
the MVNO in terms of gaining additional usage of capacity on its network, thereby
generating revenues that it would not otherwise generate. In this way, MVNOs pro-
vide the mobile network operator with an opportunity to increase economies of
10
Ofcom, ‘The International Communications Market Report 2016’, Glossary, 205 <https://w ww.ofcom.org.
uk/__data/a ssets/pdf_fi le/0 026/95642/ICMR-Full.pdf>.
11
See Ofcom, ‘Strategic Review of Digital Communications’ Discussion Document, 16 July 2015, 74.
610
scale in respect of its network utilization. They also provide MNOs with an added
revenue stream. MVNOs may allow the mobile network operator to acquire traffic
from market niches that would not otherwise be as effectively reached under its
own branding strategy.
However, there will clearly be a potential risk of loss of customers to the MVNO.
There has been a lot of debate around the role that regulation needs to play in
facilitating MVNOs. While MVNO access was initially applied as a regulatory
remedy in some national markets in Europe where collective significant market
power (SMP) was identified, a lighter regulatory approach has been followed in
recent years as competition has developed in mobile service markets. In the UK,
MVNO arrangements are not facilitated by regulation requiring mobile network
operators to enter into an MVNO arrangement, but are left as a commercial matter
to be negotiated between the mobile network operator and the potential MVNO.
Further, an MVNO is not required to comply with any specific MVNO regulations
in addition to the requirements set out in the General Conditions. The approach
has been replicated in most other European jurisdictions.
There are now a number of successful MVNO arrangements in place in the
United Kingdom, with Virgin Mobile’s telecommunications supply agreement
with BT (which acquired Everything Everywhere in January 2016) being the most
notable, with three million UK subscribers.12 The number of MVNOs has increased
over the last decade and, based on the most recent reliable information, there are
around 1,000 such arrangements in place across the globe.13 With service pro-
viders increasingly offering bundled products to their customer base, the addition
of mobile products to their portfolio will increase the attractiveness of MVNOs.
The backbone of any MVNO arrangement is the ability of the MVNO to resell
under its own branded service the airtime it has purchased from the mobile net-
work operator. This ability to purchase wholesale extends beyond the basic pur-
chase of minutes to SMS, MMS, GRPS and 3G/HSDPA capacity as well.
To conclude, for mobile network operators, there are a number of reasons for
allowing MVNOs to use their networks. MVNOs offer an ability to lease excess
capacity on the operator’s purchased spectrum, thereby facilitating a greater re-
turn on sunk network investment costs, with the responsibility for marketing and
customer support often being dealt with by the MVNO. In this way, MVNOs pro-
vide the mobile network operator with an opportunity to generate increased econ-
omies of scale in respect of its network utilization. They also provide MNOs with
an added revenue stream. MVNOs may also provide the mobile network operator
12
<http://w ww.virginmedia.com/c orporate/media-c entre/press-r eleases/v irgin-media-a nd-bt-a gree-
new-mobile-deal.html>.
13
<https://w ww.gsmaintelligence.com/research/? file=e83c8a3cb0f8a685fb8ab53fff206518&download>.
61
with the ability to reach market niches that would not otherwise be reached under
its own branding strategy.
11.3.1 Introduction
Although some types of capacity agreement, such as leased lines and IRUs, have
in effect become commoditized, thus leaving little scope for negotiation of indi-
vidual terms, there remain a number of key issues to bear in mind in every cap-
acity agreement.
As with any written contract, certain provisions will usually be included in a
capacity agreement, in particular payment terms, limitations on and exclusions
of liability, termination, confidentiality, and traditional boilerplate provisions. In
addition, the following provisions will need particular attention in any bespoke
capacity agreement.
understand the balance between the margin a supplier may be making on pro-
viding the services and the potential impact on the customer if the services are not
provided at all or suffer a quality degradation. Clearly resilience and diversity have
roles to play here in appropriate circumstances. Many SLAs will be backed up by
service credit regimes which will provide some limited redress if the contracted
availability is not achieved without the customer having to claim and prove its loss.
It is widely acknowledged that leased lines play a vital role in business com-
munications in the UK and other markets, both at retail and wholesale levels. It
is therefore important that the markets for these services operate effectively and
competitively. As such, it is not surprising that Ofcom and other regulatory author-
ities have focused regulatory attention on these markets.
Where an operator has had an SMP determination made against it,14 it may be
subject to minimum service level requirements. This may mean that SLAs offered
in a commercial context can be subject to the prior approval of the regulatory au-
thority, typically in the form of an approved reference offer. Adherence to these
SLAs is seen by regulators as critical to ensuring a functioning competitive en-
vironment, and regulatory authorities can apply various regulatory tools to en-
sure that access providers do not discriminate between their competitors and,
for example, their downstream arm in terms of quality of service levels provided.
This can even include the operational separation of the access provider’s network
management from its downstream retail arm and, in extreme cases, their outright
structural separation. In the UK, BT and Ofcom agreed in March 2017 to the im-
plementation of stronger operational separation measures within the BT Group in
the context of Ofcom’s strategic sector review.15 This includes the incorporation of
Openreach, BT’s network management division, as a separate legal entity and the
implementation of various other safeguards aimed at ensuring equivalency both
in terms of network access and ongoing service levels.
Dark fibre raises interesting SLA issues in that the apparatus attached to each
end of the fibre is the user’s responsibility, but the connection may be the fibre
owner’s (or a third party’s) responsibility. It is important that this is clearly ad-
dressed in any dark fibre agreement to ensure clarity of obligations.
Further, there are a number of different product specifications for dark fibre at
an International Telecommunication Union (ITU) level, each with different prop-
erties, depending on the type of fibre, its length, and its purpose.16 The capacity of
14
See further consideration of SMP obligations in Chapter 4, at Section 4.6 and Chapter 8.
15
<https:// w ww.ofcom.org.uk/ a bout- o fcom/ l atest/ m edia/ m edia- releases/ 2 017/ b t- a grees- t o- l egal-
separation-of-openreach>; <http://www.btplc.com/UKDigitalFuture/Agreed/index.htm>; <https://www.ofcom.
org.uk/__data/assets/pdf_file/0035/98855/Openreach-consultation-2017.pdf>.
16
eg ITU-T G.652 (11/16) remains perhaps the most commonly deployed fibre.
613
11.3.3 Term
The ideal term of an agreement will vary depending on a number of factors. In rela-
tion to capacity agreements, particular factors to bear in mind include:
Service Continuity: a longer term contract provides more security than a shorter
term, in that ongoing costs are known and the purchaser will have the comfort
of knowing that it will not have to find replacement capacity on short notice;
Cost: a longer term can usually be purchased more cheaply (relatively speaking)
than a short term. If the capacity being purchased is a fundamental part of the
purchaser’s service offering, the purchaser may want a longer term contract to
ensure continued supply;
Scalability: if an operator is expecting to expand quickly, it will want an option to
increase capacity quickly and simply. In which case, it may be sensible to seek
to tie all related capacity agreements to the same termination/expiry date for
ease of internal administration;
Risk Aversion: short term contracts lower the barriers to exit in the event of a market
downturn. During the dotcom boom, operators entered into very long-term
contracts, but then struggled to exit those contracts when the bubble burst. This
is the counterpoint to the issue of contractual security noted above;
Competition: in a very competitive market, prices can be driven down as innov-
ation and technology develops. Short term contracts enable purchasers to take
advantage of new service offerings such as enhanced speed or lower latency.
Notwithstanding this, purchasers may try to ‘future proof’ a capacity agree-
ment by establishing a periodic price review mechanism. The effectiveness
of such a mechanism will ultimately depend on the bargaining/purchasing
power of the purchaser, particularly if this price review mechanism requires
the vendor to approve either the instigation of the periodic review exercise itself
or the revised price (or both).
11.3.4 Access
As in the case of simple leased lines, the majority of dark fibre agreements are
now commoditized standard terms. Key issues to bear in mind, however, remain
614
the same as when early dark fibre agreements were negotiated. Given that the
purchasing operator is responsible for attaching electronic communications
equipment and lasers to light the fibre at each end (unless the vendor is providing
a managed service), the purchaser will want to ensure that it has appropriate and
sufficient access (and, depending on the type of arrangement, location or co-
location) rights at each end. This will require the detailed description of network
demarcation points in the agreement. The seller will usually want as long a term
as possible, but with the option of early termination should it decide to alter the
physical architecture of its network.
11.3.6 Testing
Capacity agreements will often include provision for testing prior to handover. In
IRU agreements, there may or may not be testing procedures for the entire cable
system. An agreement should certainly contain testing and acceptance provisions
for the IRU circuits themselves.
arrangements tend to focus particular attention on the laws that the satellite op-
erator and customer must comply with. The satellite operator will generally be re-
quired to ensure that all approvals, consents, and licences are obtained in respect
of the leasing of the satellite capacity. On the other hand, the customer will often
be burdened with the requirement to ensure that it complies with all applicable
laws in force in the country of transmission of the up-l ink signal and, more oner-
ously, in the countries where the footprint from the satellite is able to be received
(usually referred to as ‘landing rights’). The Audiovisual Media Services Directive
(previously the Television Without Frontiers Directive) can be highly relevant
here to lightening some of the regulatory burdens in relation to TV broadcasts
and on-demand services, but any detailed discussion is beyond the scope of this
chapter.17
Force majeure and exclusions of liability: The relationship between force majeure
and unavailability is particularly important in transponder leasing agreements.
With the risks of adverse atmospheric conditions, such as solar flares and sun out-
ages, and also issues such as micrometeoroid impacts and unavoidable conjunc-
tion with space debris, being highly relevant, the satellite operator will be keen
to ensure that the definition of force majeure and exclusions of liability provide
appropriate protection for the satellite operator. It is important that the customer
understands the implication of these and what arrangements are available to miti-
gate the consequences.
Rights of pre-emption and restorability: As the ultimate aim of any satellite oper-
ator is to ensure that all its capacity is used, there will be scenarios where capacity
cannot be offered to those requesting it, whether due to excess capacity requirements
on the allotted transponder or by the downtime of alternative transponders. The sat-
ellite operator therefore has to prioritize allocation of capacity and, where designated
capacity is not available, offer alternative capacity. The relevant terms used to de-
scribe the prioritization are whether there is a right of pre-emption over the capacity
and whether the capacity is restorable:
alternative transponder on the same satellite, and if the satellite fails, then the
customer will be provided with capacity on an alternative satellite to the extent
reasonably practicable; and
• where the satellite capacity is not restorable, then in the event the transponder
or satellite fails, the customer will not be provided with alternative capacity.
The customer’s needs and what it is willing to pay, together with the strength of
the customer’s negotiating position, will ultimately dictate what priority it has
over other customers. The following scenarios illustrate the hierarchy of customer
rights:
It is very important to look carefully at how these terms are used in different trans-
ponder leases, including whether pre-emption rights are in favour of the satellite
operator or the customer.
Earth stations: It will be usual for satellite operators to require its customers to
ensure that the earth stations used for transmitting the signals to the satellites are
licensed and meet certain requirements. The customer will generally take the risks
arising from non-compliance with these requirements.
Right to re-sell: The extent to which the customer under a transponder leasing
arrangement may resell any allotted capacity, if at all, is clearly a key issue and
should be stated.
Fixed/occasional-use capacity: Historically, satellite capacity tended to be leased
on a permanent basis. However, the requirement for the occasional-use of satellite
capacity has become increasingly important in the satellite capacity sector. In re-
spect of broadcasting, the transmission of live broadcasts away from the studio
(eg to Anfield or the Millennium Stadium) has contributed to an increase in the
number of occasional-use satellite capacity arrangements.
Rights of cancellation: The customer’s right to cancel any allotted capacity will
ultimately depend on the negotiated position. It is unusual for a customer to be
able to terminate allotted capacity once it has entered into a long-term supply ar-
rangement without paying early termination charges. In respect of occasional-use
capacity, there are general rights to withdraw capacity on agreed notice provided
that a minimum commitment is achieved over the term of the agreement.
Pricing: The MVNO will be able to work independently of the host mobile network
operator and can set its own pricing structures, though clearly it will need to pay
the charges it has agreed to pay to the mobile network operator. If the MVNO
access is mandated under the national telecommunications regime, then the
host operator may be subject to price control obligations, including those that
require the recovery of incurred costs only. If this is not the case, the pricing
practices of the host MNO may need to be compliant with the principles of ex
post competition law, including those relating to discriminatory pricing for
example.
Branding rights: This is a key issue given that a customer may be leveraging a well-
known consumer brand that is a fundamental element of its business plan.
Many Tesco Mobile subscribers may not know that the underlying network
used is the O2 network.
Term/exclusivity/minimum commitment: As one would expect, these issues are at
the heart of the commercial deal. The MVNO will need to ensure that it has the
ability to move to a different mobile operator or at least use this possibility as
leverage when terms are renegotiated and extended.
Forecasting: MVNO agreements can include traffic forecasting provisions re-
quiring the MVNO to furnish the host MNO with reasonably accurate traffic
forecasts. Traffic forecasting allows the MNO to take the measures necessary
to ensure the availability of sufficient capacity for network functionality in the
future. The MVNO agreement may also require accurate traffic forecasting, and
the MVNO may be subject to financial penalty if its forecast is out by more than
a certain fixed percentage of the actual traffic handled.
Promoted services: Contractual provisions will apply in respect of the services that
the MVNO is able to provide and promote. In the field of mobile communica-
tions, new services are regularly developed as a way of creating extra revenues.
MVNOs will require flexibility to procure and offer these services to their own
customers. However, the mobile network operator will be keen to ensure that
the MVNO does not profit from these opportunities to the detriment of the mo-
bile network operator’s customer base.
Customer services and billing: The MVNO agreement will document who will be
responsible for customer services and billing. If the MVNO has such responsi-
bility, the contractual arrangements will need to document how information is
passed between the systems and how customer service issues are ticketed and
resolved.
618
Today’s electronic communications market is a long way from the market of the
1990s, when specific individual licences were required by operators offering
services either in the UK or in undersea capacity internationally. In jurisdictions
outside the UK, of course, different regulatory regimes will apply, although there
is now significant harmonization throughout the EU Member States as a result of
the EU communications regulatory regimes. The focus of this chapter, however, is
on the UK capacity market.
As discussed further in Chapter 6, under the Communications Act 2003, General
Conditions apply to anyone who provides an electronic communications service18
or an electronic communications network19 in the UK. A capacity agreement, de-
pending on the type of capacity being purchased and the other provisions of the
contract, may be a contract for provision of an electronic communications service
or an electronic communications network (or both). In either case, both the buyer
and the seller of the capacity will have to ensure that they comply with the relevant
General Conditions that apply to them, and these will vary on a case-by-case basis,
because differing conditions apply to operators, depending on what services or
networks they are providing and whom they are providing to.
Many capacity agreements are also access services and, as such, may be regu-
lated under the regime established by the Access and Interconnection Directive.20
Access agreements are addressed further in Chapter 8 but in brief, if an operator
has been found by Ofcom, or an equivalent regulatory authority in another EU
Member State, to have significant market power21 in a relevant market, that regula-
tory authority may impose conditions on those operators, in particular in relation
to ensuring that charges are on cost-orientated terms and are not discriminatory.
18
Defined in the Communications Act 2003 (as amended), s 32(1).
19
Defined in the Communications Act 2003 (as amended), s 32(2).
20
Directive 2002/19/EC on access to, and interconnection of, electronic communications networks and
associated facilities, OJ L 108/7, 24 April 2002 as amended in 2009.
21
Addressed in the Communications Act 2003, s 78.
619
It should be noted that regulated service markets are subject to periodic review.
The outcome may lead to either of the following: (i) a revised market definition
may mean that certain services formerly subject to inclusion (and regulation)
within a specific market are no longer so subject or vice versa; and (ii) the regu-
latory obligations that are imposed in respect of a regulated market during one
review period can be amended or withdrawn for the next review period and/or
new regulatory obligations can be imposed. These periodic changes can have an
important impact on commercial arrangements that are concluded in respect
of regulated services. If, for example, charge controls are removed as an ex ante
market remedy, the access provider will likely have some more commercial flexi-
bility in terms of the prices it offers for its services (although it may be subject to
other ex ante regulatory measures, including a margin squeeze test, and the ap-
plication of ex post competition law). The example of the business connectivity
market in the UK provides a good example of how ex ante regulation can evolve
over time.
In its market review of 2004, Ofcom found that Kingston Communications had
SMP in the Hull area and BT had SMP in the rest of the UK for the retail market
for low bandwidth ‘traditional interface’ (TI) markets (such as analogue circuits
or digital circuits using SDH and PDH transmission), the wholesale markets for
low and high bandwidth TI terminating segments, and the ‘alternative interface’
(AI) market (eg Ethernet) for terminating segments at all bandwidths. BT was also
found to have SMP in the UK market for trunk segments.
As a result of these findings, a number of obligations were imposed on BT and
Kingston, including obligations to supply, requirements not to discriminate un-
duly between customers, requirements to publish prices, terms and conditions,
and in some cases, price controls.
In a further market review in 2008,22 Ofcom found that progress towards more
effective competition had varied considerably by market. It proposed that:
• outside the Hull area, a separate market now existed for wholesale AI services at
bandwidths over 1Gbps and that this market was effectively competitive, so no
SMP regulation was required;
• a new geographical market in high bandwidth wholesale terminating segments
existed in Central and East London and that again it was competitive and no
SMP regulation was required; and
• the market for low bandwidth TI retail leased lines in Hull was now competitive
and so could be deregulated.
22
Ofcom, ‘Business connectivity market review’, 8 December 2008 <http://w ww.ofcom.org.uk/consult-
ations/bcmr08/>.
602
In 2013, Ofcom identified a set of markets at retail and wholesale level for con-
tinuing ex ante regulation.23 In summary, these markets were the following: (i) the
retail market for very low bandwidth leased lines in the UK, excluding the Hull
area; (ii) the wholesale market for TI regional trunk segments in the UK; (iii) the
retail market for low bandwidth TI leased lines in the Hull area; and (iv) the re-
tail market for low bandwidth alternative interface (AI) leased lines in the Hull
area. As a result of these findings, a number of obligations were imposed on BT and
Kingston, including obligations to supply (grant access), requirements not to dis-
criminate unduly between customers, requirements to publish prices, terms and
conditions, and in some cases, price controls.
In 2016, Ofcom once again identified a set of retail and wholesale business
connectivity markets for continued ex ante regulation.24 This included both
terminating and trunk segments of leased lines. On the basis of the technologies
used, Ofcom differentiated between contemporary interfaces (or CI forming part
of a single product market for contemporary interface symmetric broadband ori-
gination (or CISBO) services) and TI. TI leased lines include services which use
legacy analogue and Time Division Multiplexing (TDM)-based interfaces. CI
leased lines, on the other hand, include services which use Ethernet and WDM.
Specifically with regard to the CISBO product market, Ofcom defined a number of
geographic markets including: the Central London Area, the London Periphery,
the Rest of the UK, and Hull.
Ofcom’s 2016 market review found that the competitive landscape had changed
since 2013 sufficiently to justify the deregulation of a number of business connect-
ivity markets. Ofcom found BT to have SMP in the CISBO market for the London
Periphery and the Rest of the UK, and introduced a new passive ex ante remedy on
BT to supply access to dark fibre. According to Ofcom, the objective of this remedy
was to allow alternative providers to assemble a wider range of inputs in order
to differentiate their leased lines services offer and to compete more effectively
with BT.
BT appealed Ofcom’s 2016 market review decisions concerning the market
definition and the imposition of the dark fibre remedy. The Competition Appeal
Tribunal (CAT) ruled that Ofcom made some specific errors in relation to the
market definition.25 On 20 November 2017, the CAT ordered Ofcom to revoke its
23
Ofcom, ‘Business connectivity market review’, 28 March 2013.
24
Ofcom, ‘Business connectivity market review’, 28 April 2016 <https://w ww.ofcom.org.uk/consultations-
and-statements/c ategory-1/business-connectivity-market-review-2015>.
25
British Telecommunications v Office of Communications (BCMR), 1260/3/3/16 <http://w ww.catribunal.
org.uk/2 37-9285/1260-3-3-16-British-Telecommunications-.html>.
612
SMP findings and regulatory conditions (including the dark fibre remedy) for all
CISBO services.
On 23 November 2017, Ofcom removed all regulation in the CISBO markets,
including the dark fibre remedy.26 It then made temporary market identifica-
tions and market power determinations, and imposed temporary SMP conditions
and directions on BT.27 Ofcom consulted on whether it would be appropriate to
add a restricted form of dark fibre access to this package of temporary measures.
On 12 April 2018, Ofcom published a Statement setting out its decision not to intro-
duce this temporary remedy.28
As noted above there has been a lot of debate around the part regulation needs
to play in facilitating MVNOs. However, the light regulatory touch currently ap-
plied seems to have been the right approach at least so far. In the UK, an MVNO
is not required to comply with any specific MVNO regulations in addition to the
general requirements set out in the General Conditions. As noted earlier, however,
some countries have chosen in the past to impose regulatory requirements on
MNOs aimed at facilitating the access MVNOs require.
There are significant regulatory issues and considerations in relation to the
provision of satellite services, including under the ITU Radio Regulations. 29
These are mainly an issue for the satellite operator but if a customer under a
transponder lease and associated arrangements is providing its own uplink ser-
vice it will need to comply with the relevant requirements set out in the General
Conditions and Wireless Telegraphy Act 2006 and regulations issued under it.
Consideration will also need to be given to any regulation relevant to the landing
of the signal in relevant jurisdictions and associated licensing requirements for
earth stations.
In addition, of course, operators are also bound by the rules of general competi-
tion law, in particular Articles 101 and 102 of the Treaty on the Functioning of the
European Union.30
26
Ofcom, ‘Business Connectivity Market Review 2016’, ‘Revocation of certain measures imposed in the
business connectivity markets’, 23 November 2017 <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 018/
108018/BCMR-Revocation-Notification.pdf>.
27
Ofcom, Business Connectivity Markets, ‘Temporary SMP conditions in relation to business connectivity
services’, 23 November 2017 (BCMR Temporary Conditions Statement) <https://w ww.ofcom.org.uk/_ _data/
assets/pdf_fi le/0 019/108019/BCMR-Temporary-Conditions.pdf>.
28
Ofcom, ‘Statement on adding dark fibre to the temporary remedies for business connectivity markets’, 12
April 2018 <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 015/112902/dark-fibre-statement.pdf>.
29
See further Chapter 16, at Section 16.3.2.
30
See further Chapter 10.
62
under 1 millisecond, and (iii) improved energy efficiency, although it remains un-
clear at the time of writing how exactly this technology will be deployed on the
ground.31
Operators today widely acknowledge the cost and difficulties inherent in
constructing competing infrastructures. This is particularly true in the mobile
market, where 3G and 4G networks (as a result of the higher frequency range
and the requirement to transmit ever growing amounts of information) require
smaller cells, more base stations and additional macrocells, microcells, and
picocells than the 2G network. This trend will only continue with the develop-
ment of 5G technology. Heavier network deployment requirements give rise to a
corresponding increase in build-out and maintenance costs. As a result, mobile
network operators have been entering into infrastructure sharing agreements
for a number of years. These agreements, which may take a variety of forms,
from creating a joint venture vehicle to build, run, and maintain the infrastruc-
ture, to the simple sharing of space, masts or buildings, allow network operators
to make vital savings in terms of capital and operational expenditure in an in-
creasingly competitive environment. Care has been taken, however, to ensure
that such arrangements do not have an anti-c ompetitive effect, particularly in
urban areas where the deployment of network infrastructure is less costly than
in rural areas. It is interesting to note that the existence of such arrangements
has also been an important factor in the analysis of mobile market consolida-
tion, including in the UK.
Lastly, it is clear that one of the major themes emerging in recent years has
been the continued drive at EU level for further deregulation across Europe. The
number of markets identified as susceptible to ex ante regulation in the European
Commission’s Recommendation on Relevant Markets is currently four,32 although
Member States can and do regulate more markets where market failure (SMP)
is identified. The slow but steady roll-back of ex ante regulation in this manner
means that there are fewer markets where capacity agreements are concluded
under regulated terms and conditions, and correspondingly more markets where
operators have the commercial freedom (subject to ex post competition law re-
quirements) to negotiate terms and conditions.
31
The authors would like to thank Barry Jennings, Legal Director Bird & Bird, for his assistance with up-
dating this section of the chapter.
32
Commission Recommendation of 9 October 2014 on relevant product and service markets within the
electronic communications sector susceptible to ex ante regulation in accordance with Directive 2002/21/EC
of the European Parliament and of the Council on a common regulatory framework for electronic communi-
cations networks and services, (2014/710/E U), Annex, OJ L 295/79, 11 October 2014.
642
652
12
From the point of view of European electronic communications law, most MNEs
and the entities comprising them would legally qualify as end-users.1 This is so be-
cause they are purchasing telecommunications services predominantly for their
own use and not for resale. However, several aspects of their service requirements
make them different from other end-users.
1
According to Article 2(n) of the EU Framework Directive, ‘end-u ser’ means a user not providing public
communications networks or publicly available electronic communications services.
672
operators are still likely to be in better position than other MNEs when it comes to
leasing capacity and purchasing other services internationally. This may be due to
their expertise, economies of scale and scope, or licensing requirements that may
in some countries prohibit corporate self-provision of services.2
12.1.2 Outsourcing
The above elements may put international operators in a unique market position to-
wards MNEs. A natural additional step for firms such as BT Global Services wanting
to offer MNEs a one-stop global communications solution would therefore be the
bundling of connectivity with the associated terminal equipment and IT together
with the necessary support and maintenance services. Such full outsourcing service
is expected to offer greater global efficiency. While traditional monopoly bundles of
terminal equipment such as computers and telecommunications services that were
challenged in historic cases3 might be obsolete, the complexity of the ICT used by
the MNEs and the associated need for the interworking of software, hardware, and
connectivity would be expected to put providers of full service solutions in a bene-
ficial market position. This may be the case with anything from video-conferencing
solutions to operating international product call centres.
However, outsourcing MNEs’ communications solutions to the provider of net-
work connectivity is no longer the only option, and there may be other companies
able to provide a more efficient and cheaper yet still global outsourced service. The
first reason for that is that universal internet (IP) connectivity makes it increasingly
difficult to bundle any additional equipment or service with it. An additional obs-
tacle for that may be emerging ‘net neutrality’ laws, although the latter might be
more lenient regarding specialized corporate services such as virtual private net-
works (VPN).4 The second reason is the diminishing role of specialized on-premises
equipment such as public branch exchanges (PBX), which may be increasingly
substituted by software running on standard end-user computer equipment or in
the cloud.5 Modern age telecommunications outsourcers can therefore offer their
2
Whereas some countries such as Germany only subject to licences or authorization ‘public’ telecommu-
nications services ie those services that are offered to the public (eg Germany, Telekommunikationsgesetz
(TKG) § 6), others such as the UK regulate the provision of any (tele)communications services (see s 32 of the
Communications Act 2003).
3
eg Re Cordless telephones in Germany [1985] 2 CMLR 397 and the Computer Inquiry cases, in Chapter 5,
at 5.2.5.1.
4
See Art 3(5) of the Regulation (EU) 2015/2120 of the European Parliament and of the Council of 25
November 2015 laying down measures concerning open internet access and amending Directive 2002/22/
EC on universal service and users’ rights relating to electronic communications networks and services and
Regulation (EU) No 531/2012 on roaming on public mobile communications networks within the Union.
5
See ‘Telecomulonimbus: cloudification will mean upheaval in telecoms’, The Economist, 12 April 2017.
628
6
See eg ‘Report Enabling the Internet of Things’, BoR (16) 39, 12 February 2016.
7
<https://slack.com/>.
8
See eg <http://searchunifiedcommunications.techtarget.com/a nswer/Is-UC-as-a-Service-s ynonymous-
ith-cloud-UC>.
w
629
9
<http://searchcloudcomputing.techtarget.com/definition/I nfrastructure-a s-a-Service-IaaS>.
630
States.10 Still, mobile voice and messaging apps, better affordability of mobile data
and ubiquitous WiFi, lately coupled with the ability of employees of large com-
panies being able to use their own mobile terminal equipment (ie ‘Bring-Your-
Own-Device’), promise to move a great deal of corporate mobile communications
to the OTT app or CaaS level, increasingly bypassing mobile operators’ own voice
and SMS services.
10
See Arts 3 and 4 of the Commission Implementing Regulation (EU) 2016/2286 of 15 December 2016 laying
down detailed rules on the application of fair use policy and on the methodology for assessing the sustain-
ability of the abolition of retail roaming surcharges and on the application to be submitted by a roaming pro-
vider for the purposes of that assessment.
11
See further Chapter 16, at Section 16.4.1.
12
See eg Case 21/72, International Fruit Company NV and others v Produktschap voor Groenten en Fruit
[1972] ECR 1219, para 27.
613
Within the EU, the system of general authorizations in individual Member States
does not provide for a reliable ‘passport’ when the same entity wants to offer elec-
tronic communications services outside the Member State where it is subject to
general authorization.13 Obtaining such an authorization would in most Member
States mean notifying the relevant Member State’s NRA.14 Despite the general EU
Treaty principle of free movement of services, in UPC v NMHH15 the ECJ ruled that
Member States are not precluded from requiring undertakings which supply elec-
tronic communications services in their territory to register those services with
the same Member State’s NRA, regardless of their general authorization based on
registration in another Member State. The ECJ, however, also ruled that such pro-
viders cannot be required to establish in the Member State a branch or a legal en-
tity separate from that located in the Member State from where they provide the
service, which means that on-site work at the relevant MNE premises may always
be provided by travelling personnel or subcontractors.16
One should note, however, that such limitations to free movement of service
would not apply to those types of telecoms-related outsourced services that do not
qualify as electronic communications ie that do not ‘wholly or mainly’ consist in
the ‘conveyance of signals on electronic communications networks’.17 It is there-
fore important to consider, if and to what extent do modern services for MNEs fall
under traditional definitions of regulated communications services.
13
See further Chapters 4 and 6.
Art 3(2) of the amended Authorisation Directive. The UK does not require notification unless a certain
14
services revenue threshold has been reached according to Ofcom’s Tariff Tables: Administrative Charges for
the Networks and Services Sector. A one-stop shop system whereby BEREC forwards the notifications to indi-
vidual NRAs concerned is currently proposed under Art 12(3) of the Proposal for a Directive of the European
Parliament and of the Council establishing the European Electronic Communications Code, COM(2016) 590
final, 14 September 2016.
15
Case C‑475/12, UPC DTH Sàrl v Nemzeti Média-és Hírközlési Hatóság Elnökhelyettese, Judgment of the
Court (Second Chamber), 30 April 2014.
16
Ibid. 17
See Art 2 of the amended EU Framework Directive.
18
See eg amended Authorisation Directive, Arts 3 and 6, and the Annex.
632
19
BEREC Report on OTT Services, January 2016, at 16. 20
Ibid.
21
See Chapter 5, at 5.2.5.4.
22
Art 2(6) and (7) of the Proposal for a Directive of the European Parliament and of the Council establishing
the European Electronic Communications Code, COM(2016) 590 final, 14 September 2016.
23
Chilean regulator SUBTEL, eg, lists video as one of the possible components of IP telephony. See <http://
www.subtel.gob.cl/telefonia-ip/>.
24
According to Swiss regulator BAKOM, ‘a telecommunications service provider is . . . a natural or legal
person who itself transmits or arranges to transmit information using telecommunications for third parties
and assumes responsibility for the provision of the promised service in respect of these third parties within
the framework of a contractual relationship under private law.’ Guide to the Registration Form for Providing
Telecommunications Services, 1 May 2010.
63
25
Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal
aspects of information society services, in particular electronic commerce, in the Internal Market, OJ L 178,
17 July 2000.
26
According to Directive (EU) 2015/1535 of the European Parliament and of the Council of 9 September
2015 laying down a procedure for the provision of information in the field of technical regulations and of rules
on Information Society services, OJ L 241/1, 17 September 2015, Art 1 (1)(b), an ‘information society service’
is ‘any service normally provided for remuneration, at a distance, by electronic means and at the individual
request of a recipient of services’.
27
See eg § 6 Telekommunikationsgesetz (TKG), German Federal Telecommunications Law.
28
Art 2(c) of the amended EU Framework Directive.
29
See Explanatory Memorandum to the Proposal for a Directive of the European Parliament and of the
Council establishing the European Electronic Communications Code, COM(2016) 590 final, 14 September 2016.
364
for the EU ePrivacy Regulation recognizes the problem of integrated online en-
vironments, and expressly extends the scope of e-privacy obligations to ‘services
which enable interpersonal and interactive communication merely as a minor
ancillary feature that is intrinsically linked to another service.’30 Such services
are otherwise excluded from telecommunications market regulation even under
the proposed European Electronic Communications Code.31 On the other hand,
regulators may altogether adopt a functional approach and look at the transmis-
sion purpose of the service instead of the provider’s control over the conveyance
element. For example, BNetzA, the German regulator, considered Gmail a public
electronic communications service, which was upheld by a German administra-
tive court: the only purpose for using Gmail was to convey messages from sender
to recipient, and even though Google itself does not provide the actual signal
transmission between the servers that are involved in email delivery, the court
held that this transmission via the open internet must still be attributed to Google
because the transmission was triggered by Gmail users hitting the ‘send’ button
and facilitated by Google’s email servers.32 However, upon appeal, the Higher
Administrative Court in Münster in February 2018 ruled that the case be referred
to the European Court of Justice in order to clarify ‘whether internet-based e-mail
services provided via the open internet, which do not themselves provide internet
access, need to be covered as transmission of signals over electronic communica-
tion networks’.
30
Art 4(2) of the Proposal for a Regulation of the European Parliament and of the Council concerning the re-
spect for private life and the protection of personal data in electronic communications and repealing Directive
2002/58/EC (Regulation on Privacy and Electronic Communications), COM(2017) 10 final, 10 January 2017.
31
According to Art 2(5) of the Proposal for a Directive of the European Parliament and of the Council
establishing the European Electronic Communications Code, COM(2016) 590 final, 14 September 2016, ‘inter-
personal communications service’ means ‘a service normally provided for remuneration that enables direct
interpersonal and interactive exchange of information via electronic communications networks between a
finite number of persons, whereby the persons initiating or participating in the communication determine its
recipient(s); it does not include services which enable interpersonal and interactive communication merely as
a minor ancillary feature that is intrinsically linked to another service.’
32
Administrative Court of Cologne, Judgment of 11 November 2015.
653
made easier through WTO rules and a general relaxation in licensing regimes
worldwide, actual provision of service may require access to network elements of
the host country incumbent operator(s).
Several inputs may be largely the same as in case of the provision of other end-
user communications. Whereas the capacity required for the MNEs will be higher
than in case of residential or small business markets, there might be little or no
difference in the network elements used. New entrants are in both cases expected
to require facilities such as unbundled local loops, backbone connectivity, which
may include active transmission via IP/MPLS networks,33 or even direct access to
submarine cable landing stations. All these facilities may be subject to wholesale
regulation irrespective of whether they are required to provide the service in the
residential or in the business market.34
However, corporate markets are often particularly difficult to tackle for new en-
trants because they require high-quality services not always available in the in-
cumbent operators’ standard offerings. This is often due to the regulators’ focus on
the retail market where the bargaining power of the end-users is lower. Moreover,
one has to keep in mind that the market for MNE services is a lucrative market that
the incumbent operators are not comfortable to lose to the new entrants—who
might be larger than them internationally and might enjoy an established rela-
tionship with the MNEs in their home countries. Limiting their wholesale offering
to low-speed TDM leased lines that have long been the regulators’ focus35 whilst
restricting access to high-speed Ethernet links and IP/MPLS backbone may be a
possible anti-competitive strategy.
With the specific requirements of the business markets in mind, the European
Commission in 2014 redefined the previously regulated market for the terminating
segments of leased lines, transforming it into the market for ‘wholesale high-quality
access’ (known as the new ‘Market 4’).36 This separate market definition builds on
the idea that the last mile remains the key bottleneck for corporate connectivity in
33
See eg <http://w ww.commverge.com/Solutions/I PCoreEdgeNetworks/I PMPLSNetworks/t abid/169/
Default.aspx>.
34
See further Chapter 8, Section 8.5.
35
It was only in 2007 that European Commission adapted the wholesale leased lines market to the require-
ments of the IP data-d riven communications by adding ‘irrespective of techology used’. See Commission
Recommendation on Relevant Product and Service Markets within the electronic communications sector
susceptible to ex ante regulation in accordance with Directive 2002/21/EC of the European Parliament and
of the Council on a common regulatory framework for electronic communications networks and services
(Second edition).
36
Market 4 of the Commission Recommendation 2014/710/EU of 9.10.2014 on relevant product and service
markets within the electronic communications sector susceptible to ex ante regulation in accordance with
Directive 2002/21/EC of the European Parliament and of the Council on a common regulatory framework for
electronic communications networks and services.
63
EU Member States, and that regulating the physical building blocks of the network
such as the copper local loop or the resale of regular residential broadband prod-
ucts would not enable sufficient level of competition in the business market.37 The
exact scope of the regulated wholesale products may vary from Member State to
Member State, and may comprise not only broadband transmission but also (fully
‘unbundled’ or ‘dark’) fibre access and (shared) wavelength division multiplexing
(WDM) fibre access, which uses different colours of light to differentiate multiple
operators’ signals in a single optical link.38
While some of the products such as WDM may be suited to large business offer-
ings by technology alone, others such as asymmetric broadband via copper pairs
or hybrid copper-fibre networks may require additional characteristics: guaran-
teed availability and high quality of service in all circumstances, including service
level agreements (SLAs), 24/7 customer support, short repair times and redundancy;
high-quality network management, not only in the access network but also in the
backhaul network segment, resulting in upload speeds appropriate for business use
and very low contention; the possibility to access the network at points which have
been defined according to the geographic density and distribution of business such
as business parks; or the possibility to offer separate Ethernet continuity eg through
an additional header allowing for several layers of virtual LANs.39 These wholesale
regulated features would normally be reflected in end-user agreements between the
operators and MNEs or other large corporate entities.
37
‘Leased lines may be provided using a range of technologies. Legacy options (so-c alled “traditional”
interface leased lines) include low-bandwidth analogue leased lines and digital lines at a wide range of band-
widths, for example, via SDH/PDH or TDM-based technologies. These are usually point-to-point connections.
Increasingly, leased lines are offered over Ethernet-based technologies, allowing more flexibility, normally at
a lower cost, and can be both PtP and PtMP. Ethernet-based leased lines, in particular carrier-g rade Ethernet
with larger frames, have been found substitutable to legacy traditional leased lines in most Member States.’
Explanatory Note to the Commission Recommendation of 9 October 2014 on relevant product and service
markets, at 47.
38
See eg UK Case UK/2016/1849: Market for wholesale high-quality access provided at a fixed location in
the United Kingdom and Dutch Case NL/2017/1960: Wholesale high-quality access provided at a fixed loca-
tion (Market 4).
39
Explanatory Note to the Commission Recommendation of 9 October 2014 on relevant product and service
markets, at 47.
40
<https://w ww.iana.org/numbers>.
673
• network access, which may comprise both the Internet and PSTN access;
• transmission services, which may comprise virtual private network (VPN) be-
tween locations, calls and text messaging (SMS) at designated rates;
41
CEPT/ECTRA Decision of 2 December 1999 On European Telephony Numbering Space (ETNS) conven-
tions (ECTRA/DEC(99)04) Revised version by written procedure 12 January 2001.
42
See also Chapter 4, at Section 4.5.
43
Associating numbers with geographic areas is also a premise of Recommendation ITU-T E.164 (11/2010).
44
See <https://w ww.bundesnetzagentur.de/S haredDocs/D ownloads/E N/BNetzA/P ressSection/P ress
Releases/2017/27012017_ r ufnummern.pdf?_ _blob=publicationFile&v=2>.
638
with a view to the integration of the legacy systems such as company computers
and phones with their CaaS counterparts. One should further note that, even in
a cloud communications environment, a traditional operator role would still be
required to ensure the connectivity between the MNE premises and the hosted so-
lution or the MNE’s own data centre, for example via leased Ethernet links, which
may be provided in a contract with the same or a different provider.
As for the cloud service itself, a key new dilemma emerges in relation to the solu-
tion hosting. Whereas in most retail internet-based services the physical location
of the hosting server may be irrelevant at least from the end-user perspective, the
situation is likely to be different in case of corporate (tele)communications for the
reasons of capacity, latency, security, and regulation, the latter potentially com-
prising rules on personal data exports or data residency. Accordingly, the MNE
and the service provider will need to agree on a solution that will satisfy both the
MNE’s commercial needs and the regulatory requirements. Where location of the
hosted CaaS solution is critical for the company’s security purposes, the company
may choose to host the solution on their own servers locally, or in their own local
or remote data centre. One should keep in mind though that security in such cases
is more likely to be a matter of company policies than actual risks: major cloud
providers have been focusing on minimizing cloud security risks for years, util-
izing the ability of cloud solutions to provide redundancy or even distributing files
over multiple data centres to avert the risks of data loss and contain the impact of
potential data breaches. That said, data residency requirements may apply in dif-
ferent countries for various reasons. EU data protection laws have long restricted
export of personal information to countries that do not sufficiently address privacy
concerns.48 Such laws may further address government control over information
flows or security concerns, such as in Russia and China.49
Apart from the location of the hosted solution, the contract will need to address
the type of hosting. In a multi-tenant public cloud solution, a single instance of
software running on a server is used by several end-users (tenants), whereby
this instance would be divided or partitioned in order to ensure data confiden-
tiality.50 Whereas multi-tenancy is a typical solution for small businesses, MNEs
may use single-tenancy private cloud solutions whereby each company uses
its own instance of the software application and supporting infrastructure51
See Art 25 of the Directive 95/4 6/EC and Article 44 of the Regulation (EU) 2016/679. See further Chapter 13.
48
See Determann, L, ‘Local Data Residency Requirements for Global Companies’ (Baker & McKenzie,
49
52
See eg European Commission Recommendation of 7 May 2009 on the Regulatory Treatment of Fixed and
Mobile Termination Rates in the EU, 2009/396/EC.
614
ISDN30 because they support up to thirty voice telephone channels. These chan-
nels are then allocated to individual users by PBX on the corporation’s premises,
or by the telephone operator’s own central location switching solution (Centrex).
ISDN30 may by the regulators still be perceived as bottleneck and offered at regu-
lated rates.53 In the CaaS context, however, ‘per user’ charging does not refer to
the number of user channels enabling communication via the access network but
rather to the number of employees that are simultaneously able to use the commu-
nications platform, regardless of where they are physically, and independently of
the capacity of the access network that a cloud provider would normally have no
control of.
Capacity-based charging models have traditionally been used for large cor-
porate communications. Point-to-point connections such as leased lines, including
Ethernet links, or dark fibre have been charged for based on capacity such as the
speed of the link (Mb/s) or the number of optical fibres leased. Capacity-based
charging can also be used in an IaaS model whereby the customer company is
charged per number of individual hardware items and their capacity. With the
introduction of flexible, software-based cloud communications technologies that
enable constant access control and usage monitoring, capacity-based charging is
becoming ‘smarter’. Communications capacity actually used and additional fea-
tures may be charged for in addition to or instead of the basic communications
software licence. Such charging practices are converging with innovative usage
fees in the context of ‘pay per use’ software licensing otherwise well-k nown for
SaaS products.54
53
See eg Commission Decision concerning Case UK/2014/1607: Fixed narrowband wholesale services
in the United Kingdom; fixed narrowband retail services in the Hull Area; and retail ISDN2 exchange line
services in the United Kingdom excluding the Hull Area. Comments pursuant to Article 7(3) of Directive 2002/
21/EC.
54
<https://apprenda.com/l ibrary/g lossary/definition-payperuse-software-saas/>.
624
634
Part V
13
13.1 INTRODUC TION
The right to privacy of communications is one of the most enduring and widely rec-
ognized of the constellation of rights known as privacy law. While privacy remains a
notoriously elusive concept, our communications activities have been consistently
recognized as a fundamental element of our private life, placed side-by-side with no-
tions of family and home. While the terminology used in international instruments
and national constitutions may have evolved over time to reflect changing technolo-
gies, from ‘correspondence’ to ‘communications’,2 the centrality of communications
privacy to our private life remains undisputed.
Despite its constant presence, the nature of communications privacy has evolved
greatly since it made its appearance in the UN Declaration. In 1948, the primary
relationship being governed was that between the individual and the state, pro-
tecting the former from arbitrary interference by the latter. In modern regimes,
1
With thanks to Christopher Millard for writing and updating the chapter in the first four editions, as well
as reviewing this new chapter.
2
See the Universal Declaration of Human Rights (1948), at Art 12, and the Charter of Fundamental Rights
of the European Union (‘Charter’) (2000), at Art 7, respectively.
64
manifest most clearly within European Union (EU) law,3 other communication
privacy relationships have also come to be recognized as requiring express legal
protection and regulatory intervention. First, with liberalization and privatization,
the state-owned incumbent became distinct from the state and part of a global in-
dustry comprising a multitude of private entities. In the course of entrusting these
private entities with the content of our communications, they obtain vast amounts
of data about our communications activities, both qualitative and quantitative,
embodying substantial potential value in an information economy. Controlling
the commercial interests of service providers to use and abuse this data has
therefore become an important component of modern communications privacy.
Second, a distinction must be made between the privacy interests of the person
that contracts with a provider for the provision of services, generally referred to as
a subscriber, and a user of that service. Two key examples being an employer and
employee or a parent and child. To what extent should the law intervene to protect
the privacy interests of the user from the capability of the subscriber to interfere in
the user’s private communications? Lastly, there is a privacy relationship between
the actual parties to the communication, the calling and the called party or sender
and recipient. Each of these four distinct privacy relationships, User–State, Service
provider–Subscriber, Subscriber–User, and User–User, has become ever more im-
portant in our communications environment.
This chapter examines how each of these four privacy relationships has be-
come the subject of legal intervention, designed to protect and balance the various
privacy interests of the parties. In the EU, such legal intervention has appeared
in the form of data protection laws, specifically the Data Protection Directive
(DPD),4 the PEC Directive and the General Data Protection Regulation (GDPR),5
and proposed reform of the PEC Directive.6 While data protection law is closely
related to and overlaps with the right to privacy, it also contains some unique
characteristics that distinguish it from traditional notions of privacy law, as most
clearly expressed in the Charter’s recognition of them as separate rights.7 First,
the data protection right is applicable to all data about a person, whether that
data can be said to be of a private or public nature. Second, personal data can
3
ie Directive 02/58/EC concerning the processing of personal data and the protection of privacy in the elec-
tronic communications sector, OJ L 201/37, 31 July 2002 (‘the PEC Directive’).
4
Directive 95/4 6/EC on the protection of individuals with regard to the processing of personal data and on
the free movement of such data, OJ L 281/31, 23 November 1995.
5
Regulation (EU) 2016/679 on the protection of natural persons with regard to the processing of personal
data and the free movement of such data, OJ L 119/1, 4 May 2016.
6
Proposal for a Regulation concerning the respect for private life and the protection of personal data in
electronic communications, COM(2017) 10 final, 10.1.2017 (‘ePrivacy proposal’).
7
Charter at Art 7, ‘Respect for private and family life’ and Art 8, ‘Protection of personal data’.
674
8
See eg Lynsky, O, The Foundations of EU Data Protection Law (OUP, 2015); and Tzanou, M, ‘Data Protection
as a Fundamental Right next to Privacy? “Reconstructing” a not so New Right’, (2013) 3(2) International Data
Privacy Law 94.
9
See recital 1.
10
See cases C-4 65/0 0, Rechnungshof v Österreichischer Rundfunk [2003] 3 CMLR 10, at para 70; C-92/0 9 and
C-93/0 9, Volker and Markus Schecke GbR and Hartmut Eifert v Land Hessen [2012] All ER (EC) 127, at para 52;
and Tele2 Sverige AB v Post-och Telestyrelsen [2017] 2 CMLR 30.
11
OJ C 277, 5 November 1990.
12
Directive 97/6 6/EC concerning the processing of personal data and the protection of privacy in the tele-
communications sector, OJ L 24/1, 30 January 1998.
684
Protection Directive or GDPR. Conversely, the GDPR expressly states that it does
not impose ‘additional obligations . . . in connection with the provision of publicly
available electronic communication services . . . in relation to matters for which
they are subject to specific obligations with the same objectives . . .’.13
Directive 97/66/EC and the PEC Directive were transposed into UK law through
a combination of primary legislation and secondary regulation. Article 5(1) and (2)
were implemented through Part I of Regulation of Investigatory Powers Act 2000,14
as replaced by the Investigatory Powers Act 2016, while the remainder is contained
in the Privacy and Electronic Communications (EC Directive) Regulations 2003.15
The PECR has been amended on a number of occasions, but the primary reform
occurred in 2011, transposing the amendments made to the PEC Directive by the
‘Citizens’ Rights’ Directive in 2009.16
This chapter examines the complex nature of each of the four privacy relation-
ships, focusing on two particular aspects. First, the obligations placed upon tele-
communications providers to regulate their conduct in order to either prevent
abusive practices by them or third parties, or facilitate the exercise of privacy rights
by their subscribers and users. Second, the evolving legal treatment of key types
of personal data generated through the use of telecommunications services: com-
munications content, traffic and location data, and subscriber data.
Before examining the different privacy relationships, it is necessary to examine
the terms and terminology used in EU law to set the regulatory boundaries of the
communications privacy regime.
13. 2 DEFINING TER MS
13
GDPR, at Art 95. 14
Explanatory Notes to the Regulation of Investigatory Powers Act, at para 9.
15
SI 2003/2 426 (PECR), which replaced the Telecommunications (Data Protection and Privacy) Regulations
1999, SI 1999/2093.
16
Privacy and Electronic Communications (EC Directive) (Amendment) Regulations 2011, SI 2011/1208.
17
This has been consistent throughout its history, from the International Telegraph Convention (1865), at
Art 5, to the current ITU Constitution (2014), at Art 37(1). See further Chapter 16.
694
18
eg Huvig v France (1990) 12 EHRR 528, paras 8 and 25. In Bărbulescu v Romania (5 September 2017), the
Grand Chamber of the ECtHR notes that the notion of ‘correspondence’ is ‘not qualified by any adjective, un-
like the term “life”. And indeed, the Court has already held that, in the context of correspondence by means of
telephone calls, no such qualification is to be made’ (para 72).
19
Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data,
Strasbourg, 28 January 1981 (European Treaty Series No. 108) (Cmnd 8341) (London: HMSO, 1981). See also
Recommendation R 95(4) ‘on the protection of personal data in the area of telecommunication services, with
particular reference to telephone services’.
20
Ibid, at Art 3(2)(b). See generally Walden, I and Savage, N, ‘Data Protection and Privacy Laws: Should
Organisations be Protected?’, (April 1988) 37 The International and Comparative Law Quarterly 337.
21
The 2016 Proposal would make this explicit, although it would seem implicit under the current regime.
The Internet of Things is the context for the current concern, where personal devices around the home com-
municate with each other, but it would be equally applicable to industrial EDI systems, such as deployed in the
automotive and transportation industries.
22
See van der Sloot, B, ‘Do Privacy and Data Protection Rules Apply to Legal Persons and Should They?
A Proposal for a Two-t iered System’, (2015) 31 Computer Law and Security Review 26, at 32.
23
eg Klass v Germany (1978) 2 EHRR 214, re: telephone conversations and Copland v UK (2007) 45 EHRR 37,
re: email and internet usage.
24
PEC Directive, at Art 3.
560
25
Proposal for a Directive establishing the European Electronic Communications Code, COM(2016) 590
final, 14 September 2016 (2016 Proposal), at Art 2(5): ‘a service normally provided for remuneration that en-
ables direct interpersonal and interactive exchange of information via electronic communications networks
between a finite number of persons, whereby the persons initiating or participating in the communication
determine its recipient(s); it does not include services which enable interpersonal and interactive communi-
cation merely as a minor ancillary feature that is intrinsically linked to another service.’
26
ePrivacy proposal, at Art 4(2).
27
eg The Guardian, ‘Publishers call for rethink of proposed changes to online privacy laws’, 29 May 2017,
<https://w ww.theguardian.com/media/2 017/m ay/2 9/p ublishers-c all-f or-r ethink-of-proposed-c hanges-
to-online-privacy-laws> and Business Insider UK, ‘European regulators are about to kill the digital media
industry’, 14 August 2017, <http://u k.businessinsider.com/european-regulators-a re-about-to-k ill-t he-
d igital-media-i ndustry-2017-8>.
28
Oftel, ‘Guidelines for the interconnection of public electronic communications networks’, at 6.1.
29
Halford v United Kingdom (1997) 24 EHRR 523, at para 56.
615
any limitation in the PEC Directive must be viewed in the context of the con-
tinuing applicability of the DPD and GDPR. As well as being limited to publicly
available services, the PEC Directive is also limited to communications between
a ‘finite number of parties’, excluding information that is broadcast to the public
(Article 2(d)). The rationale being that the broadcaster, as sender of the communi-
cation, is using a medium that is not inherently confidential, although personal
data obtained about any recipient of the broadcasting service would be covered
(recital 16).30
While this chapter generally refers to service providers as the regulated entities,
the PEC Directive regime can extend both to the operators of networks as well as
suppliers of ‘terminal equipment’, from mobile handsets to desktop computers.31
Member States may require that networks and equipment have ‘specific technical
features’ to implement the provisions of the PEC Directive, but only as long as they
do not interfere with the functioning of the EU Single Market (Article 14). In respect
of the latter, EU law expressly states that ‘radio equipment’ sold in the EU should
comply with the ‘essential requirements’, one of which ‘incorporates safeguards
to ensure that the personal data and privacy of the user and of the subscriber are
protected’.32 Terminal equipment is also increasingly viewed as an element of a
person’s private life,33 more akin to his home or indeed even more private, as noted
by the US Supreme Court in Riley v California:
a cell phone search would typically expose to the government far more than the
most exhaustive search of a house: A phone not only contains in digital form many
sensitive records previously found in the home; it also contains a broad array of
private information never found in a home in any form –unless the phone is.34
Reference has already been made to the qualitative and quantitative explosion
in data being generated and processed by service providers in our modern com-
munications environment as we spend more of our lives online. In terms of com-
munications privacy, however, not all data is equal and it needs to be further
distinguished into three broad categories: communications content, traffic data,
and subscriber data. There is widespread consensus, reflected in both privacy and
30
See generally Walden, I and Woods, L, ‘Broadcasting Privacy’, (2011) 3(1) Journal of Media Law 117.
Commission Directive 2008/6 3/EC on competition in the markets for telecommunications terminal
31
data protection laws, that the processing of each category engages our private life
to differing degrees, although to what extent is increasingly debated.
Communications content is defined by the PEC Directive as ‘any information ex-
changed or conveyed between a finite number of parties’ (Article 2(d)), which, as al-
ready noted, excludes broadcast content. Key to understanding this definition is that
it is not the private or confidential nature of the content itself that triggers the regime,
since it could involve both publicly available and non-personal information, but the
act of communication between specific parties that is being protected as private.
While communications content is broadly defined, two modes of communi-
cating content are further distinguished: ‘calls’ and ‘electronic mail’. A definition
for a ‘call’ was provided in the original PEC Directive, but was subsequently de-
leted in 2009, despite its continued usage, including in the ePrivacy proposal. It is
generally used in the context of voice telephony services, rather than data services.
‘Electronic mail’ encompasses any type of message content, including voice, but
it must be capable of being stored, either in the network or the recipient’s terminal
equipment (Article 2(h)). A live ‘call’, therefore, would fall outside the definition,
unless it was being recorded. While ‘calls’ are referred to throughout the PEC
Directive, the distinction between calls and electronic mail only becomes relevant
in the context of unsolicited communications.
Traffic data is defined in the PEC Directive as:
any data processed for the purpose of the conveyance of a communication on an
electronic communications network or for the billing thereof (Article 2(b))
This data represents the attributes of communications activities. The when, where,
and how of our communication’s conduct. While data concerning a person’s lo-
cation (the where) will clearly fall within the definition of traffic data, the PEC
Directive also governs the processing of location data generated in relation to
the use of communication services, through GPS, cell site triangulation, or WiFi
hotspots, but distinct from the communication process itself. For fixed line com-
munication services, location data is revealed by a billing address (as subscriber
data) and, in broad terms, by the use of a geographic number (as traffic data), but
it is the capabilities of digital mobile networks and devices that have given rise to
privacy concerns and regulatory intervention (recital 35).
Historically, traffic data has not been considered to engage our privacy interests
at the same level as access to the content of our communications. While this view
is increasingly contested, the CJEU seems to continue to differentiate between ac-
cess to the content of communications, which can ‘affect adversely the essence’ of
our rights to privacy and data protection,35 from traffic data, despite commenting
35
Tele2 Sverige AB v Post-och Telestyrelsen [2017] 2 CMLR 30, at para 101. A similar sentiment can be found
in ECtHR jurisprudence, see Malone v United Kingdom (1985) 7 EHRR 14.
653
36
Tele2, n 35, at para 99.
37
This equality represents a shift from Directive 97/6 6/EC, which only addressed the confidentiality of
communications content (Art 5(1)).
38
See Home Office, ‘Acquisition and Disclosure of Communications Data’, at para 2.31.
39
(2016) 63 EHRR 17.
40
Concurring opinion of Judge Pinto de Albuquerque, in Sazbó and Vissy v Hungary, 12 January 2016, at 1.
41
eg case C-275/0 6, Promusicae v Telefonica de Espana SAU [2008] 2 CMLR 17.
564
the form of the LEA exercising investigative powers, in the latter, it is the courts
that represent the state, authorizing the privacy intrusion, to ‘protect the rights
and freedom of others’ (ECHR, Article 8(2)).
As with other topics covered in this book, this law enforcement need has
resulted in the imposition of both ex ante and ex post regulatory obligations
on service providers. In terms of ex ante obligations, this requires service pro-
viders to conduct themselves in certain ways in order to be in a position to as-
sist law enforcement at some point in the future. The two key examples of this
are obligations to build an ‘intercept capability’ into a network or service,42
and requirements to retain data concerning usage of the network or service for
designated periods of time. Ex post obligations arise when an LEA contacts the
service provider to request the obtaining and disclosure of data in respect of a
specific investigation. Such requests may be in respect of targeted suspects or
be bulk requests in respect of particular categories of communication. Again,
as with other topic areas, there are two jurisdictional dimensions to law en-
forcement obligations: whether the applicable service falls within the regu-
lated sphere and whether the obligations are applicable to service providers
located outside the state but providing services into the state. This section
examines the rules imposed on service providers to assist LEAs, primarily
under EU and UK law.
13.3.1 EU law
One of the first occasions that issues of state interference with an individual’s
communications was formally addressed by the EU was a 1995 resolution of the
Council of the European Union, as an aspect of police cooperation in the fight
against terrorism and serious crime.43 It was triggered by concerns that the on-
going liberalization of the telecommunications sector and rapid technological de-
velopments were generating problems for Member States in terms of interception
capabilities. The main part of the Resolution comprises an Annex detailing a set of
requirements that law enforcement agencies had with regard to the intercept cap-
abilities provided by service providers, including access to ‘call associated data’.44
42
eg in the US, the Communications Assistance for Law Enforcement Act Pub. L. No. 103–414, 108 Stat. 4279
(1994).
43
Council Resolution of 17 January 1995 on the lawful interception of telecommunications, OJ C 329/1, 4
November 1996.
44
Defined as ‘Signalling information passing between a target service and the network or another user.
Includes signalling information used to establish the call and to control its progress (e.g. call hold, call hand-
over). Call associated data also includes information about the call that is available to the network operator/
service provider (e.g. duration of connection).’
65
In response, the Article 29 Working Party (A29WP), established under the Data
Protection Directive, reminded Member States of the need to respect privacy and
detailed its own list of minimum requirements for national law to protect funda-
mental rights.45 Later that same year, the A29WP got in first and, in response to
proposals for data retention from the G8 high-tech crime subgroup, issued a rec-
ommendation stating that ‘traffic data should in principle not be kept only for law
enforcement purposes’.46
The PEC Directive, and its predecessor Directive 97/66/EC, address state inter-
ference in two provisions. First, Member States are required to ensure the con-
fidentiality of communications and related traffic data, by prohibiting ‘listening,
tapping, storage or other kinds of interception or surveillance of communications’
(Article 5(1)). This default position may then be made subject to national law ex-
ceptions where based on a ‘necessary, appropriate and proportionate measure’ in
order to:
. . . safeguard national security (i.e. State security), defence, public security, and
the prevention, investigation, detection and prosecution of criminal offences or of
unauthorised use of the electronic communication system (Article 15(1)).
This effectively replicates the privacy exceptions detailed in Article 8(2) of the
ECHR. However, the provision then goes on to state that it is permissible for
Member States to adopt measures ‘providing for the retention of data for a limited
period’ (Article 15(1)) which ran counter to the opinion of the A29WP. Following
the terrorist attacks in Madrid in 2004 and London in 2005, the political mood was
in favour of mass retention of traffic data, which resulted in the adoption of the
Data Retention Directive in 2006.47 As a consequence of the extended competence
of EU law following the Lisbon Treaty, including the incorporation of the Charter
of Fundamental Rights, the CJEU has entered into the fray on User-State relation-
ships concerning communications privacy. Article 15(1) has since been subject to
considerable scrutiny by the CJEU, first in relation to the validity of the legislative
measure on data retention and second in relation to the applicable scope. In 2014,
the CJEU struck down the Data Retention Directive as invalid for being incompat-
ible with Articles 7 and 8 of the Charter.48 In 2016, the CJEU held that Article 15(1)
45
Recommendation 2/99 ‘on respect of privacy in the context of interception of telecommunications’
(WP18), 3 May 1999.
46
Recommendation 3/99 ‘on the preservation of traffic data by Internet service providers for law enforce-
ment purposes’ (WP25), 7 September 1999.
47
Directive 2006/2 4/EC on the retention of data generated in connection with the provision of publicly
available electronic communications services or of the public communications networks and amending
Directive 2002/58/EC, OJ L 105/5 4, 13 April 2006 (‘Data Retention Directive’).
48
See case C-293/12, Digital Rights Ireland Ltd v Minister for Communications, Marine and Natural
Resources [2014] 3 CMLR 44 (DRI).
56
precludes legislation that provides for the ‘general and indiscriminate retention
of all traffic and location data’.49 In addition, it held that retained data should not
be accessible by LEAs without prior review by ‘a court or an independent admin-
istrative authority’ for the purpose of tackling ‘serious crime’ and the data should
be retained in the EU.50 These judgments and their implications for data retention,
communications privacy, and, by association, service providers are not yet fully
worked through. However, any future initiative at EU level will need to be a pre-
ventative measure, targeted in some manner and restricted to ‘serious crime’.51 For
the UK, the government has conceded in the ongoing domestic litigation52 that the
threshold must be raised to ‘serious crime’53 and that there is a need for some form
of independent prior authorization before accessing retained data.54 The UK’s in-
tended departure from the EU could eventually lessen these constraints, although
not if they were considered an important consideration in any ‘adequacy’ deter-
mination under the GDPR.55
Under the PEC Directive, service providers are required to establish internal
procedures for responding to data requests by LEAs,56 although no distinction is
made between ex ante and ex post obligations. In addition, data about the number
of requests made by LEAs, the legal basis for the request, and the service provider’s
response should be reported to a competent authority, to enable oversight of the
exercise of these LEA powers. This provision implies that any data obtained from
service providers must be mediated through the service provider, rather than per-
mitting LEAs direct access to a service provider’s network. This echoes the senti-
ment of the ECtHR in Zakharov, when it notes that disclosing an authorization to
a service provider before obtaining access to a user’s data ‘is one of the most im-
portant safeguards against abuse’ (para 269).
49
Tele2, n 35, at 112.
50
Two further references have been made to the CJEU on issues arising from these cases: C-207/16,
Ministerio Fiscal, OJ C 251/7, 11 July 2016; and C-475/16, K, OJ C 428/8, 21 November 2016.
51
Tele2, n 35, at 108. It is somewhat ironic that the Data Retention Directive does not refer to prevention as
an aim (Art 1(1)), the word having been removed at the request of the Parliament.
52
Secretary of State for the Home Department v Tom Watson MP & ors [2015] EWCA Civ 1185.
53
Home Office, Investigatory Powers Act 2016: Consultation on the Government’s proposed response to the
ruling of the Court of Justice of the European Union on 21 December 2016 regarding the retention of commu-
nications data, November 2017 (‘Consultation 2017’). The Government intends to insert a bespoke definition,
distinct from the general definition at IPA, s 263(1), lowering the sentencing threshold from three years to six
months, as well as the likelihood from ‘reasonably expected’ to ‘capable of’ being so sentenced.
54
Consultation 2017, the government intends to make LEA access to communications data subject to ap-
proval by the Investigatory Powers Commissioner, except for national security purposes. The IPC will delegate
the function to a new Office for Communications Data Authorisations.
55
Art 45.
56
PEC Directive, at Art 5(1b), inserted by Directive 2009/136/EC (Citizens’ Rights’), Art 2(9), OJ L 337/11, 18
December 2009.
657
Information released by the former NSA contractor Edward Snowden from June
2013 onwards57 revealed the extent to which service providers cooperate with law
enforcement in terms of disclosing data. This resulted in considerable public con-
sternation and distrust towards service providers. In response, service providers
have tried to become more transparent about such practices, publishing ‘trans-
parency’ reports, in addition to any reporting obligations they have to oversight
authorities.58
13.3.2 UK law
In all jurisdictions, LEAs are granted a range of powers to assist in the pre-
vention, investigation, detection, and prosecution of criminal conduct. Such
powers include the right to engage in covert and coercive techniques, including
interfering with a person’s communication privacy, such as the interception of
communications. 59 Given the inevitable privacy intrusion involved, the exercise
of such powers will generally require some form of prior authorization, under
executive, judicial, or administrative procedures. In addition, not all LEAs are
equal, with intelligence agencies having distinct responsibilities to protect
national security and related enhanced powers. It is beyond the scope of this
section to examine the multitude of laws in the UK governing the conduct of
national security, criminal, or regulatory law enforcement agencies, but it will
be assumed that a requesting LEA is acting lawfully when making a request.60
Instead, the focus is on those laws that impose obligations on operators or ser-
vice providers to assist LEAs in the course of an investigation. UK law has re-
cently undergone reform, with parts of the current regime under the Regulation
of Investigatory Powers Act 2000 (RIPA) being replaced by the Investigatory
Powers Act 2016 (IPA), which also folds other existing powers into it.61 As such,
the following analysis takes into account both, due to their similarities, but con-
centrates on the latter.
In compliance with Article 5(1) of the PEC Directive, the IPA establishes a
general criminal prohibition on intentionally intercepting a communication
Google was the first, see <https://t ransparencyreport.google.com>. See also Vodafone’s ‘Law
58
‘in the course of its transmission’ (s 3(1)).62 The IPA then details the various
circumstances in which interception conduct has lawful authority, including
under other statutory powers or court orders (s 6). While the IPA is the primary
statutory provision on interception, other legislation could also be applicable
to equivalent conduct, including the Wireless Telegraphy Act 2006, Computer
Misuse Act 1990, and the Data Protection Act 1998.63
As discussed in Section 13.2 and earlier in this section, an initial issue to determine
is with whom any obligations lie? Under the IPA, the regulated entity is a ‘telecommu-
nications operator’, defined as a person that:
(a) offers or provides a telecommunications service to persons in the United
Kingdom, or
(b) controls or provides a telecommunication system which is (wholly or partly)—
(i) in the United Kingdom, or
(ii) controlled from the United Kingdom. (s 261(10))
The first element is a replication of the RIPA definition, while the Data Retention
and Investigatory Powers Act 2014 (DRIPA) inserted the second element into RIPA,64
before being replicated in the IPA. The amendment was designed to address per-
ceived legal uncertainties about the scope of the first phrase.65 It makes clear that
the definition should be not understood as being the same as that of an ‘electronic
communication service’ under current EU telecommunications law,66 but rather
encompasses a much broader range of service providers, including ‘web-based
62
This period includes any storage of a communication in a telecommunication system (s 4(4)). See
Edmondson and others v R [2013] EWCA Crim 1026. An offence of interception under UK law can be traced
back to the Telegraph Act 1868, s 20.
63
Sections 48, 1, and 55 respectively. Under the IPA, Ofcom is authorized to intercept for the purpose of the
enforcement of the Wireless Telegraphy Act 2006 (Art 48).
64
DRIPA, s 5.
65
See Walden, I, Computer Crimes and Digital Investigations (2nd edn, OUP, 2016), at 4.125 et seq.
66
See further Chapter 4, at Section 4.2.
659
See Home Office, ‘Interception of Communications Code of Practice’, January 2016, at 3.8.
67
69
eg IPA, s 43(3) (in relation to targeted interception), s 85 (in relation to the acquisition of communications
data), and s 97 (in relation to data retention); although not all obligations that have extraterritorial application
are enforceable through civil proceedings, eg s 97(2).
70
eg IPA, ss 43(5) and 85(4).
60
to minimize the amount of data generated and retained (eg Snapchat). In such an
environment, what is the scope of the duty on an operator to respond to a request?
Under the IPA, an operator is required to do that which is ‘reasonably practic-
able’,71 which although not further defined would imply that the LEA has to take an
operator as they find them, ie if the applicable system does not generate the data,
then the duty has been discharged. From a state perspective, to reduce such uncer-
tainty, the IPA provides that the Secretary of State may issue a notice to an operator
requiring them to implement a specific ‘technical capability’ (s 253).72 While the
exact requirements would be tailored to the operator, secondary legislation details
the types of capability that the operator may be required to implement, including
the timescales for disclosure, capacity (ie number of requests that can be handled
simultaneously) and the ability to ‘remove electronic protection applied by or on
behalf of the telecommunications operator to the communications or data’.73 This
last requirement has generated concern that operators could be required to build
back doors into the encryption systems they deploy, which may create a vulner-
ability that could be exploited for malicious purposes. The requirement repeats
an identical requirement under RIPA,74 but has significantly extended applica-
tion given the broader definition of a ‘telecommunication service’. However, the
requirement should not impact encryption systems applied, or controlled, by
subscribers and users.75 In meeting the requirements of a ‘Technical Capability’
notice, operators may be required to comply with industry standards, such as
those developed by the European Telecommunications Standards Institute.76 The
operator will also have an ongoing obligation to consider the requirements when
developing new systems and services.
Once a notice has been implemented, then the threshold for what is ‘reason-
able practicable’ for an operator becomes compliance with that capability.77
Building such a capability can obviously represent a significant compliance cost
for an operator and therefore the IPA makes provision for the possibility of pay-
ments being made by the government (s 249).78 From a market perspective, such
payments can be seen as preventing such capabilities from becoming a barrier
71
eg ibid, at ss 43(4), 66(3), and 128(5).
72
An operator can refer a notice back to the Secretary of State for review (s 257). When reviewing a notice,
the Technical Advisory Board and a Judicial Commissioner must be consulted.
73
The Investigatory Powers (Technical Capability) Regulations 2018, SI 2018/353.
74
RIPA, s 12 and the Regulation of Investigatory Powers (Maintenance of Interception Capability) Order
2002, SI 2002/1931.
75
eg end-to-end encryption, where the key is generated by and maintained on the user’s device (eg
WhatsApp). RIPA, Part III, contains powers to access protected data from a user.
76
See <http://w ww.etsi.org/technologies-clusters/technologies/lawful-i nterception>.
77
eg IPA, at s 43(6), 66(4), and 128(6).
78
Such financial assistance may be in the form of a grant, loan etc. (s 250(3)).
61
79
GCE, at 11.2. ‘Records’ mean ‘data or information showing the extent of any network or service actually
provided to an End-User and any data or information used in the creation of a Bill for an End-User’ (11.6(f)). In
the US, ‘telephone toll records’ must be retained for 18 months (47 CFR § 42.6). When considering the retention
period, the FCC received a request from the US Department of Justice requesting that the period be extended
from six months to eighteen months specifically because such records are ‘often essential to the successful
investigation and prosecution of today’s sophisticated criminal conspiracies’ (See Federal Register, vol. 50,
no. 149, 2 August 1985, at 31397.
80
PEC Directive, at Art 6(1). See further Section 13.4.
81
In Russia, Federal Law No. 374-FZ will require the retention of content for a period of six months. It is cur-
rently scheduled to come into force on 1 July 2018, although there have been calls for it to be delayed to 2023
due to the need to build sufficient data storage facilities.
82
Walden, n 65, at 4.224 et seq.
83
ie The Data Retention (EC Directive) Regulations 2009, SI 2009/859 (‘2009 Regulations’).
84
As of November 2017, the government has issued retention notices to fewer than twenty-five operators.
See Consultation 2017, at 14.
62
DRI, at para 67.
85
63
for collateral privacy intrusion, the regimes governing ‘bulk’ data collection and
equipment interference are much more narrowly prescribed.86
While interception and acquisition involves the direct obtaining of data from
an operator’s systems, equipment interference is primarily directed at obtaining
data from equipment being used by a suspect, such as a mobile device or net-
work. As such, the role of the operator in equipment interference is more facilita-
tive, assisting the LEA to carry out the equipment interference in order to enable
the LEA to obtain communications, equipment data, or any other information.87
Indeed, the applicable code of practice notes that in certain situations the LEA
may have to choose whether to acquire data with the assistance of an operator or
independently, a decision that should include considerations of proportionality.88
The Code details such proportionality criteria including the need to cause ‘least
possible interference to . . . others’, which could be read as steering LEAs away from
obtaining assistance; as well as whether the conduct has any implications ‘for the
privacy and security of other users of equipment and systems’ (4.18), on which an
operator would seem well placed to advise. Yet again, service providers can have a
mediating role in governing the User–State relationship.
The conditions for the issuance and implementation of warrants and notices
under the IPA are beyond the scope of this chapter, but it is worth noting that the
recipient service provider will not be in a position to challenge the validity of the
grant of the warrant or notice, eg with regard to purpose or proportionality, but
only in relation to procedural irregularities on the face of the documentation or its
service.89 Indeed, the IPA imposes an obligation on an operator to report ‘any rele-
vant error’90 to the Investigatory Powers Commissioner (s 235(6)).
Finally, service providers are under a duty not to disclose the existence or con-
tent of any notice or warrant they have received, although the consequences differ
between the ex ante obligations, which are enforceable through civil proceed-
ings,91 and the ex post obligations, for which disclosure is a criminal offence.92 This
reflects the fact that an unauthorized disclosure in the latter case could interfere
with a live investigation.
IPA, Pt 6.
86
Ibid, at s 99(2). A ‘communication’ is defined at s 135, while ‘equipment data’ is defined at s 100.
87
88
Home Office, ‘Draft Equipment Interference Code of Practice’, February 2017, at 7.11.
89
eg under the Home Office code of practice under RIPA, Acquisition and Disclosure of Communications
Data (March 2015), details of what a notice should include are specified (at 3.47).
90
IPA, s 231(9), which is defined as non-compliance by a public authority with a requirement, which is sub-
ject to review by a Judicial Commissioner, and is identified in a code of practice.
91
Ibid, at s 95(5), regarding data retention, and s 255(8) regarding a technical capability notice.
92
Ibid, eg at s 59 (interception), s 82 (acquisition of communications data).
64
While it is beyond the scope of this chapter to offer a detailed examination of the gen-
eral data protection regime and its potential applicability to the different parties dis-
cussed in this chapter,93 it is important to understand the status of a service provider
under data protection law, since the general applicability of the regime differs de-
pending on the nature of the role they play.
EU data protection law distinguishes between a ‘controller’ and a ‘processor’ of
personal data. The former is the person that ‘determines the purposes and means’ of
the processing, while the latter processes the data on the controller’s behalf.94 Under
the DPD regime, the regulatory obligations reside solely with the controller, while the
processor is only subject to contractual obligations (Article 17(3)). Under the GDPR,
this changes somewhat, with the processor being made subject to certain direct
regulatory obligations (Article 28), although the general model remains the same.
However, in the PEC Directive and the ePrivacy proposal, the controller/processor
distinction is not expressly used in the substantive provisions to distinguish the re-
spective roles of the service provider, subscriber, or user.95 This potentially extends
liability for non-compliance to persons acting in the role of a processor. Under PECR,
for example, it states that ‘a person shall neither transmit, nor instigate the transmis-
sion of’ unsolicited communications (reg 20), which could include an intermediary
service provider who supplies a facility to enable the calls to be made.96
In a traditional telecommunications context, a service provider would likely
be characterized as a processor in respect of the communications content they
transmit on behalf of their users, but a controller in respect of subscriber data
and traffic data.97 However, as a processor, a service provider could only act on
the instructions of the controller, unless permitted by law,98 which would permit
a service provider to disclose communications content under a law enforcement
order, such as an interception warrant, or where an express compliance exemp-
tion is provided for.99 Given these constraints, service providers may want to be
93
See eg Rosemary Jay’s Data Protection Law & Practice (4th edn, Sweet & Maxwell, 2014), and Guide to the
General Data Protection Regulation (Sweet & Maxwell, 2017).
94
GDPR, at Art 4(7) and (8) respectively.
95
Recitals 10 and 32 of the PEC Directive reference the roles of controller and processor.
96
Such an outcome would require evidence of the required fault on behalf of the intermediary, which is
more likely to be that of intention rather than knowledge.
97
DPD, at recital 47. 98
DPD, at Art 16 and GDPR, at Art 28(3)(a).
99
Under the Data Protection Act 1998 (DPA), s 1(4), a person on whom an obligation to process data is im-
posed by law, becomes the data controller in respect of that data and processing purpose; while under s 29(3),
a data controller is exempt from its non-d isclosure obligations where it is for the purpose of the prevention or
detection of crime.
65
100
See Kuan Hon, W, Millard, C, and Walden, I, ‘Who is Responsible for “Personal Data” in Cloud
Computing?—The Cloud of Unknowing, Part 2’, (2012) 2(1) International Data Privacy Law, Vol. 3.
101
DPD, at Art 8(1) and GDPR, at Art 9(1). Referred to as ‘sensitive personal data’ under the DPA, s 2.
102
Note that this period may differ from any regulatory obligation to maintain billing records, see n 79 above.
103
Opinion 1/2003 ‘on the storage of traffic data for billing purposes’ (WP69), 29 January 2003.
6
104
DPD, at Art 2(h), GDPR, at Art 4(11).
105
See Case C-199/12, Josef Probst v mr.nexmet GmbH, 22 November 2012 [2013] CEC 913.
106
eg <https://dma.org.uk/article/eprivacy-lobbying-and-the-threat-to-b2b-marketing-and-telemarketing>.
107
ie Framework Directive, at Arts 20 and 21.
108
See A29 Working Party Opinion 13/2011 ‘on Geolocation services on smart mobile devices’ (WP 185),
at 4.2.1.
109
PEC Directive, Art 6(5) and PECR, reg 8(2).
110
PEC Directive, Art 9(3) and PECR, reg 14(5)(a)(ii).
111
A29WP, Opinion ‘on the use of location data with a view to providing value-added services’ (WP 115), 25
November 2005.
67
112
DPD, at Art 17, GDPR, at s 2.
113
See Directive 2002/21/EC on a common regulatory framework for electronic communications networks
and services (as amended in 2009) (Framework Directive), at Chapter IIIa ‘Security and Integrity of Networks
and Services’, Arts 13a and 13b. See Communications Act 2003, ss 105A–105D, also Ofcom guidance (8 August
2014) and update (18 December 2017).
114
See Directive 2016/1148/E U concerning measures for a high common level of security of network and
information systems across the Union, OJ L 194/1, 19 July 2016 (NIS Directive). The NIS Directive is applicable
to ‘operators of essential services’ and ‘digital service providers’ that are not public communications networks
or publicly available electronic communication services (Art 1(3)), although these terms include an internet
exchange point’, which interconnect IP networks, and a ‘cloud computing service’ (Annex II and III), which
are clearly integral to modern communication systems and services.
115
Regarding compliance, see ENISA report, ‘Analysis of security measures deployed by e-communication
providers’, December 2016.
116
ePrivacy proposal, at 1.2 and 3.5. 117
California Civil Code § 1798.29 and 1798.80 et seq.
68
118
Commission Regulation (EU) No. 611/2013, OJ L 173/2 , 26 June 2013 (‘Notification Regulation’), adopted
under Art 4(5) of the PEC Directive.
119
Ibid, at Art 2(2), which also states that detection is deemed when the provider has ‘sufficient awareness’.
120
EA/2016/0110, 30 August 2016.
121
Commission Regulation (EU) No. 611/2013, n 118, at Art 3(2), indicating the circumstances that should
be taken into account when assessing such likelihood.
122
Ibid, at Annex I and II.
123
eg SSE Energy Supply Limited (Retail Telecoms), <https://ico.org.uk/action-weve-t aken/audits-ad-
visory-v isits-a nd-overview-reports/>.
69
13.4.3 Transparency
Security breach notification is just one element of a broader raft of transparency
obligations that a controller has towards a data subject, to enable the latter to exer-
cise control over their personal data. Under the DPD, transparency was primarily
subsumed within the concept of ‘fair’ processing, but has become a distinct prin-
ciple under the GDPR.124 As noted in the previous section, service providers are re-
quired to inform subscribers of any residual risks to their personal data that would
not be addressed by security measures taken by the service provider (Article 4(2)).
The examples given are accessing ‘an open network such as the Internet or ana-
logue mobile telephony’ (recital 20); the latter reflects the state of technological
development at the time of the original measure. This transparency obligation is
the only security-related provision retained in the ePrivacy proposal.125
Under EU ‘Net Neutrality’ rules,126 providers of ‘internet access services’ have
an obligation to be transparent in any contract about how any traffic management
measures could impact the privacy of end-users and the protection of their per-
sonal data (Article 4(1)(a)). The Body of European Regulators, BEREC, has stated
that such information should include the types of personal data involved and the
measures taken to protect such data.127 These obligations are supplementary to the
general information requirements for controllers under the GDPR.128
13.4.4 Monitoring users
While Section 13.3 examined the User–State relationship in respect of the inter-
ception of communications, a service provider may also engage in conduct for
certain purposes that could amount to interception. In an age of behavioural on-
line advertising, message content may be reviewed in order to target adverts; for
cybersecurity reasons, to prevent spam or malware; or to enhance a system’s al-
gorithmic modelling of customer conduct. Of these purposes, the PEC Directive
only expressly recognizes the prevention of crime and the ‘unauthorised use of
the electronic communication system’ as legitimate reasons for interfering with
the confidentiality of a communication (Article 15(1)). Alternatively, the consent of
the communicating parties would legitimize any interception activities. However,
consent-based interception is problematic for a number of reasons: the standard
for obtaining a valid consent under data protection law is high, and raised under
the GDPR; obtaining consent from non-subscribing users is very difficult, and
consent can always be withdrawn.
European and UK law permit consent-based interception where both parties
have consented.129 The European Commission challenged UK law for being non-
compliant with EU law, because the original provision in RIPA enabled inter-
ception on the basis of ‘reasonable grounds for believing’ that consent had been
given.130 This was amended in 2011 and the concept of consent is now that under
data protection law. UK law has also long recognized that interception may occur
either incidental to, or in the course of, the provision of services.131 Under the IPA,
an operator is permitted to intercept a user’s communications for one or more of
the following purposes:
The scope of the first purpose seems particularly unclear since, subject to com-
pliance with general data protection laws, an operator may have strong commer-
cial incentives to tie the provision of a service to an act of interception, such as
monitoring communications for advertising purposes. Such uncertainty be-
comes greater as the definition of what constitutes a ‘telecommunication service’
has been extended to encompass certain OTT communications services, such as
Gmail, whose revenue model for the service is based on advertising (eg ‘cost per
click’) rather than end-user subscriptions.
In terms of preventing the viewing or publication of content, this clearly engages
an individual’s right to freedom of expression as much as their communications
privacy.132 While the IPA renders such conduct by operators lawful, questions re-
main whether it would be lawful under data protection law without either the con-
sent of the users of the service or some other legal authorization.133 An example of
129
See A v France (1994) 17 EHRR 462 and the IPA, s 44. By contrast, under US law interception is lawful if
only one party has given prior consent (18 USC § 2511(2)(c)).
130
Commission Press Release (IP/10/1215), ‘Commission refers UK to Court over privacy and data protec-
tion’ (30 September 2010).
131
eg Telecommunications Act 1984, s 45(1): ‘A person engaged in the running of a public telecommunica-
tion system who otherwise than in the course of his duty . . .’
132
In terms of ECHR, Art 8, see Golder v United Kingdom (1979) 1 EHRR 524: ‘Impeding someone from even
initiating correspondence constitutes the most far-reaching form of “interference” ’ (para 43).
133
See A29WP Opinion 2/2006 ‘on privacy issues related to the provision of email screening services’ (WP
118), 21 February 2006, at III(c).
671
the latter would be an injunction obtained by rights holders under the Copyright
Designs and Patent Act 1988, s 97A, to block or impede access by subscribers and
users to infringing content being streamed over the internet.134 However, the CJEU
has held that an internet filtering system would be unlawful were it to be applic-
able to all communications; applied indiscriminately to all customers; used as a
preventative measure and for an unlimited period.135
Under the ePrivacy proposal, access to communications content is presumed
to generate ‘high-r isks to the rights and freedoms of natural persons’ and there-
fore any such processing would require user consent and it must not be possible
to provide the service without such processing; or with the prior consultation with
the data protection supervisory authority, under a procedure detailed in the GDPR
(Article 36(2) and (3)).136 The former combines consent and necessity, which raises
the legitimacy threshold beyond that required for any other type of personal data,
including the ‘special categories’ under the GDPR. The latter is a quasi form of
prior authorization, which to date has only been required where state agencies are
involved. Were the current ePrivacy provisions on communications content to be
adopted, UK service providers will be constrained to a much greater degree than
under existing law, which is likely to reduce their ability to generate value from
their customers.
13.4.5 Directories
Under the New Regulatory Framework, the provision of directories of subscribers
is a part of universal service provision, as is a right for subscribers to be included in
such directories.137 From a data protection perspective, however, such directories
represent a public source of personal data about the subscriber, containing their
telephone number and home address, made available in either printed or elec-
tronic form or via a directory enquiry service. To reconcile these potentially di-
vergent interests, the PEC Directive requires that subscribers be fully informed
before being included in a directory, including any functionality that the direc-
tory may have, such as search functions, or transmission to a third party, such as
a competing provider of a directory enquiry service (recital 39 and Article 12(1)).
Subscribers should then be free to choose whether to be included in the directory
or not, which data relating to them is included, and to correct or withdraw such
134
eg FAPL v British Telecommunications & ors [2017] EWHC 480 (Ch).
Case C-70/10, Scarlet Extended SA v Societe Belge des Auteurs, Compositeurs et Editeurs SCRL (SABAM)
135
data, all at no charge (Article 12(2)). As with traffic and location data, this provision
treats subscriber data akin to sensitive data, legitimizing the processing solely on
the basis of consent.138
Of the four privacy relationships examined in this chapter, the most complex is
that between a subscriber to a communication service and a user of that service.
A subscriber is the party that has a contract with the service provider and, as al-
ready mentioned, may be a natural or a legal person.139 Payment may be made
under the contract, whether on a prepaid or post-paid basis, or there may be no
payment where the counter-performance by the user is being exposed to adver-
tising. While a user could also be a legal or natural person,140 for the purposes of
the PEC Directive, a user only means natural persons. The two most obvious ex-
amples of the Subscriber–User relationship are that between an employer and em-
ployee and between a parent and child. An employee may call friends from their
desktop during work hours to chat about the weekend or access the internet to
search for a holiday; while parents may provide their child with a mobile device
designed to track them when out and about.141 In what circumstances, however,
should a user be able to preserve the privacy of their communications activities
when using a service for which the subscriber has contracted and may be required
to pay for? Relying on the consent of a user to any monitoring can be even more
problematic than in other areas of communications privacy. In an employment
context, can consent ever be said to be ‘freely given’; while in a family context,
when does a child have the capacity to grant consent?142 The PEC Directive ad-
dresses two specific scenarios, one primarily applicable to employer–employee
relations; the other more relevant to that of parent–child.
138
The scope of such consent is examined in Case C-5 43/0 9, Deutsche Telekom AG v Bundesrepublik
Deutschland, 5 May 2011, at paras 54–67.
139
Framework Directive, at Art 2(k). 140
Ibid, at Art 2(h).
141
eg Family Locator: <https://itunes.apple.com/gb/app/family-locator-g ps-phone-t racker/id588364107?
mt=8>
142
Under the GDPR, consent is not freely given where there is a ‘clear imbalance’ between the data subject
and controller (recital 42); while children’s consent is subject to a special rule (Art 8).
673
143
RIPA, s 4(2), IPA, s 46. The Telecommunications (Lawful Business Practice) (Interception of
Communications) Regulations 2000, SI 2000/2699, and The Investigatory Powers (Interception by Businesses
etc. for Monitoring and Record-keeping Purposes) Regulations 2018, SI 2018/356.
144
RIPA, s 3(1), IPA, s 44.
145
See Working Document ‘on the surveillance of electronic communications in the workplace’ (WP 55, 29
May 2002) and Opinion 2/2017 ‘on data processing at work’ (WP 249, 8 June 2017). Also the ICO’s Employment
Practices Code, Part 3 ‘Monitoring at work’.
146
Judgment of the Grand Chamber, 5 September 2017 (Application no. 61496/0 8).
647
ECtHR held that the state had failed to adequately protect the individual’s Article
8 rights. Despite the fact that the employer had told the employee that he should
not use work computers for personal use, he had not been give prior notification
that enforcement of this restriction could result in the monitoring of his commu-
nications, especially the content of his messages. However, the Court went further
and noted that even if such notification had been part of the original restriction,
‘an employer’s instructions cannot reduce private social life in the workplace to
zero’ (para 80). This suggests that employers must provide employees with some di
minimis means of communicating for personal reasons during work hours.
13.5.2 Itemized bills
The second scenario concerns itemized billing. Where a subscriber receives an
itemized bill, it will generally detail the different categories of call made and their
associated cost. Indeed, such detail may be provided to evidence compliance with
a regulatory requirement to ensure that subscribers ‘can monitor and control ex-
penditure’, as is their right.147 As a consequence, subscribers may be given a record
of calls made and received by users of the service. To try and protect users from
such disclosure, the PEC Directive offers two potential solutions. First, the sub-
scriber shall have the right not to receive an itemized bill (Article 7(1)), which places
the onus on the subscriber to choose to refrain from receiving such data. Second,
Member States are made subject to a general obligation to implement measures
to ‘reconcile the rights of subscribers . . . with the right to privacy of calling users
and called subscribers’ (Article 7(2)). While it is left to the Member States to deter-
mine what measures to take, the provision calls for sufficient availability of ‘alter-
native privacy enhancing methods of communications or payment’, specifically
referencing the deletion of certain digits from a billed number and calling cards
(recital 33).
Under UK law, this obligation to take measures has been placed on Ofcom.148
Under the General Conditions of Entitlement, a service provider must there-
fore not include in an itemized bill calls made that are free of charge to the
subscriber, such as a helpline number; while the obligation to provide an
itemized bill does not apply where the subscriber is accessing the service on a
pre-paid-basis.149
147
Universal Services Directive, at Art 10(2) and Annex, Part A(a). See also the CJEU decision C-411/02,
Austria v Commission (2004) ECR I-8155, in which Austria was held to have failed in its obligation to ensure
itemized bills ‘sufficiently detailed to ensure effective verification and control by the consumer’.
148
PECR, at reg 9(2). 149
GCE, at Conditions 12.4 and 12.5 respectively.
657
13.5.3 Call-blocking
The provisions on lawful business practices and itemized bills are designed to pro-
tect the user’s privacy from the subscriber. The PEC Directive also contains a provi-
sion offering protection in reverse. Service providers are required to offer subscribers,
by a simple means and free of charge, the ability to prevent users automatically for-
warding calls to the subscriber’s terminal.150 Preventing such ‘nuisance’ calls (recital
37) can be viewed as a classic expression of privacy as the ‘right to be let alone’,151
which is elaborated further in the PEC Directive with respect to unsolicited commu-
nications in the User–User relationship (at Section 13.6.3 below).
The ePrivacy proposal removes any reference to the need to reconcile the inter-
ests of the subscriber and user, which implies that the Subscriber–User relation-
ship is considered to require no further regulatory intervention.
The final privacy relationship, between the communicating parties, has a more
limited direct impact on service providers, but has generated some of the most
high profile communication privacy issues in recent years, particularly in relation
to ‘cookies’ and ‘spam’. The ‘cookie’ rule has resulted in large numbers of web-
sites presenting banners and other mechanisms to site visitors about how the site
deploys cookies and requesting consent, while controlling unwanted, unsolicited
calls and emails remains a major concern for the general public. The latter issue
engages service providers because of the potential for network congestion.
13.6.1 Cookies
At the time the PEC Directive was being considered, at the beginning of the cen-
tury, the issue of ‘cookies’ had come to public attention as usage of the World Wide
Web was exploding. A cookie is a simple text file which a website can deposit, or
‘set’, on a site visitor’s web browser in order to improve the user experience, by
enabling the site to identify returning users through an exchange of data held on
the visitor’s device. There is a wide range of different cookies being used now, as
well as many other techniques that operate in different ways but have similar ob-
jectives, ie to identify, measure, and track the conduct of visitors interacting with
an online resource.152 The policy and regulatory concern with such techniques
150
Art 11 and PECR, reg 17.
151
Warren, S and Brandeis, L, ‘The Right to Privacy’, (1890) 4 Harvard Law Review 193.
152
Broadly, cookies are distinguished into session and persistent, first and third party.
67
were two-fold: their deployment was not transparent to users; and the placing
and accessing of data on the user’s device, which is itself part of a person’s pri-
vate sphere. Indeed, the original standard that defined such techniques was spe-
cifically designed to be hidden from the user for reasons of efficiency. Such lack of
transparency was seen as fundamentally unacceptable from a privacy and data
protection perspective,153 so the PEC Directive included a provision to address this
concern.
The original provision required that those that use electronic communication
networks ‘to store information or to gain access to information stored in the ter-
minal equipment of a subscriber or user’ must provide clear and comprehensible
information about the purposes for such processing and a right to refuse such pro-
cessing (Article 5(3)). The right to refuse was qualified to the extent that any such
processing was necessary to enable or facilitate the transmission of a communi-
cation or to provide an ‘information society service’ that had been explicitly re-
quested. However, as with all transparency obligations, websites could comply by
simply adding the required information to the details already included in privacy
policies, which are rarely read. Notification was therefore increasingly viewed
as insufficient to address the privacy concerns and, in 2009, the provision was
strengthened to require that consent be obtained. This then led to the appearance
of banners, pop-ups, and other creative solutions to give users an opportunity to
consent, whether by opting in or opting out.154 Ironically, the ICO initially chose to
adopt an explicit consent approach to cookies, but promptly experienced a huge
loss of analytics about how its own website was being used by the public, as most
visitors chose not to opt in! In 2013, the ICO eventually changed its stance and
adopted an opt-out, implied consent approach.155
The ePrivacy proposal now devotes a whole article to the issue. It adopts a default
position that collecting data from a user’s terminal equipment, even where the de-
vice itself emits the data, or using its processing capabilities is prohibited unless
an exception is applicable. Following the general scheme of the Data Protection
Directive and the GDPR, the justifications for processing can be broadly divided
into consent and certain specified necessities. In terms of consent, the ePrivacy
proposal makes the concession that its expression may be obtained ‘by using
the appropriate technical settings of a software application enabling access to
the internet’ (Article 9(2)), which would replace the need for some of the cookie
153
eg A29WP Recommendation 1/99 ‘on invisible and automatic processing of personal data on the internet
performed by software and hardware’ (WP17), 23 February 1999.
154
See <www.allaboutcookies.org> and <https://w ww.youronlinechoices.com>.
155
See <https://ico.org.uk/about-t he-ico/news-a nd-events/c urrent-topics/changes-to-cookies-on-our-
website/>.
76
banners and other mechanisms that are currently used. However, users will need
to be notified at six-monthly intervals of their right to withdraw their consent
(Article 9(3)). The necessity-based justifications include those already present, as
well as web audience measuring carried out by the provider of the requested infor-
mation society service (Article 8(1)(d)).
With regard to device emissions, the classic example of the potential for abuse
involved Google’s Street View service, when it was discovered that between 2008
and 2010 the mapping cars were also collecting payload data, such as passwords
and credit card details, sent over open WiFi networks.156 The ePrivacy proposal re-
fers to the emergence of new services that track a person’s movement based on
device scanning (recital 25). As a consequence, it states that the only justification
for collection is in order to establish a connection or where ‘a clear and prominent
notice is displayed’ (Article 8(2)). The latter requirement is similar to the approach
taken to the collection of images by CCTV cameras.157 In both cases, the unique
feature is that an individual is outside his home in a public space, where expect-
ations of privacy are accordingly reduced.158
156
See <https://epic.org/privacy/streetview/>.
ICO, ‘In the picture: A data protection code of practice for surveillance cameras and personal informa-
157
tion’, June 2017.
158
See Von Hannover v Germany (2005) 40 EHRR 1 and Von Hannover v Germany (2012) 55 EHRR 15.
159
PEC Directive, at Art 8(2), (3), and (4) respectively.
678
160
See Ofcom Guidelines for the provision of Calling Line Identification Facilities and other related services
over Electronic Communications Networks (v2), 26 April 2007.
161
PECR, reg 12; eg <http://bt.custhelp.com/app/a nswers/detail/a _ id/9952/~/how-do-i-w ithhold-my-
telephone-number%3F>.
162
PEC Directive, Art 10 and PECR, regs 15 and 16. See also Commission Recommendation (2003/558/EC)
on the processing of caller location information in electronic communication networks for the purpose of
location-enhanced emergency call services, OJ L 189/49, 29 July 2003.
163
Civil Contingencies Act 2004, s 1. 164
PECR, reg 16A.
679
165
eg the debates surrounding the WCIT, specifically in relation to Art 7 on ‘Unsolicited bulk electronic
communications’. See further Chapter 16, at Section 16.3.4.1.
166
On consent, see A29WP Working Document 2/2013 providing guidance on obtaining consent for cookies
(WP 208), 2 October 2013.
167
The A29WP suggests that ‘similar’ be judged from ‘the objective perspective (reasonable expectations)
of the recipient’. See Opinion 5/2004 ‘on unsolicited communications for marketing purposes’ (WP90), 27
February 2004, at 9. Under US law, the equivalent term is ‘established business relationship’ (47 USC § 277(a)
(2), defined at CFR §64.1200(f)(5)).
168
PECR, regs 19 and 21, as amended by SI 2016/524.
169
eg in May 2017, the ICO fined Keurboom Communications Ltd £400,000 for a serious breach of PECR,
reg 19. See <https://ico.org.uk/media/action-weve-t aken/enforcement-notices/2014013/mpn-keurboom-ltd-
20170503.pdf>.
170
PECR, reg 2(4), which includes ‘anything that performs the function of a line’.
680
subscriber to that ‘line’ over which the communication is made. Both may be held
liable, the subscriber for permitting his line to used,171 although the subscriber’s
liability is a form of accessory or secondary liability, where the fault element will
differ from that of the principal, generally requiring intent in respect of the aiding
and abetting and knowledge in respect of the ‘essential matters which constitute
the offence’.172
Similar rules in the US have given rise to a question concerning who is the
‘sender’ of an unsolicited communication. Yahoo! is subject to a class action for
sending unsolicited messages in breach of the Telephone Consumer Protection
Act.173 In this case, when a user sent a message to another person, the recipient
also received a ‘welcome’ message from Yahoo! advertising their services. In this
scenario, should the incorporated ‘welcome’ message be viewed as a distinct un-
solicited commercial communication, or is the service provider simply taking ad-
vantage of an unsolicited communication initiated by the user? In the UK, the ICO
has noted a distinction between the sender of a message and the ‘instigator’ of
a message, where companies ask message recipients to forward marketing mes-
sages to friends and family, so-called ‘viral marketing’.174 The instigator remains
subject to the PECR rules because they have ‘encouraged’ the sending of the
message. Applying this to the Yahoo! scenario, by offering users the facility to send
messages to persons not using the service, Yahoo! could be considered as having
instigated the message.
Under UK law, the PECR lays down distinct rules for each type of unsolicited
communication: automated calling systems, facsimile systems, voice call and
‘electronic mail’.175 One mechanism designed to support the regime is the estab-
lishment of national ‘registers’ to enable subscribers to opt out of receiving un-
solicited communications on a general basis. The ‘Telephone Preference Service’
and related services are operated by the Direct Marketing Association, under
contract to the ICO which has the statutory obligation to establish and maintain
registers.176
The ePrivacy proposal retains the provisions on unsolicited communications
(Article 16), as well as requiring providers of ‘publicly available number-based
interpersonal communication services’ to deploy state-of-the-a rt measures to
171
PECR, regs 19(3), 20(3), 21(2), and 22(4). 172
Johnson v Youden [1950] 1 KB 544, at 546.
173
47 USC § 227. See <http://w ww.reuters.com/a rticle/u s-y ahoo-classaction/yahoo-must-face-class-
action-over-text-messages-u-s-judge-idUSKBN0UI21R20160105>, also <http://w ww.yahootcpaclass.com>.
174
ICO, ‘Guide to the Privacy and Electronic Communications Regulations’, 20 May 2016, at 22.
175
PECR, regs 18–26.
176
PECR, regs 25–26. Responsibility for registers was transferred from Ofcom, by SI 2016/1177. See <https://
www.mpsonline.org.uk/t ps/i ndex.html> and <http://w ww.fpsonline.org.uk/f ps/>.
618
limit unwanted calls for users and to provide users with the facility to block calls
from specific numbers or anonymous sources (similar to the CLI provisions) and
prevent automatic call-forwarding (Article 14).
As this chapter has (hopefully) shown, communications privacy should not be seen
as simply a minor branch of privacy and data protection law. Each privacy relation-
ship illustrates different facets of both regimes in uniquely interesting ways. From a
privacy perspective, the User–State relationship and, to a lesser degree, the Service
Provider–Subscriber relationship concerns the conditions under which exceptions to
the individual right may operate. Any interference must be in accordance with law, for
a legitimate purpose and necessary and proportionate. By contrast, the Subscriber–
User and User–User relationships often involve a balancing exercise between the
individual rights of both parties, extending beyond privacy to include freedom of ex-
pression and other rights. The latter therefore tends to require a more complex and
nuanced legal response than the former.
Under data protection law, the perspective differs. For the controller, the con-
straints are similar to privacy with processing always needing to be legitimized
through necessity or with a data subject’s consent. For data subjects, the regime
enables the exercise of control over their personal data through consent, often im-
plemented through action, using ‘a simple means and free of charge’. Under the
general regime, the service provider as controller has to justify the legitimacy of the
processing activity. Under the sectoral regime, however, it can be argued that the re-
stricted focus on consent-based processing, combined with the data subject being a
controller in respect of communication content, shifts greater responsibility on to the
data subject.
Both privacy and data protection law treat different types of data differently
(content, traffic, location, and subscriber data), each according to the degree to
which such data is perceived to represent our private life; although data protection
law has narrowed such differences. While subscriber data seems clearly distin-
guishable, the case for greater parity of treatment between content and traffic data
has become stronger as the latter becomes ever more revealing.
Compliance with data protection laws is becoming an ever more important
issue for controllers and processors, especially with the substantially strength-
ened sanctions regime under the GDPR. For communication service providers, the
additional constraints under the PEC Directive have long been a source of griev-
ance, when compared with others in the Internet supply chain. If the ePrivacy pro-
posal proceeds in its current form, this additional burden will become applicable
628
See Chapter 9.
177
638
14
CONVERGENCE
14.1 INTRODUC TION
1
Directive 2010/13/E U on the coordination of certain provisions laid down by law, regulation or adminis-
trative action in Member States concerning the provision of audiovisual media services, [2010] OJ L 95/1. The
2010 restatement was a consolidation of the original 1989 Directive with its amendments in 1997 and 2007
(each cited in Section 14.2, below).
2
See for instance Barendt, E, Hitchens, L, Craufurd-Smith, R, and Bosland, J, Media Law: Text, Cases, and
Materials (Harlow: Pearson, 2014); the chapters by Ballard, T (‘Broadcasting’) and Woods, L (‘Regulation
684
and extra-legal regulation of the media sector’ in Goldberg, D, Sutter, G, and Walden, I (eds), Media law and
practice (Oxford: OUP, 2009); Keller, P, European and International Media Law (Oxford: OUP, 2011); Oster, J,
European and International Media Law (Cambridge: CUP, 2017).
3
Autronic AG v Switzerland (1990) 12 EHRR 485; Mustafa v Sweden (2008) 52 EHRR 803; Yildirim v Turkey
[2012] ECHR 2074.
658
the principle of broadcasting as a Treaty service,4 and a lengthy debate on the ap-
propriate form of regulation,5 the Television Without Frontiers (TVWF) Directive6
of 1989 represented the first legislative intervention.
The AVMSD was adopted in 2007 with a transposition date of 19 December 2009.7
Prior to its adoption, fundamental issues of content regulation (including in rela-
tion to material available on the internet) were discussed at length, and the Court
of Justice also made a key intervention through its decision in Mediakabel,8 on the
status of an emerging platform (near-v ideo-on-demand) and the relationship be-
tween Television Without Frontiers and the Electronic Commerce Directive.9 The
AVMSD made significant revisions to the original Directive, including the intro-
duction of two tiers of regulation, as discussed in this section. In 2010 the three
Directives were consolidated (without further amendment) as Directive 2010/13/
EU, and it is that consolidated version which is referred to in this chapter. In 2016,
the European Commission proposed further amendments to the Directive; as of
the publication of this edition, these proposals are still being considered by the
Union’s legislative bodies.
The many (and sometimes contradictory10) objectives of the EU’s intervention
are best explained in Recital 11, noting the need to ‘avoid distortions of competi-
tion, improve legal certainty, help complete the internal market and facilitate the
emergence of a single information area’.
The Directive applies to ‘audiovisual media services’, further defined as:
4
Italy v Sacchi [1974] 2 CMLR 177.
5
Ward, D, The European Union, Democratic Deficit, and the Public Sphere (Oxford: IOS Press, 2002), 55–57.
6
Council Directive 89/552/EC on the coordination of certain provisions laid down by law, regulation or
administrative action in Member States concerning the pursuit of television broadcasting activities [1989] OJ
L 298/2 3; amended by Directive 97/36/EC of the European Parliament and of the Council of 30 June 30 1997
[1997] OJ L 202/6 0.
7
Directive 2007/65/EC amending Council Directive 89/552/E EC on the coordination of certain provisions
laid down by law, regulation or administrative action in Member States concerning the pursuit of television
broadcasting activities, OJ L 332/27, 18 December 2007.
8
Mediakabel BV v Commissariaat voor de Media [2005] ECR I-4 891.
9
Directive 2000/31/EC on certain legal aspects of information society services, in particular electronic
commerce, in the Internal Market [2000] OJ L 178/1; see further Chapter 15.
10
On the tension between the ‘free flow’ and ‘cultural policy’ aspects of the Directive, see Oster, n 2, 145.
11
See Chapter 4.
68
This approach does have some substantive consequences for content regulation,
meaning that for example, many non-economic activities are outside the scope of
the AVMSD, as are services not making use of networks (eg DVD). Therefore, even
though a wide range of audiovisual services across platforms might meet some of
the criteria set out in the Directive, many are immediately excluded due to failure
to meet other tests; this was the subject of much discussion during the negotiation
of the Directive, with the Commission pointing out that it was not proposing to
regulate the internet.12
Television Without Frontiers dealt with ‘television broadcasting’ and did not
apply to what were termed ‘communication services’ (Article 1(a)). In the 1997
amendments, this definition was retained, although it was noted (at Recital 8) that
it was essential that ‘Member States should take action with regard to services
comparable to television broadcasting in order to prevent any breach of the fun-
damental principles which must govern information and the emergence of wide
disparities as regards free movement and competition’. But in the AVMSD, which
regulates ‘audiovisual media services’ divided into two mutually exclusive sub-
categories, services are no longer defined in purely technical terms of reception
and broadcasting. Instead, a television service (the first category) is defined as an
‘audiovisual media service provided by a media service provider for simultan-
eous viewing of programmes on the basis of a programme schedule’ (Article 1(1)
(e)). Although the language of ‘linear’ is not found, the concept of simultaneous
viewing on the basis of a schedule captures its essential aspects. In turn, an on-
demand (non-linear) service (the second category) is an ‘audiovisual media ser-
vice provided by a media service provider for the viewing of programmes at the
moment chosen by the user and at his individual request on the basis of a cata-
logue of programmes selected by the media service provider’. The media ser-
vice provider is the person who has ‘editorial responsibility for the choice of the
audiovisual content of the audiovisual media service and determines the manner
in which it is organised’.
A programme is defined as moving images (with or without sound) which
constitute ‘an individual item within a schedule or a catalogue established
by a media service provider and the form and content of which are compar-
able to the form and content of television broadcasting’. The Directive also sets
out what it describes as examples of programmes: ‘feature-length films, sports
12
Ballard, M, ‘EU regulation attacked as censorship’ The Register 19 May 2006. <http://w ww.theregister.
co.uk/2006/05/19/eu_censorship/>; Valcke, P and Stevens, D, ‘Graduated regulation of “regulatable” content
and the European Audiovisual Media Services Directive: one small step for the industry and one giant leap for
the regulator?’ (2007) 24 Telematics & Informatics 285, 295.
678
13
In particular, see the report prepared for Ofcom: Marsden, C, et al, ‘Assessing Indirect Impacts of the EC
Proposals for Video Regulation’ (31 August 2006) <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 021/
32088/v ideoregulation.pdf>.
14
Council of the European Union, ‘General approach on a proposal for a Directive of the European
Parliament and of the Council amending Council Directive 89/552/E EC’ (13 November 2006, document
15277/0 6) <http://register.consilium.europa.eu/pdf/en/0 6/st15/st15277.en06.pdf>.
68
and magazines (Recital 28), and services ‘where any audiovisual content is merely
incidental to the service and not its principal purpose’ (Recital 22).
14.2.2 Beyond the EU
The Directives are applicable within the European Economic Area (EEA), and so
bring Norway, Iceland, and Liechtenstein within the AVMS system. (Indeed, as
with EU states, a large number of broadcasters addressing audiences in Norway
are established in the UK, including the major Norwegian commercial channels,
and Norwegian-language services of global broadcasting companies).
The original TVWF Directive was adopted alongside a similar but not iden-
tical Council of Europe instrument: the European Convention on Transfrontier
Television. (A ‘disconnection clause’ in the ECTT ensured that EU states would
apply the Directive between them, but the ECTT where a non-EU state was con-
cerned.) Many (but not all) EU states have also ratified the Convention. The ECTT
was updated in parallel with the 1997 reforms to TVWF, but an attempt to make
further changes in response to the AVMSD foundered on the EU’s objections to
Member States ratifying international agreements in areas of the EU’s exclusive
external competences.15 Nonetheless, the ECTT (effectively the pre-AVMS EU law
on television services) continues to apply, with particular reference to relations
with the non-EU states that have ratified it (notably Turkey and Ukraine), and per-
haps in due course the United Kingdom when it leaves the European Union. (The
impact of Brexit on broadcasting regulation is discussed in Section 14.2.5.)
15
Mac Síthigh, D, ‘Death of a Convention: Competition between the Council of Europe and European Union
in the Regulation of Broadcasting’, (2013) 5 Journal of Media Law 133.
16
See further Chapter 15. 17
See further Chapter 4.
698
parameters of the Directive. For example, Swedish law provides for further restric-
tions on advertising directed at children, and French law requires further use of
European content than that required by the Directive.18 This is generally accept-
able, but such rules can only apply to those services subject to the jurisdiction
of that state (as set out in Article 4), and the state must also be aware that if an
audiovisual media service provider is dissatisfied with a non-AVMSD provision, it
can relocate to another Member State while still continuing to be received by audi-
ences within the original Member State.
In limited circumstances, states can place restrictions on services regulated in
other Member States. The procedures differ as between television and on-demand
services. In the case of television, Article 3(2) allows derogation from the freedom
of reception principle where the provisions on minors or incitement to hatred are
‘manifestly, seriously and gravely infringed’ on at least three occasions in a year
and subject to a consultation process. The procedure for proscription in the UK in-
volves a notification by Ofcom and a decision by the Secretary of State, and is set out
in the Broadcasting Act 1990, ss 177–178 (for satellite)19 and the Communications
Act 2003, ss 329–332 (for cable, DTT and internet services);20 the power has been
exercised on six occasions21 and also considered and confirmed by the domestic
courts.22
The scope for derogation is wider in respect of on-demand services, permitted
by Article 3(4) where a necessary and proportionate response to the prejudice to
(or ‘serious and grave risk’ thereof) public policy, public health, public security or
the protection of consumers. A notification procedure applies, although it can be
partially bypassed in urgent cases. For television only, Article 4 sets out two pos-
sible courses of action where a broadcaster under the jurisdiction of one state is
‘wholly or mostly directed’ at the territory of another, where the latter (recipient)
state has adopted ‘rules of general public interest’ beyond the Directive. The first
18
Garzaniti, L and O’Regan, M, Telecommunications, broadcasting and the Internet: EU competition law
and regulation (3rd edn, London: Thomson Reuters, 2010) 283.
19
Where the quality of the service is unacceptable (further defined as where the service repeatedly con-
tains ‘matter which offends against good taste or decency or is likely to encourage or incite to crime or to lead
to disorder or to be offensive to public feeling’) and such an order is compatible with the UK’s international
obligations and in the public interest. Enforcement includes a range of criminal offences (supplying decoder
equipment, advertising on the service, etc).
20
In broadly similar terms to the provision regarding satellite, but also including the prohibition of the in-
clusion of the proscribed service in a cable service or multiplex.
21
The statutory instruments are 1993/1024 (Red Hot Television), 1995/2917 (XXX TV Erotica), 1996/2557
(Rendez-Vouz), 1997/1150 (Satisfaction Club Television), 1998/1865 (Eurotica), 1998/3083 (Eros TV), and 2005/
220 (Extasi TV).
22
R v Secretary of State for National Heritage ex parte Continental Television [1993] 2 CMLR 333 (refusal
to prevent SI by injunction, referral to ECJ); R v Secretary of State for Culture, Media & Sport ex parte Danish
Satellite Television & Rendez-Vouz Television International [1999] 3 CMLR 919.
960
option is for the recipient state to contact the originating state, with the originating
state having the obligation to request that the broadcaster comply with the rules of
the originating state. The advice of the Contact Committee set up by the Directive
can be sought. Should the results not be satisfactory (to the recipient state), and
the broadcaster is established in the other state ‘in order to circumvent the stricter
rules’, the recipient state can take action against the broadcaster, subject to limita-
tions23 and the subsequent ratification of the action by the Commission. This pro-
vision builds on Court of Justice decisions regarding conflicts of this nature.24
Moreover, services originating in other states can also be the subject of regu-
lation through the general law of a state. However, this must not, as the CJEU has
said, amount to ‘secondary control of television broadcasts’ already regulated in
the country of origin or the preventing of retransmission of such services per se.25
23
Article 4(3): ‘Such measures shall be objectively necessary, applied in a non-d iscriminatory manner and
proportionate to the objectives which they pursue’.
24
See, for example, TV10 SA v Commissariaat Voor de Media [1995] 3 CMLR 284.
25
Case C-2 44/10, METV and Roj TV v Germany [2012] 1 CMLR 32, on national law on the regulation of as-
sociations and the principles of international understanding, confirming an earlier decision in the context of
consumer protection Konsumentombudsmannen v De Agostini C-3 4/95 [1998] 1 CMLR 32.
26
See for example Commission v UK C-222/94 (use of different criteria for jurisdiction); VT4 C-56/9 6 (broad-
casters established in more than one state); see further Wagner, M, ‘Revisiting the Country-of-Origin Principle
in the AVMS Directive’, (2014) 6 Journal of Media Law 286.
691
27
Garzaniti and O’Regan, n 18, 280, Walden, I, ‘Illegal content’ in Goldberg, D, Sutter, G and Walden, I (eds),
Media law and practice (Oxford: OUP, 2009) 452; DCMS, ‘Audiovisual Media Services Directive: consultation
on proposals for implementation in the UK’ (on file with author) 65. See subsequently the Wireless Telegraphy
Act 2006, s 9A and Ofcom, ‘Notice of Proposed Changes to Satellite Services licences resulting from the AVMS
Directive’ 25 January, 2010. <http://stakeholders.ofcom.org.uk/binaries/consultations/satellite_ services/
summary/condoc.pdf> on changes to the licensing regime for satellite earth stations.
629
multiple EU/EEA states, and even for a significant proportion of services addressed
to a single state or language-based audience. As of 2016, the 1222 television services
licensed by Ofcom28 can be classified as 337 targeting the UK (including forty-two
for the UK and Ireland29), 138 operating on a pan-European basis including the UK,
110 targeting non-EU/EEA states, and, over half of the total, 637 targeting other EU/
EEA states (483 at single states and 154 at two or more, normally on a linguistic basis
(eg France and French-speaking Belgium; Greece and Cyprus)). Conversely, only
a handful of the services available in the UK are established elsewhere in the EU
(around twenty-five), and many of those are pan-European services (eg Euronews) or
public services (eg RTÉ in Ireland, France24 in France). If appropriate provision is not
28
Figures (dated October 2016) derived from the MAVISE database held by the European Audiovisual
Observatory and further classified by the author.
29
In practice a greater share than 42 of the 337 are addressed to the UK and Ireland, given the similar Sky
platform marketed in both states; the low number may be a feature of how the data is reported. Note that some
services are also identified as addressed to Ireland (ie part of the 438 listed here as ‘single state’) although are
very similar to UK-addressed services—most likely in order to carry separate advertising.
639
made within the Brexit settlement, UK-established services will no longer fall within
the country of origin rules and so can be subject to re-regulation by each and every
EU state; this may have a particular impact on pan-European services (whether from
US or European companies), and on collections of services targeting single states all
established in the same London facilities, because being established within the EU
has long been an important facilitator of access to the European audiovisual market.
In some states, commercial broadcasters operating in the language of that state are
predominantly UK-established; few other areas of the single market have seen such
a concentration of establishment in a single Member State. The UK government has
declared its attention to promote free trade, specifying its desire to ‘(support) the
continued growth of the UK’s broadcasting sector’.30 Given the paucity of provisions
on broadcasting in international trade agreements (let alone the stronger country of
origin provisions in the Directive), this will present a particular challenge.
14.3 UK L AW A ND L IC ENSING
‘The United Kingdom’s exiting from and new partnership with the European Union’ (Cm 9417, 2017) 35.
30
Commission v Belgium C-11/95 [34]: ‘it is solely for the Member State from which television broadcasts
31
emanate to monitor the application of the law of the originating Member State applying to such broadcasts
and to ensure compliance with Directive 89/552, and [ . . . ] the receiving Member State is not authorized to ex-
ercise its own control in that regard’. The overwhelming majority of channels available in the UK are licensed
by Ofcom.
964
punishable by a fine but proceedings must have the consent of the Director of
Public Prosecutions. Occasional challenges to licence refusals are heard in the
courts.32
A digital terrestrial television (DTT) service will be licensed as a DTPS or a tele-
vision broadcasting service, and be carried on a licensed multiplex. A DATS is in
practice a data service delivered via DTT. In practice, these are the services col-
lectively branded and presented to consumers as ‘Freeview’.
Cable, satellite, and internet services are typically licensed as TLCS. It is how-
ever the provider of each service (ie channel) who is accountable to Ofcom, rather
than the cable or satellite provider (eg Virgin Media or Sky) to which end users
subscribe. Local delivery services (delivering radio/T V services by cable) were for-
merly the subject of licensing (under the Broadcasting Act 1990, Part 2) but this is
no longer the case (abolished by the Communications Act 2003, s 213).
A television broadcasting service entails the provision of programmes with a
view to being broadcast (ie by wireless telegraphy), so as to be available for re-
ception by the public. Multiplexes, TLCS, DTPS, and restricted services are ex-
cluded from this definition. In essence, these services are the established non-BBC
public service channels: the ITV licences, Channel 4, Channel 5, and S4C Digital
(qualifying services in the Broadcasting Act 1996, s 2, excluded from definition of
DTPS in the Broadcasting Act 1996, s 1).33
Certain services will also be the subject of Wireless Telegraphy Act licences. For
example, a multiplex will require a spectrum licence under Section 8 of the Act,
and Ofcom has a duty to secure (so far as practicable) sufficient capacity for the
‘qualifying services’ (the television broadcasting services referred to above) under
Section 7.
The provider of any of these services is the person with ‘general control’ over the
programmes, services, and facilities contained in the service, even though others
may have control the making of a particular programme or over the distribution of
the service over an ‘electronic communications network’ (Communications Act, s
362). Ofcom revoked the licence of Press TV in January 2012 on the grounds that
the body holding the licence did not have general control over the service.34
32
See for instance Re Blast 106 [2015] NICA 16 (refusal to extend a community radio licence quashed (and an
extension ordered) as a fair hearing was not granted).
33
The former analogue services also fell within the definition of television broadcasting services, until
digital switchover was completed in 2012.
34
<http://stakeholders.ofcom.org.uk/binaries/enforcement/broadcast-l icence-conditions/press-t v-revo-
cation.pdf>. Ofcom had already made a number of adverse findings against Press TV on other grounds; see
Johnson, H, ‘Regulation of “Foreign Broadcasters”—Standards of Fairness and Impartiality Unobtainable
Objectives in Reality?’, (2012) 17 Communications Law 25. The service is still available through various means
(including satellite and Internet), although no longer holds a licence from any EU Member State.
695
37
<ht t p:// s ta keholders.ofcom.org.u k/ b inaries/ e nforcement/ c ontent- s anct ions- a djudicat ions/
bangmedia-revocation.pdf>.
38
See discussion in Section 14.4.2, below.
96
14.3.1.2 TLCS
The Communications Act 2003 defines a TLCS as a service that is to be made avail-
able for reception by members of the public, consisting of television programmes
or an electronic programme guide (EPG), with a view to public reception by way of
satellite broadcasting, through a radio multiplex service, or distribution through
an electronic communications network (s 232). The other types of licensed service
are excluded from the definition. However, this is by far the biggest category of
television licence in the UK, with close to one thousand licences currently issued;40
Ofcom describes it in simple terms on its website as a licence ‘for linear televi-
sion channels provided on cable, satellite and the internet.’ The Broadcasting Act
1990 system of dividing satellite services into two regulatory categories (domestic
and non-domestic) was abolished in 1997,41 following a decision of the Court of
Justice.42
A licence is issued with little formality, requires the payment of an application
fee (£2500) and follows a detailed template. The standard licence consists of 29
conditions, based closely on the Broadcasting Acts and Communications Act, and
dealing with issues such as standards, the payment of fees, revocation, and sur-
render. Licences can only be refused on specified grounds: two are carried for-
ward from the Broadcasting Act 1990 (fit and proper person, restrictions on licence
holder) (s 3) and one is found in the Communications Act 2003 (that the service
would be likely to contravene the Broadcasting Code regarding standards or fair-
ness) (s 235). There is no requirement to provide evidence of carriage (as compared
with DTPS where an agreement between the service provider and a multiplex is a
prerequisite) and the licence does not require renewal, as long as an annual fee is
paid.43 A service available in identical form on more than one platform (eg on cable
and satellite) requires only one licence.
39
<https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 028/8 8219/Guidance-for-religious-bodies.pdf>.
40
List available at <https://w ww.ofcom.org.uk/manage-your-l icence/t v-broadcast-l icences/c urrent-
licensees> (frequently updated).
41
Satellite Television Service Regulations 1997, SI 1997/1682. 42
Commission v UK C-222/94.
43
A percentage of turnover based on bands (eg 0.0125% of the first £10m), with a minimum fee of £1000.
697
Following the implementation of the AVMSD in the UK a TLCS may now be re-
quired (if not already held for the service) for qualifying services transmitted via a
live stream on the internet. The provision of the Communications Act 2003 which
excluded services delivered via the internet has been repealed, and therefore a ser-
vice accessed via the internet is potentially licensable. However, two-way services
remain outside of the definition. A two-way service is provided through an ECN
and entails the transmission of images and/or sound by the provider to users and
by the users to the provider or other users (s 232(5)).
There are a series of services still excluded from the definition of TLCS. The first
group is essentially services that are regulated under another part of the broad-
casting legislation, eg if it is provided with a view to being broadcast as part of a
multiplex, or is a service already licensed as a television broadcasting service (ss
233(1) and 233(2)). The second group excluded entirely a number of different types
of service: two-way services (ss 232(5) and 233(4), ‘closed circuit’ services (on a set of
premises under a single occupier, not connected to an external network) (ss 233(5)
and 233(6)), and services for users with a ‘business interest’ in the programmes
(s 233(7)–(10)). Section 234 provides for the modification of ss 233–4 by order (ap-
proved by resolution of both houses of Parliament), subject to at least one of five
criteria being satisfied. These criteria include technological developments and the
practicability of applying different levels of regulation. The Secretary of State can
also by similar order exclude certain types of service from the definition of TLCS.
The ease with which multiple ‘feeds’ can be offered by a single broadcaster has
prompted specific guidance to be issued by Ofcom.44 It now states that in gen-
eral terms, multiple feeds will require multiple licences unless the programmes
(including advertisements) can be seen on all feeds at the same time, or the
services are ‘almost identical’ with ‘very occasional’ variations, or it is purely a
‘+ 1 service’ (ie exactly the same feed but broadcast with a one-hour delay) or a
dubbed/subtitled version. This would mean, for example, that a service available
on two different satellite services would only require one licence, but a service
which had different advertisements for different audiences (eg the use of two fre-
quencies) would require multiple licences.
TLCS licences are available in respect of three types of service: editorial, tele-
shopping, or self-promotional. Editorial services are subject to the general provi-
sions regarding commercial communications. Teleshopping (the supply of direct
offers to the public with a view to the supply of goods and services in return for pay-
ment, including transactional gambling, adult chat or adult sex chat, and psychic
44
Ofcom, ‘Guidance regarding the licensing position of television licensable content services broadcast
into multiple territories’, 19 October 2010 <http://l icensing.ofcom.org.uk/binaries/t v/l icensing-position.
pdf>.
968
45
See generally Starks, M, Switching to digital television: UK public policy and the market (Bristol: Intellect,
2007) 39–45.
96
Ltd (B) and Crown Castle (now Arqiva) (C/D). Following switchover and consoli-
dation, Channel 5 and S4C are carried on Multiplex 2, while Multiplex B remains
controlled by the BBC and contains HD channels only;46 further multiplexes for
HD services have been licensed to Arqiva, and to the local TV multiplex operator
(discussed below). An additional multiplex licensed in Northern Ireland carries
channels from the Republic of Ireland.
There are various obligations regarding the services included in a multiplex and
also certain reservations of capacity. Licences include a requirement to provide
the service, and conditions such as a prohibition on conveying any channel not
licensed either as DTPS/DTAS or by another EEA state. A licence including pro-
vision of the ‘core proposals’ of the licensee which were set out during the appli-
cation process and are varied from time to time. Redacted versions of the licences
(and the many amendments to them) are available.47
Digital UK is owned by all the multiplex licensees and manages the DTT plat-
form as a whole. It has a number of roles, in particular the allocation of logical
channel numbers (LCNs) to appropriately licensed/authorized DTT services, in
accordance with a published policy. Features of the policy include genre grouping
and appropriate prominence for public service channels (influenced by Ofcom’s
EPG code; see Section 14.3.3, below).
Analogue TV broadcasting was ‘switched off’ in the UK in 2012. The process
of switchover was first announced in 1994, legislated for in the Broadcasting Act
1996, and launched in 1998. As analogue transmissions ceased, additional DTT
transmissions were added and audiences were encouraged to move to a digital
platform (whether DTT or other), in a process managed by Digital UK.
48
Communications Act 2003, s 264. This review is itself an important part of broadcasting policy in the UK.
Crucially it requires Ofcom to consider all the relevant services together (in assessing the provision of public
service broadcasting) rather than as standalone services. See further Petley, J, ‘The re-regulation of broad-
casting, or the mill owners’ triumph’ (2002) 3 Journal of Media Law and Practice 131, 134–5.
07
Ofcom and others sometimes refer to the general regulation of television pro-
gramme services as tier 1 and to the PSB-specific provisions as tier 2 (‘quotas’ ie
requirements for programmes meeting various criteria to be broadcast) and tier
3 (other).51
The Communications Act 2003 also refers in various Sections to the ‘licensed
public service channels’, which are ITV1 (as defined above), Channel 4 and
Channel 5, but not the BBC services or S4C.
The regulation of these channels is through licensing, initially through the
Broadcasting Acts and further provided for in the Communications Act 2003, s 215
(Channel 3, Channel 5) and s 231 (Channel 4).
The tier 2 quotas apply to the licensed public service channels. These channels
are subject to requirements regarding news, independent productions, regional
programming, programming made outside the M25, original productions, services
for disabled audiences (beyond the general provisions applicable to all services),
and the relationship between the broadcaster and independent producers.
The tier 3 requirements are more general in nature, with a public service remit
set out in the Communications Act 2003, s 265 in respect of the licensed public
service channels (‘the provision of a range of high quality and diverse program-
ming’ for ITV1 and Channel 5 and a more detailed statement52 for Channel 4).
These channels are required to prepare an annual statement of programme policy,
having regard to Ofcom guidance and reports; significant changes must be ap-
proved by Ofcom. Ofcom has a very limited power to direct a provider to amend
49
Commonly referred to as ITV1, but still a group of regional franchises, all of which except STV (two fran-
chises, Scottish and Grampian) are controlled by ITV plc. Included here too is ITV Breakfast, the separate
franchise for mornings (formerly GMTV) bought in full by ITV plc in 2009.
50
The licence was revoked in 2010 after the service provider stopped providing it and it appears very un-
likely that a new licence will be issued. See Ofcom, ‘Report to the Secretary of State on the public teletext ser-
vice’ 1 December 2010 <http://stakeholders.ofcom.org.uk/binaries/broadcast/t v-ops/public-teletext-report.
pdf>.
51
See for example Katsirea, I, Public broadcasting and European law (The Hague: Kluwer, 2008) 129.
52
(a) demonstrates innovation, experiment and creativity in the form and content of programmes; (b) ap-
peals to the tastes and interests of a culturally diverse society; (c) makes a significant contribution to meeting
the need for the licensed public service channels to include programmes of an educational nature and other
programmes of educative value; and (d) exhibits a distinctive character.
701
its remit or to step in and regulate through specific conditions, in the case of a ser-
ious failure (not excused by economic or market conditions) to fulfil the channel’s
remit or to make an adequate contribution towards the purposes of PSB.
The Digital Economy Act 2010 inserted a new Section 198A in the Communi
cations Act 2003, of particular interest from a convergence point of view. It pro-
vides for Channel 4 Corporation functions in respect of the making of high quality
media content (including on-demand and internet services) and ‘films intended
to be shown to the general public at the cinema in the UK’. More broadly, Ofcom’s
duty to report on public service broadcasting has been supplemented53 by a fur-
ther duty to report on the contribution of material included in ‘media services’ to
the fulfilment of the Act’s public service objectives, with media services defined
in the broadest of terms as meaning publicly available services (where there is
editorial control) provided via the internet, as well as television, on-demand, and
radio services.
The BBC is the subject of a Charter and Framework Agreement.54 Since April
2017, Ofcom is responsible for the regulation of the BBC in accordance with these
instruments; prior to this date, Ofcom’s role was limited to certain aspects of con-
tent regulation, with the remaining functions being vested in the (now-d issolved)
BBC Trust.
14.3.1.6 ITV
The regulation of Channel 3 licences (ie ITV) was once a major part of UK broad-
casting law.55 However, the last franchise auction took place in 1991, in the wake
of the Broadcasting Act 1990. That process was primarily financial but with the
ability to award the licence to a lower bid for quality reasons in exceptional cir-
cumstances. These original licences were renewed after their first ten-year period
in 2001, and were then succeeded by ‘digital replacement licences’ without a new
franchise round in 2004, and current licences were renewed (with a change to the
boundaries of licence areas in Wales and western/south-western England) for a
further ten years, on revised financial terms, from 1 January 2015.56 The business
New Communications Act 2003, s 264A as inserted by the Digital Economy Act 2010, s 2.
53
55
For background, see Munro, C, Television, Censorship and the Law (Farnborough: Saxon House,
1979), ch 3.
56
The ‘initial expiry date’ was set at 31 December 2014 by the Communications Act 2003, s 224. Ofcom was
required to report on whether the services would be sustainable and could meet public service obligations (s
229), so that the Secretary of State could choose from a number of courses of action. The Secretary of State’s
decision was to allow renewal to proceed; all licencees applied for renewal and the decisions to renew were
announced by Ofcom on 20 February 2014. See <https://w ww.gov.uk/government/uploads/s ystem/uploads/
attachment_data/fi le/77982/Maria_M iller_letter_to_ed_r ichards.pdf> and <https://w ww.ofcom.org.uk/
about-ofcom/latest/media/media-releases/2014/ofcom-renews-itv-stv-utv-a nd-channel-5 -l icences>.
702
structure of ITV has also undergone significant change since 1991, with all li-
cences except those in Scotland now controlled by ITV plc after a series of mergers
and operating, in practice, as a single enterprise; a series of reductions in regional
programming has been approved. Ofcom has a further role in reviewing and ap-
proving the ITV networking arrangements under the Communications Act 2003,
ss 290–294. In practice, these arrangements are now handled by ITV plc under
agreed ‘network affiliate arrangements’ which have been accepted by Ofcom.57
14.3.1.7 Local TV
After years of review and discussion, a new system for licensing local television
services came into force in 2012.58 Two new forms of licence: Local Digital Television
Programme Service (L-DTPS) and Local Multiplex Service Licence were created.59
L-DTPS are the subject of ‘beauty contest’ licensing in selected geographical areas
(which have been identified by DCMS and Ofcom). Some content restrictions are
in place (eg no adult material or PRS chat services, with new definitions of both),
and local commitments form part of the licensing process, but licences do not con-
tain quotas (eg for the proportion of local programmes broadcast). L-DTPS and
simulcasts of such are included as public service channels for the purpose of EPG
prominence.60 Thirty-one services are currently licensed (although a majority fall
within one of two quasi-networks—‘Made in . . .’ and ‘That’s . . .’ followed by the
area name). In the first years of the new services, applications to vary licence com-
mitments (inevitably to reduce such commitments) have been frequent.61 A single
new DTT multiplex is licensed by Ofcom (the first and current holder is Comux),
with a requirement to carry the L-DTPS but with space for other services using
‘interleaved’ spectrum62 (the subject of a further WTA licence using newly reserved
spectrum).63 Many of these services are also available by way of VOD and IPTV.
57
This system has been in place since 2012; the most recent review took place in December 2016. See
<https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 018/95103/C hannel-3-networking-a rrangements-2016.
pdf>.
58
See generally Moore, M, ‘Plurality and local media’ in Barnett, S and Townend, J (eds), Media Power and
Plurality: From Hyperlocal to High-level Policy (Palgrave Macmillan, 2015).
59
Communications Act 2003, s 244 contains general powers. See The Local Digital Television Programme
Services Order 2012, SI 2012/292.
60
SI 2011/3003.
61
For instance, the London licence holder (London Live/ESTV) has made six requests for amendment since
its January 2014 award—i ncluding one which was made before broadcasts began, and two (one refused) which
required a more extensive consultation process by Ofcom on the grounds that it would be a departure from
the character of the service.
62
That is, various parts of the broadcast spectrum which are not currently used; the exact frequencies will
vary quote widely from transmitter to transmitter. The services currently carried through this system are True
Crime, Tiny Pop, and True Movies/Kix.
63
Wireless Telegraphy Act 2006 (Directions to OFCOM) Order 2012, SI 2012/293.
703
64
See discussion in Mac Síthigh, D, ‘Principles for a Second Century of Film Legislation’, (2014) 34 Legal
Studies 609, 614.
65
For detailed discussion of independent radio in the UK, see Stoller, T, Sounds of your Life: The History
of Independent Radio in the UK (New Barnet: John Libbey, 2010); Starkey, G, Local Radio, Going Global
(London: Palgrave Macmillan, 2011).
66
Broadcasting Act 1990, s 105: ability of the application to provide the proposed service, extent to which
it caters for the tastes and interests of residents in the area, broadening the range of programmes available/
catering for different tastes and interests, and evidence of demand or support for the service. See also Starkey
n 65, 114–119.
074
67
Significant changes to the regulation of community radio were brought about through the Community
Radio (Amendment) Order 2015, SI 2015/1000, increasing the class of stations permitted to carry advertising,
providing for additional renewals of licences, and supporting cross-ownership with local TV services.
68
Ofcom, ‘Radio in digital Britain’, 2009, <http://webarchive.nationalarchives.gov.uk/20100304014526/
http://ofcom.org.uk/radio/i fi/radio_d igitalbritain/d igitalbrit.pdf> [7.4].
69
Starkey n 65, 145–158; Guy Starkey, ‘Cultural Policy in the Coalition Years: Laissez-faire Regulation, the
Public Spending Squeeze and the Drive to Digital’, (2015) 24 Cultural Trends 80, 81.
70
<https://w ww.gov.uk/government/uploads/s ystem/uploads/attachment_data/fi le/591508/R adioDereg-
Final13Feb.pdf>.
71
For instance, one test is that the national digital share of all listening (including DAB and other digital
platforms eg through DTT) exceeds 50 per cent. As of 2016, the figure is 45.5 per cent. See further Ofcom,
‘The communications market: digital radio report’ (2016) <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/
0032/94838/The-Communications-Market-Digital-R adio-Report-2016.pdf>.
705
A multiplex licence includes an initial list of the services carried, which can be
varied; Ofcom can only refuse if ‘the capacity of the digital sound programme services
broadcast under the licence to appeal to a variety of tastes and interests would be
unacceptably diminished’.72 Since 2010, analogue licences due to expire have been
renewed, under new powers given to Ofcom on two occasions, so as to provide for
‘stability’ in the long transition to digital radio.73 Local DAB multiplexes can be re-
newed up to 2030 subject to requirements on maintaining coverage (in order to sup-
port the digital transition).74 ‘Small-scale’ DAB (serving smaller geographic areas),
originally the subject of a trial, has been brought into the legislative system for radio
through the Broadcasting (Radio Multiplex Services) Act 2017.
14.3.3 Infrastructure
14.3.3.1 Must-carry and must-offer
The availability of a service on a given platform is normally a matter for negotiation
between service provider and distributor, but in some EU states also by ‘must carry’
or ‘must offer’ rules. A ‘must carry’ provision requires that the service be carried on
a specified electronic communication network. Article 31 of the Universal Service
Directive allows (subject to conditions) the imposition of must-carry duties regarding
specified services on providers of ECNs ‘where a significant number of end-users
of such networks use them as their principal means to receive radio and television
broadcasts’. This is implemented in the UK through the Communications Act 2003,
s 64 and General Condition 7 for communications licences. The Court of Justice has
held that the criteria for designation must be precise regarding the applicable criteria
and the services and networks in question.75 Beyond this, there is no general right of
access for a content service to an ECS or ECN.76
Must-carry provisions are not currently in force in the UK. Early iterations of this
approach in the UK included a requirement for certain cable services to carry the
public service channels, although this was not possible between 1990 and 1996.77
Broadcasters also argue that because cable services were not required to pay copy-
right fees for retransmission of a television service,78 such a provision would not
72
Broadcasting Act 1990, s 54(6A).
73
Digital Economy Act 2010, s 32; Legislative Reform (Further Renewal of Radio Licences) Order 2015, SI
2015/2052.
74
Broadcasting Act 1996, s 58ZA, inserted by Broadcasting Act 1996 (Renewal of Local Radio Multiplex
Licences) Regulations 2015, SI 2015/9 04.
75
C-134/10, Commission v Belgium, 3 March 2011.
76
Articles 2 and 12 Access Directive. See further Chapter 15.
77
Bonner, P and Aston, L, Independent Television in Britain, volume 6 (Basingstoke: Palgrave Macmillan,
2003), 395; Broadcasting Act 1990, s 78A (inserted by the Broadcasting Act 1996).
78
Copyright, Designs and Patents Act 1988, s 73.
076
create any burden for cable providers.79 Now, however, fees are to be the subject
of negotiation between broadcasters and cable providers,80 although government
has expressed a view that the net fee should be zero.81
The public service channels and any other must-carry service are also required
(by way of the Communications Act 2003, ss 272 and 273)82 to be made available
(in so far as possible) by the service provider through ECNs used by a significant
number of end-users as their principal means of receiving television broadcasts
and through satellite services ‘used by a significant number of the persons by
whom the broadcasts are received in an intelligible form as their principal means
of receiving television programmes’. The first provision clearly echoes (but is not
required by) Article 31 of the Universal Services Directive. These ‘must-offer’ pro-
visions mean that the providers are required to offer the channel to the network
and satellite providers; the providers cannot charge their users for this service.
79
Oliver & Ohlbaum Associates, ‘PSB network platform re-transmission and access charges in the UK’ June 2011
<http://downloads.bbc.co.uk/aboutthebbc/reports/pdf/RetransmissionandAccessChargesReview.pdf>.
80
The wide scope of this provision (to include the internet) was initially confirmed in ITV v TVCatchup
[2011] EWHC 1874 (Pat) and further explained as excluding mobile access in ITV v TVCatchup [2015] EWCA Civ
204. However, as part of the same proceedings, the CJEU subsequently found that s 73 was not compatible with
EU Law (specifically, Directive 2001/29/EC on the harmonisation of certain aspects of copyright and related
rights in the information society): Case C-275/15, ITV Broadcasting v TVCatchup. Meanwhile, s 73 has also
been repealed: Digital Economy Act 2017, s 34.
81
<ht t ps:// w w w.gov.u k / g over n ment/ u ploads/ s ystem/ u ploads/ a t tach ment _ d ata/ f i le/ 2 25783/
Connectivity_Content_ a nd_Consumers_ 2013.pdf>.
82
Not in force until 2010: SI 2009/2130.
83
See further Chapter 8. See also Walden, I, ‘Who owns the media? Plurality, ownership, competition and
access’ in Goldberg, D, Sutter, G and Walden, I (eds), Media law and practice (Oxford: OUP, 2009) 42–4 6. In the
UK, the Freeview EPG is licensed as a ‘digital television additional service’: <http://w ww.digitaluk.co.uk/_ _
data/a ssets/pdf_fi le/0 017/86120/L icence_DTAS_0 44.pdf>.
84
<https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_fi le/0 031/19399/epgcode.pdf>.
70
85
eg Digital UK, ‘Logical Channel Number policy’ 3 April 2017 <http://w ww.digitaluk.co.uk/_ _data/a ssets/
pdf_fi le/0 009/86814/Digital_U K_L CN_Policy_v 6.0.pdf>; Sky, ‘Method for allocating listings in Sky’s EPG’
May 2015. <http://s 3-eu-west-1.amazonaws.com/skygroup-sky-static/documents/about-sky/regulatory-i n-
formation/method-for-a llocating-l isting-i n-skys-epg-290116.pdf>.
86
JML Direct v Freesat UK [2009] EWHC 616 (Ch); affirmed in [2010] EWCA Civ 34. Freesat is an EPG (and
associated marketing campaign and approved hardware) which includes free-to-a ir services distributed via
satellite.
87
Bulkley, K, ‘Why is Sky’s tightening of its EPG rules so sensitive?’ Broadcast, 16 November 2007 <http://
www.broadcastnow.co.uk/n ews/multi-p latform/c omment/w hy-i s-s kys-t ightening- o f-i ts- e pg-r ules-s o-
sensitive/264165>. Sky’s current approach to this secondary market (last updated 2015) is set out in section
4.2 of its allocation procedure; a major review of the EPG (by Sky) is now underway: Farber, A, ‘Sky plans EPG
shake-up’ Broadcast, 10 August 2017.
88
Directive 95/47/EC ‘on the use of standards for the transmission of television signals’, OJ L 281/51, 23
November 1995.
89
Ofcom, ‘Provision of technical platform services: guidelines and explanatory statement’ 21 September
2006 <http://stakeholders.ofcom.org.uk/binaries/consultations/t psguidelines/statement/statement.pdf>. It
was determined (pursuant to the Communications Act 2003, s 48) that Sky is the only provider within the
scope of the Directives: Oftel, ‘The regulation of conditional access’ 24 July 2003. <http://w ww.ofcom.org.uk/
static/a rchive/oftel/publications/eu_d irectives/2003/condac0703.pdf>.
90
Rapture TV v Ofcom [2008] CAT 6.
078
appropriate.91 For other services (EPG and conditional access), regulation is still in
place, and Sky publishes a detailed breakdown of prices in these categories.92 The
BBC disputed the level of (and wider necessity for) the payments (for EPG and re-
gionalization) it made to Sky;93 a 2014 agreement between the two organizations set
the fee at zero, as part of a long-term agreement that Sky would carry BBC linear and
on-demand services.94
91
Ofcom, ‘Review of Sky’s access control services regulation’, 17 March 2015, <https://w ww.ofcom.org.
uk/_ _data/a ssets/pdf_ fi le/0 025/57418/ac_ statement.pdf>; Sky, ‘Voluntary commitments by Sky in relation
to the provision of access control services’, 23 April 2015 <http://corporate.sky.com/documents/about-sky/
regulatory-i nformation/v oluntary-c ommitments-b y-s ky-i n-r elation-t o-t he-provision-of-access-c ontrol-
services.pdf>.
92
<http://c orporate.sky.com/d ocuments/a bout-s ky/r egulatory-i nformation/s ky-a nd-s ssl-p ublished-
price-l ist.pdf>.
93
Neilan, C, ‘BSkyB and BBC clash on fees’ Broadcast, 20 October 2010. <http://w ww.broadcastnow.co.uk/
news/broadcasters/bskyb-a nd-bbc-clash-on-fees/5033524.Article>; Toynbee, P, ‘How the badly maimed BBC
can stand up to parasitic Sky’ Guardian, 3 January 2012. <http://w ww.guardian.co.uk/commentisfree/2012/
jan/02/maimed-bbc-parasitic-sky>.
94
Plunkett, J, ‘BBC and BSkyB reach agreement over retransmission payments’ Guardian, 28 February
2014 <https://w ww.theguardian.com/media/2014/feb/2 8/bbc-bskyb-agree-retransmission-deal>.
95
States are not required to compile a list. By 2002 only four lists had been approved within the EU (Ward, n
5, 66–67); by 2008 this had doubled to eight ([2008] OJ C17/7). As of 2017, one further has been added ([2015] OJ
L27/37) and a tenth is under consideration. Norway, as an EEA member, also has an approved list.
96
The Commission’s approval can be challenged; see for instance Case T-385/07, FIFA v Commission (un-
successful General Court challenge to the listing of all games in the (football) World Cup final stage by the
responsible bodies in Belgium), upheld in Case C-204/11 P, FIFA v Commission (parallel cases also saw the
upholding of a similar designation in the UK); Case E-21/13, FIFA v EFTA Surveillance Authority (unsuccessful
EFTA court challenge to a similar designation by Norway).
079
is dealt with in Part 4 of the Broadcasting Act 1996,97 and the qualifying services
(reception without charge and by at least 95 per cent of the population) are the
BBC television broadcasting services (ie BBC1 and BBC2), channel 3 services (ie
ITV1), Channel 4, and (since 2000) Channel 5.98 The UK list is divided into those
events requiring live coverage (group A) (such as the Olympic Games and the FA
Cup Final) and deferred coverage (group B) (such as the Commonwealth Games
and the Ryder Cup). By way of support to the statutory provisions, the standard li-
cences for television services ensure that these provisions are complied with.99 The
UK is also responsible for ensuring that the ‘lists’ of other states are not violated
by broadcasters under UK jurisdiction which are received in those states; this was
considered and confirmed by the House of Lords in a 2001 decision.100
The right of reply is set out in Article 28;101 it requires states to provide for a right
of reply or equivalent remedy for any person (including a legal person) whose legit-
imate interests have been ‘damaged by an assertion of incorrect facts’ in a television
programme. In the UK, this is found in rule 7.11 of the Broadcasting Code as a re-
quirement for programmes to allow the subject of significant allegations to respond.
It is clear (as compared, for example, with the standalone right of reply ‘response’
provision now included in Irish legislation)102 that the approach is one where the
right of reply is dealt with as an aspect of programme-making rather than a separate,
subsequent right.
The AVMSD added a new provision (Article 15) on the right to reproduce in news
reports short extracts of broadcast material (where it is the subject of exclusive
rights). In 2013, the CJEU considered whether an aspect of this provision (Article
15(6) on limiting charges to directly incurred access costs) was incompatible with
Articles 16 and 17 of the EU Charter of Fundamental Rights (on the freedom to con-
duct a business and on property rights respectively), upholding it.103 The UK has
however not adopted specific legislation on this matter, considering that the existing
fair dealing provisions of copyright law104 and associated case law already deals with
the matter.105
97
Therefore predating the amendments of 1997 to the TVWF Directive. The aspects of listing required by
the Directive but not already included in Part 4 were subsequently added.
98
SI 2000/5 4; further services can be added pursuant to the Broadcasting Act 1996, s 98.
99
For example, condition 7 of the TLCS licence. 100
R (TVDanmark 1) v ITC [2001] UKHL 42.
101
See generally (including the influence of the Council of Europe) Keller n 2, 315; see also the broader
Recommendation 2006/952/EC on the protection of minors and human dignity and on the right of reply in
relation to the competitiveness of the European audiovisual and on-l ine information services industry.
102
Broadcasting Act 2009, s 49; Broadcasting Authority of Ireland, ‘Right of reply scheme’ May 2011 <http://
www.bai.ie/en/bai-publishes-r ight-of-reply-scheme/>.
103
Case C-2 83/11, Sky Österreich v ORF.
104
Copyright, Designs and Patents Act 1988, s 30 backed up by an exclusion of agreements which violate this
provision in the Broadcasting Act 1996, s 137.
105
DCMS, ‘EU-w ide standards for Audio Visual Media Services’ 18 December 2009.
701
The fourth set of provisions has an impact on the content of audiovisual media
services (and on the production industry) through requirements for European
production and independent production. These are commonly referred to as
quotas.
States are required by Article 16 to ensure (‘where practicable and by appro-
priate means’ and to be ‘achieved progressively, on the basis of suitable criteria’)
that a ‘majority proportion’ of transmission time of each broadcaster is comprised
of European works. When calculating transmission time for this purpose, certain
types of programme (namely news, sports events, games, advertising and tele-
shopping, and teletext) are excluded, and the requirement does not apply at all to
local broadcasts (Article 18). Compliance is a condition of Ofcom licences in the
UK.106 Services not receivable in the European Union or in a non-EU language are
also excluded in the UK.
A similar approach is taken under Article 17 to European independent produc-
tion, although the target in this case is 10 per cent of transmission time or (if per-
mitted by the state in question, as the UK does)107 10 per cent of the broadcaster’s
programming budget, including an ‘adequate’ proportion for works produced
within the five years before transmission. This is reflected in UK law in the
Communications Act 2003, s 309 in respect of a ‘range and diversity’ of inde-
pendent programmes included in non-public service DTPS (and condition 3 of the
standard DTPS licence) and is also required (by licence) for TLCS; public service
channels are dealt with separately and discussed later.
The quotas in the Directive are floors rather than ceilings, and a state can im-
pose further obligations on broadcasters under its jurisdiction, as the UK does
for the public service channels. Some overlap with European matters (eg a 25 per
cent quota for independent production, which was the subject of a vocal cam-
paign in the 1980s)108 while others are UK-specific (eg regional programming)
(Communications Act 2003, ss 266–268).
The final set of provisions concerns accessibility. Member States are required
to ‘encourage’ service providers (on demand or television) to make their services
(gradually) accessible to users with visual or hearing disabilities. Although this
general provision is now less relevant in the context of a detailed EU instrument
on accessibility, it provides a basis for national provisions on offerings such as
subtitling, sign-language, and audio description. The corresponding and more
106
Ofcom, ‘European production quotas’ 10 February 2005. <http://stakeholders.ofcom.org.uk/binaries/
consultations/e _p_q/statement/epq_ statement.pdf>.
107
<http://stakeholders.ofcom.org.uk/binaries/consultations/e _p_q/statement/epq_ statement.pdf 5>.
108
The ‘25% Campaign’: Bonner, P and Aston, L, Independent Television in Britain, volume 5
(London: Macmillan, 1998) 185. Now found in the Communications Act 2003, s 277.
71
109
Communications Act 2003, s 303; Code on Television Access Services. Note that smaller channels can
make a financial contribution (eg to the British Sign Language Broadcasting Trust) as an alternative to pro-
viding signed programmes.
110
For on-demand services, the language of the Directive is simply reproduced (Communications Act 2003,
s 368C(2)), but there are no equivalents in the legislation akin to those for television.
111
See discussion of ss 3(2)(e) and 3(4)(g) in R (Gaunt) v Ofcom [2011] EWCA Civ 692 [7].
721
112
The CJEU has considered the interpretation of the underlying AVMS provisions: Case C-2 44/10, METV.
113
The right of a company to bring a complaint is confirmed by the Court of Appeal in R v Broadcasting
Standards Commission (ex parte BBC) [2001] QB 885.
114
Ofcom, ‘Procedures for the consideration and adjudication of Fairness & Privacy complaints’ 3
April 2017 <http://stakeholders.ofcom.org.uk/binaries/broadcast/g uidance/june2011/fairness-privacy-
complaints.pdf>.
115
Ofcom, ‘Procedures for investigating breaches of content standards for television and radio’, 3 April
2017 <http:// s takeholders.ofcom.org.uk/ b inaries/ b roadcast/ g uidance/ j une2011/ b reaches- c ontent-
standards.pdf>.
116
See discussion in Bonner and Aston n 108, 382.
731
117
While one of the many matters regulated through the Broadcasting Code, Ofcom has specific powers
(introduced in 2017) to suspend and revoke licences on this ground: Broadcasting Act 1990, s 111B, as inserted
by Digital Economy Act 2017, s 91.
118
<https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 023/105269/I man-F M-Revocation-Notice.pdf>.
119
Ofcom, ‘Procedures for the consideration of statutory sanctions in breaches of broadcast licence’, 3 April
2017, <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 030/71967/P rocedures_ for_consideration.pdf>.
120
Traveller Movement v Ofcom [2015] EWHC 406 (Admin) (unsuccessful challenge by a disappointed
complainant, primarily to the way that Ofcom gives broadcasters but not those who have complained the
opportunity to respond to a preliminary view in standards complainants); R (DM) v Ofcom [2014] EWHC 961
(Admin) (unsuccessful challenge by a broadcaster, referring to apparent bias and to irrationality).
121
R (Gaunt) v Ofcom [2011] EWCA Civ 692, affirming R (Gaunt) v Ofcom [2010] EWHC 1756 (Admin). See
further Hare, I, ‘Insulting Politicians on the Radio?’, (2012) 4 Journal of Media Law 29,
741
breach of Section 2 of the Code (but no sanction was applied) by Ofcom. The ar-
gument advanced was that Ofcom (a public authority for the purposes of Section
6 Human Rights Act 1998) had violated Article 10 of the European Convention
on Human Rights (ECHR) through a disproportionate restriction on freedom of
expression. Gaunt established standing without difficulty and put forward argu-
ments including his statements being a value judgement rather than a statement
of fact, the importance of political speech, drawing on a range of decisions from
the European Court of Human Rights (ECtHR), and the lower impact of radio as
compared to television. He did not however challenge either the statutory provi-
sions or the Code itself, and these provisions confirmed for ECHR purposes that
the restriction was prescribed by law and pursuing a legitimate goal, leaving pro-
portionality as the issue before the court.
The analysis at first instance (in the High Court) is short; while the importance
of freedom of expression is recognized (including ‘heated and even offensive dia-
logue which retains a degree of relevant content’, this does not extend to ‘gra-
tuitous offensive insult or abuse’. Ofcom’s decision (including the lack of sanction
on either Gaunt or TalkSport) was therefore found to be justified. This was con-
firmed by the Court of Appeal in a more detailed analysis; Neuberger MR noted
that Ofcom should be ‘slow’ to find a violation of Section 2 of the Code, in the
light of Article 10 and this particular interview, but that such a finding could not
be excluded. He concluded that ‘the publication of the Finding, which contained
no sanction, other than the stigma of the publication of an adverse finding by the
statutory regulator’, was not an interference with Gaunt’s Article 10 rights. Ofcom’s
approach to the matter was not criticized in any way and the finding itself was
found to be proportionate.
SI 2009/2979.
122
751
123
Ofcom, ‘Proposals for the regulation of video on demand services’, 14 September 2009, <http://stake-
holders.ofcom.org.uk/binaries/consultations/vod/summary/vod.pdf>; Ofcom, ‘The regulation of video on
demand services’, 18 December 2009, <http://stakeholders.ofcom.org.uk/binaries/consultations/vod/state-
ment/vodstatement.pdf>.
124
Tambini, D, Leonardi, D and Marsden, C, Codifying Cyberspace: Communications Self-regulation in the
Age of the Internet (London: Routledge, 2008) 99.
125
Woods, L, ‘Internet protocol TV: ATVOD’, in Marsden et al (eds), Options for and Effectiveness of Internet
Self-and Co-regulation: Phase 2 case study report (RAND Europe, Cambridge 2008), 181–182.
126
Criteria for designation of a regulatory body are set out in the Communications Act 2003, s 368B(9).
Further amendments to the Communications Act 2003 facilitated co-regulation through powers to require
notification, charge fees and require the retention of recordings; these provisions were required to be notified
to the European Commission under the Technical Standards Directives: see SI 2010/419, inserting s 368D(3)
(zb) into the Communications Act 2003; Directive 98/3 4/EC laying down a procedure for the provision of in-
formation in the field of technical standards and regulations, [1998] OJ L 204/37; Directive 98/4 8/EC of the
European Parliament and of the Council of 20 July 1998 amending Directive 98/3 4/EC [1998] OJ L 217/18.
127
<https://w ww.ofcom.org.uk/t v-radio-a nd-on-demand/i nformation-for-i ndustry/on-demand>.
128
Communications Act 2003, s 386N.
761
• its principal purpose is the provision of programmes the form and content of
which are comparable to the form and content of programmes normally in-
cluded in television programme services;
• access to it is on-demand;
• there is a person who has editorial responsibility for it;
• it is made available by that person for use by members of the public;
• that person is under the jurisdiction of the United Kingdom for the purposes of
the Audiovisual Media Services Directive.
Further details were found in Ofcom’s Scope Guidance and ATVOD’s ‘Guidance
on who needs to notify’, A failure to notify was initially addressed through a pro-
visional view, leading where applicable to a Scope Determination (recording a
breach of Sections 386BA and 386D(ZA) of the Communications Act 2003 (notifi-
cation and fees respectively)), normally applying the above-mentioned criteria in
Section 368A(1) of the Communications Act. Determinations could be appealed
to Ofcom.
The question of how to apply the new system to on-demand film services
emerged at an early stage. VOD is a platform particularly suited for film; early
projects (including NVOD) promoted the ability to access a catalogue of movies
(including those first shown in cinemas). However, this can mean that there is a
substantial regulatory difference between supply on a physical disc (ie DVD) and
in electronic format through VOD. The former is subject to the Video Recordings
Act 1984 (discussed below),129 while the latter is beyond its scope (although the
BBFC offers to rate VOD works on a non-statutory basis). On-demand film services
can therefore be required to notify if they fall within scope. The matter was con-
sidered at some length during the drafting of the scope guidance, particularly by
reference to the similarity (explained by the BBFC but rejected by Ofcom) between
an on-demand film service and a DVD retail store.
129
The 1984 Act, as amended, was re-enacted in 2010 in order to address an initial failure to notify the
European Commission under the Technical Standards Directives.
130
The definitions in the Directive have also been considered by the CJEU, in C-374/14, New Media Online v
Bundeskommunikationssenat, 21 October 2015.
71
Press Complaints Commission (PCC) and then the Independent Press Standards
Organisation (IPSO)) applied the Editors’ Code to the websites of newspapers. The
AVMSD excluded ‘electronic versions of newspapers’ (Recital 28). Eight service
providers were then the subject of Scope Determinations (concerning audiovisual
material on newspaper websites), and all eight refused to notify and/or appealed
the Determination to Ofcom.131
In Ofcom’s first ruling, concerning Sun TV,132 it reviewed the Directive and
ATVOD’s guidance, and set aside ATVOD’s decision. The principal purpose test
was that a relevant service would be more likely to have its own homepage, to have
a separate section of a website where the audiovisual material is catalogued and
accessed, to be styled/presented as a television channel, to have more than bite-
sized clips, to have limited or no links between audiovisual and other material, to
have more audiovisual than written material (with the audiovisual material being
the primary means), and overall not said to be integrated into nor ancillary to an-
other service. ATVOD withdrew the other Scope Determinations in this category
on the basis of these Ofcom findings.
Following the Sun TV decision on principal purpose, Ofcom also heard a range
of appeals on the ‘comparability’ ground concerning other sectors.133 These de-
cisions saw Ofcom trying to operationalize the ‘television-like’ language of the
Directive, in contexts such as the duration of clips, the quality of production, and
formal features like titles and credits.
Quite a number of these appeals came from the providers of sexually explicit
websites.134 Initially, a key (and unsuccessful) ground of appeal was that the con-
tent would not be permitted under the terms of the Broadcasting Code, because it
was classified as R18 by the BBFC (or equivalent),135 and so did not meet the com-
parability test. Ofcom emphasized that the requirement was not that the services
being compared be identical, interpreting ‘normally included’ as a general term
rather than a specific reference to the current UK Broadcasting Code. Findings
that adult services are on-demand audiovisual media services has particular con-
sequences, as the content requirements on the protection of minors (discussed
below) require the use of access control.
131
The Independent Video; FT Video; The Guardian YouTube Channel; Guardian Video; Telegraph TV; Sun
Video; News of the World Video; Sunday Times Video Library; Elle TV.
132
<http://stakeholders.ofcom.org.uk/binaries/enforcement/vod-services/sunvideo.pdf>; see further Katsirea, I,
‘Electronic press: “press-like’’ or “television-like’’?’ (2015) 23 International Journal of Law and Information
Technology 134.
133
For a detailed discussion of these appeals, see Mac Síthigh, D, ‘ “TV-l ike”: Aesthetics, Quality and Genre
in the Regulation of Video-on-demand Services’, (2017) 14 Journal of British Cinema and Television 464.
134
For a detailed analysis of these decisions, see Petley, J, ‘The Regulation of Pornography on Video-on-
demand in the United Kingdom’, (2014) 1 Porn Studies 260.
135
Broadcasting Code s 1.17 contains this prohibition.
781
136
Ofcom, ‘Sexually explicit material and video-on-demand services’ <http://stakeholders.ofcom.org.uk/
binaries/i nternet/explicit-material-vod.pdf>.
137
<http://old.culture.gov.uk/i mages/publications/E Vletter-to-ed-r ichards-3aug2011pdf.pdf>.
791
and material refused classification by the BBFC (or its equivalent) could not be in-
cluded at all.138
138
Audiovisual Media Services Regulations 2014, SI 2014/2916, amending Communications Act 2003, s
368E(d)(2).
139
Note however that BBFC decisions have an impact on broadcasting; the Broadcasting Code, for instance,
prohibits the broadcast of a work refused classification or rated R18, and restricts the broadcast of works rated
18 to after 9pm on most services.
140
Video Recordings Act 1984, s 12; see further Hunter, IQ, ‘Naughty Realism: the Britishness of British
Hardcore R18s’, (2014) 11 Journal of British Cinema and Television 152.
141
The grounds for exemption were significantly narrowed in 2014: Video Recordings Act 1984 (Exempted
Video Recordings) Regulations 2014, SI 2014/2097.
142
Digital Economy Act 2017, pt 3.
702
provisions; its powers will also include ordering ISPs to block access to non-
compliant services and to notify payment service providers of non-c ompliant
activity.
Finally, Section 127 of the Communications Act 2003 contains criminal offences
of the improper use of a public ECN. A similar provision first appeared (regarding
telephone) in the Post Office (Amendment) Act 1935, but the current provision has
become one of the more commonly tried offences in relation to Internet commu-
nication.143 While this provision does not apply to programme services (as defined
in the Broadcasting Act, see 14.3.1), it is one which easily applies to on-demand
services (whether subject to Part 4A of the Communications Act or not), PRS (see
14.7), and the internet in general. S127(1) deals with sending or causing to be sent a
message or matter that is ‘grossly offensive or of an indecent, obscene or menacing
character’, while s 127(2) deals with messages known to be false sent for the pur-
pose of causing ‘annoyance, inconvenience or needless anxiety’. The House of
Lords held in 2006 that the offence is the sending of the message rather than its re-
ceipt.144 Further elaboration of the scope of this clause came in Chambers (a high-
profile case stated to the High Court from Doncaster Crown Court concerning
an allegedly menacing message about a temporarily closed airport posted on
Twitter)145 and in CPS guidelines on prosecuting cases involving communications
sent via social media.146
14.6 A DV ER TISING
143
See eg Sutherland, C and Dowling, S, ‘The nature of online offending’ (Home Office Research Report 82,
October 2015).
144
DPP v Collins [2006] UKHL 40.
145
Chambers v DPP [2012] EWHC 2157 (Admin) (confirming that the post was a message sent by a public
ECN, but that it was not menacing); see further McGoldrick, D, ‘The Limits of Freedom of Expression on
Facebook and Social Networking Sites: a UK Perspective’, (2013) 13 Human Rights Law Review 125; Laidlaw, E,
‘What is a joke? Mapping the path of a speech complaint on social networks’ in Gillies, L and Mangan, D, (eds),
The Legal Challenges of Social Media (Edward Elgar, 2017).
146
CPS, ‘Guidelines on prosecuting cases involving communications sent via social media’ (March
2016) (identifying priority categories of credible threats of violence, harassment and similar behaviour, and
breaches of court orders and statue, and a fourth category, with a high threshold for deciding to prosecute, of
other conduct falling within s 127 or other provisions).
147
ASA, ‘History of ad regulation’ <http://a sa.org.uk/Regulation-E xplained/H istory-of-Ad-Regulation.
aspx>.
712
The ASA is funded by a levy (eg 0.1 per cent of spend on display advertising and
broadcast airtime). The levies are collected by two industry boards, made up of
trade associations such as the Advertising Association and the Periodical Publishers
Association, which share the same staff and premises: the Advertising Standards
Board of Finance (ASBOF) and Broadcast Advertising Standards Board of Finance
(BASBOF). The Boards appoint the chair of the ASA. The codes are written by the
Committee of Advertising Practice (CAP) and the Broadcast CAP, which are made
up of the same members as the finance boards; CAP/BCAP also provide copy advice
services to advertisers.
It is also the industry that is responsible for much of the enforcement of the code.
In principle it is subject to judicial review even for its non-broadcast functions.149 An
Independent Reviewer is funded by ASBOF/BASBOF and conducts reviews of ASA
decisions where new evidence is available or if there is a substantial flaw in the pro-
cess or adjudication.
SI 2008/1277.
148
R v ASA ex parte The Insurance Service (1990) 2 Admin LR 77; R (Debt Free Direct) v Advertising Standards
149
handling, although NTSB can also act without an ASA referral.151 Through CAP, de-
cisions are also enforced in the form of ‘ad alerts’ regarding ASA decisions, asking
for the withdrawal of trade privileges (eg bulk mail discount), require pre-vetting
of an advertiser, and asking an internet advertising platform to remove ads.
Importantly, the ‘non-broadcast’ remit of the ASA is wide-ranging (in tele-
communications terms, it includes mobile phone, email, VOD, and internet ad-
vertising) and has been extended a number of times; it now includes not just
paid-for online space (eg banner ads, which have been considered by the ASA
since 1995) and the fast-g rowing keyword advertising sector, but also promotional
claims on an advertiser’s own website or on a social networking site, as part of the
extended ‘online remit’ which has been in place since March 2011. A substantial
proportion of ASA decisions now relate to internet advertising, and new sanctions
include a list of misleading online advertisers152 and the possibility of a keyword
advertisement being placed which would direct search engine users to informa-
tion on the breach of the code on the ASA’s website.
151
The Consumer Protection from Unfair Trading Regulations 2008, s 19, requires that enforcement au-
thorities ‘shall have regard to the desirability of encouraging control of unfair commercial practices by such
established means as it considers appropriate’.
152
<https://w ww.asa.org.uk/codes-a nd-r ulings/non-compliant-online-advertisers.html>.
153
Ofcom, ‘The regulation of video on demand services’ 18 December 2009 <http://stakeholders.ofcom.org.
uk/binaries/consultations/vod/statement/vodstatement.pdf> 66.
723
providers neither notify the ASA nor pay a fee. The system therefore combines
aspects of broadcast and non-broadcast advertising.
The requirements of the AVMSD are those applicable to audiovisual commer-
cial communications on all audiovisual media services, ie that advertising be rec-
ognisable (and not subliminal or surreptitious), not prejudice respect for human
dignity or promote discrimination, nor encourage behaviour prejudicial to health
and safety or grossly prejudicial to the protection of the environment. Tobacco
and prescription medicine or treatment advertising is not permitted; advertising
of alcohol is restricted ie not aimed at minors, not encouraging immoderate con-
sumption (Articles 9 and 10). The protection of minors is also covered, in general
terms (‘may not cause physical or moral detriment’) and specific (ie exploitation
of credulity, inexperience or ‘pester power’ and showing of minors in dangerous
situations).
154
Contracting Out (Functions Relating to Broadcast Advertising) and Specification of Relevant Functions
Order 2004.
155
<http://stakeholders.ofcom.org.uk/binaries/consultations/reg_broad_ ad/statement/mou.pdf>.
156
Upheld by the House of Lords R (Animal Defenders International) v Secretary of State for Culture, Media
and Sport [2008] UKHL 15, but ECtHR decisions go in different directions, eg VgT v Switzerland (2001) 34 EHRR
159, TV Vest v Norway (2009) 48 EHRR 51 (political advertising bans, violation), and Murphy v Ireland (2003) 38
EHRR 212 (religious advertising ban, no violation).
742
14.6.4.2 Content
In practice, the detailed rules on the content of advertising are found in the 32-
part BCAP Code of Broadcast Advertising. Some are taken directly from the
Communications Act eg rule 7 on political advertising, and in the case of televi-
sion, form part of the implementation of the AVMSD, but many are much more
detailed and not specifically mentioned in the Act eg rule 9 with ten rules on envir-
onmental claims, rule 20 with five rules on motoring. Particularly detailed provi-
sion is made regarding medicines and health (rule 11), weight control (rule 12) and
food and nutrition (rule 13), which also incorporate aspects of UK and EU law on
medicine and nutrition more generally.
Clearcast, a company owned by a number of broadcasters (including ITV,
Channel 4, Channel 5, Sky, and Turner), provides a ‘pre-t ransmission clearance’
service for television (and VOD) advertising. It took over from the ITV-owned (but
funded by multiple broadcasters) Broadcast Advertising Clearance Centre (BACC)
in 2008.158 The participating broadcasters159 require advertisements to be approved
by Clearcast. Approval does not bind the ASA, but Clearcast’s reasons for approval
will be included in the analysis; in a 2016 decision, it was confirmed that Clearcast
decisions are not amendable to judicial review.160 Ofcom licences require broad-
casters to make arrangements for ‘advance clearance’ of advertisements.
14.6.4.3 Scheduling
The AVMSD contains in chapter VII a number of sections on television advertising,
alongside the general requirements applicable to all audiovisual media services.
157
Bonner and Aston vol 5, n 108, p 262.
158
Barnett, E, ‘BACC to change ownership and rebrand as Clearcast’ Media Week 11 September 2007. <http://
www.mediaweek.co.uk/news/737144/>. At the launch of Clearcast, IDS (the advertising sales division of
Virgin Media) was a shareholder, but it has since been dissolved (the Virgin channels were sold to Sky, and
UKTV advertising sales formerly handled by IDS are now handled by Channel 4); Viacom was also listed as a
shareholder.
159
Listed at <http://k b.clearcast.co.uk/w iki/26/channels-we-clear-for>.
160
Diomed Direct v Clearcast (High Court, May 2016).
752
Timing and placement rules are included in Articles 19, 20, and 23.
Advertisements and teleshopping must be kept separate from editorial content
(by optical, acoustic and/or spatial means); the integrity of programmes should
be protected, with isolated ‘spots’ being the exception other than for sports events.
20 per cent of a clock hour of a television broadcast (ie 12 minutes) can be dedi-
cated to advertising and teleshopping; Member States can set a lower limit for all
or even for some broadcasters (eg pay TV), subject to considerations of proportion-
ality.161 In general, the number and duration of breaks is a matter for national law,
although the CJEU has intervened to give a broad definition to what falls within
the definition of an advertising spot.162 General provisions on the minimum ‘gap’
between breaks are no longer included, but for some genres and types eg films,
there may be one interruption for each 30 minutes.
These rules are implemented and added to in the UK in the Code on the
Scheduling, Transmission and Amount of Advertising (COSTA), which is pro-
vided for in Section 322 of the Communications Act. In particular, while the
AVMSD no longer regulates the overall amount of advertising per day (ie as op-
posed to the amount of advertising in a given hour, which is capped at 12 min-
utes), COSTA adds a restriction of nine minutes average minutes per hour of
advertising in a day (and three minutes teleshopping), which was confirmed
after review in 2011.163 Advertising on public service channels is subject to fur-
ther restrictions. It is limited to seven minutes per hour overall (eight minutes
per hour between 6pm and 11pm) instead of nine, and a break can be no longer
than three minutes 50 seconds, of which three minutes 30 seconds is advertise-
ments. Longer breaks are permitted during what are defined as films in COSTA
(ie films made for TV and cinematographic works, which in practice includes
‘single dramas’).164
The scheduling of radio advertising is a matter for domestic law. Chapter 10 of
the Broadcasting Code requires separation of spot advertising from program-
ming, but does not restrict solus advertising (a single spot instead of a break), nor
the length and number of breaks or the overall or hourly amount of advertising.
BCAP includes (in respect of both radio and television) a range of further re-
strictions on scheduling in rule 32, eg on advertisements broadcast in or adja-
cent to programmes ‘principally directed at or likely to appeal particularly to’
under 18s, or avoiding ‘unsuitable juxtapositions’ between programmes and
advertisements.
ofcom.org.uk/broadcasting/broadcast-codes/advert-code/ad-m inutage>.
164
First introduced as a trial in 2011, and made permanent in a 2014 amendment to COSTA.
762
14.6.5.2 Participation TV
When a programme is designed primarily to promote a premium rate service
(PRS) (see 14.7), then the programme is considered and regulated as an adver-
tisement rather than a programme.166 This is considered in part in the Court of
Justice decision in Kommunikationsbehörde v ORF.167 The characterization of
what is referred to as Participation TV (PTV) as advertising rather than ‘editorial’
means that the content requirements of BCAP and potentially the scheduling
requirements of COSTA will apply. On the other hand, the Broadcasting Code
restriction on charging for participation in programmes (to PRS) does not apply
to advertising.
PTV has presented a number of regulatory challenges in the UK. It is now de-
fined as a programme which has as its primary purpose the promotion of PRS or
viewer-paid interaction through other mechanisms. Many of these services, such
as adult chat or dating services, make virtually all their income through PRS. PTV
is now treated as a type of audiovisual commercial communication (ie teleshop-
ping), meaning that the licence for a PTV service must specify that it is a tele-
shopping service, and it is subject to the BCAP code rather than the Broadcasting
Code, with Ofcom remaining as the regulatory authority. A specific section within
165
Medicinal products which are subject to a marketing authorization within the meaning of Directive
2001/8 3/EC.
166
Ofcom, ‘Participation TV: regulatory statement, rules on the promotion of premium rate services’ 3 June
2010 <http://stakeholders.ofcom.org.uk/binaries/consultations/participationtv3/statement/statement.pdf>.
167
C-195/0 6, [2007] ECR I-8 817.
72
Section 120 of the Communications Act 2003 defines a PRS as the provision of content
by means of an ECN or allowing an ECS user to make use (by means of that service) of
a facility available to users of that ECS. An example of the former is a chat service (live
or recorded) and a telephone voting system is an example of the latter. There must be
a charge for the service, paid in the form of a charge for use to the provider of the ECS
through which the service is provided.
Certain services are directed by Ofcom to be ‘controlled PRS’, and are regulated by
the Phone-paid Services Authority (formerly PhonepayPlus, and before that ICSTIS)
and Section 120 of the Communications Act, being required to comply with the PSA
Code approved by Ofcom under s 121. The PSA is also subject to judicial review.168
Other PRS (ie within s 120 but outside the direction) may be within the scope of the
Act but not considered controlled PRS, in which case code compliance is voluntary.
There are five types of controlled PRS: a service with a charge of more than 10
pence per minute, calls to services with certain prefixes (087, 090, 010, and 118) with
a charge of more than 5 pence per minute, and three services of any charge: chat-
line, sexual entertainment service, or internet dialler software. A service accessed
through an international call is not a controlled PRS. Ofcom also excludes from
the definition of controlled PRS platforms where ‘the Communications Provider
providing the ECS is also the service provider providing the service delivered by
means of the ECS’, although the Payforit service (where other goods and services
can be charged to a mobile phone bill) remains controlled PRS.
This system succeeds a licence-based system (where compliance with the code
was a condition of a telecoms licence), no longer possible after the introduction
of authorization and the abolition of telecommunications licensing in 2002/3.
The maximum penalty for breach of the Communications Act s 120 is currently
£250,000169 (s 123); directions to communications providers that the provision of a
particular PRS be suspended can also be issued (s 124). A Tribunal of three mem-
bers without a commercial interest in PRS (one legally qualified and two lay), drawn
from a panel, has a wide range of sanctions at its disposal, including formal warn-
ings, fines, mandatory refunds for consumers, barring of services, and prohibitions
on individuals. The PSA can take action to recover a debt (eg unpaid fines).170
168
ICSTIS v Andronikou (liquidators of Allied Communications Ltd) [2007] EWHC 2307 (Admin).
169
This is the maximum per breach, so a single service provider can be fined in excess of this amount in the
case of multiple breaches: Consumer Rights Act 2015, s 80.
170
ICSTIS v Andronikou (liquidators of Allied Communications Ltd) [2007] EWHC 2307 (Admin);
PhonepayPlus v Ashraf [2014] EWHC 4303 (Ch).
729
171
An unsuccessful argument regarding the application of the derogation was made in R (Ordanduu and
Optimus Mobile) v PhonepayPlus [2015] EWHC 50 (Admin).
703
A key theme in the review of legislation and policy regarding broadcasting re-
mains that of technological convergence.173 The Audiovisual Media Services
Directive constituted a move in this direction, although it is also based on a very
significant distinction between linear and non-l inear services, applying radically
different regulatory models to each. These is of course a crucial third category, ie
those services not within the scope of the Directive—a lthough the interpretation
of the Directive, and the proposed changes to it, moves services in and out of this
category.174
The adoption of on-demand services continues. This is a story that has been
characterized by many false dawns, with early VOD trials taking place in the
UK over 30 years ago, and rhetoric on interactive television reaching its height
at the turn of the century. VOD is now widely available on the internet, particu-
larly through the ‘catch-up’ services provided by familiar broadcasters (eg the
BBC’s iPlayer, Channel 4’s All4), by subscription (Amazon Prime, Netflix),175 and
electronic sell-through (iTunes). VOD services delivered via the internet can be
accessed on other devices (games consoles, smartphones, tablets, and internet-
enabled television sets). Since the last edition of this book, Netflix in particular has
seen significant growth, and it and other providers become more active as com-
missioners of original content alongside what was for long their core catalogues of
works previously broadcast or exhibited elsewhere. The BBC moved its BBC3 ser-
vice to on-demand only, and new devices and services (eg YouView, Freeview Play)
provide end users with access to linear and on-demand services through a single
interface. Perceived disparities also influence the ongoing process of reviewing
regulatory requirements, although sometimes on the premise of levelling down
172
Ayre, R, ‘Report of an inquiry into television broadcasters’ use of premium rate telephone services in pro-
grammes’ 18 July 2007 <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 020/42914/report.pdf>; House of
Commons Culture, Media & Sports Committee, ‘Call TV quiz shows’ 25 January 2007, HC 72 <http://w ww.
publications.parliament.uk/pa/c m200607/c mselect/c mcumeds/72/7203.htm>.
173
The author’s scepticism regarding convergence as the basis for regulation is set out further in Mac
Síthigh, D, Medium Law (Routledge, 2017).
174
For instance, the ‘comparability’ test discussed in this chapter is under review, and (as noted above) a
new category for video sharing platforms (in essence sitting between on-demand services and non-AVMS
services) may be established.
175
Amazon Prime Video (formerly LoveFilm) continues to have mix of electronic sell-through and sub-
scription services.
713
and sometimes levelling up. For instance, Ofcom asked in 2016 for comments on
whether it should permit some post-watershed content to be aired at other times
where PIN restriction is in place.176 A clear influence was competition from on-
demand services, where although PINs and other forms of access control are used,
these methods apply twenty-four hours a day. In the other direction, the new re-
quirements for sexually explicit web material in the Digital Economy Act 2017 were
proposed as ways of ensuring that requirements currently applicable (in varying
ways) to on-demand and DVD services would also apply to the web more generally.
The relationship between broadcasting and telecommunications continues
to be a difficult one. Although the definitions applying to services delivered via
various types of ECN are relatively clear, this does not mean that there is polit-
ical or industry consensus on the appropriateness of applying any particular ap-
proach to regulation to a given platform or type of service available on a platform.
The Leveson Inquiry, set up in response to the ‘phone-hacking’ allegations against
the News of the World and wider concern about press standards, also considered
issues of media regulation more generally, including the distinction between print,
broadcast, and internet regulation. A House of Lords committee also reported at
length on how to rethink media regulation in light of technological change.177 In
July 2013, the Department for Culture, Media, and Sport178 set out various thoughts
in a Green Paper, including the merits of a ‘common framework for media stand-
ards’ and whether EPG prominence rules could be further developed.179
A proposed Communications Bill could have addressed some of these issues,
but more importantly usefully consolidated the fragmented regulation of broad-
casting (spread across the Broadcasting Acts 1990 and 1996, the Communications
Act 2003, and subsequently amended including for the purposes of transposing
the AVMSD and the Digital Economy Act 2010), removed otiose categories (per-
haps analogue television and the public teletext service), and recognized the
changes pertaining to ITV (including the future of non-BBC public service broad-
casting). Despite some helpful discussion papers and events from DCMS, though,
no Bill ever came forward. Now, while the AVMS Directive is being reviewed, the
UK is also preparing to withdraw from the European Union, with obvious con-
sequences for the actors and priorities in the UK content and broadcast sectors,
and so the effect of the broadcasting and communications legislation. Piecemeal
amendment may be the main means of responding to change for some time.
176
<https://w w w.ofcom.org.uk/c onsultations-a nd- s tatements/c ategory-3/m andatory- d aytime-p in-
protections>.
177
House of Lords Select Committee on Communications, ‘Media convergence’ (HL 154, 2012–2013).
178
Now, after a 2017 rebrand, the Department for Digital, Culture, Media, and Sport.
179
<ht t ps:// w w w.gov.u k/ g over nment/ u ploads/ s ystem/ u ploads/ a t tachment _ d ata/ f i le/ 2 25783/
Connectivity_Content_ a nd_Consumers_ 2013.pdf>.
723
73
15
Christopher T. Marsden
• The general liability of ISPs for third party content and its enforcement;
• The self-regulatory schemes1 established by ISPs to take down illegal content (eg
the Internet Watch Foundation);
• New laws and regulations regarding permissible Quality of Service (QoS) be-
tween ISPs, which is termed ‘network neutrality’. European law provides the
1
See definitions in Leveson, B, ‘An Inquiry Into the Culture and Ethics of the Press, Politicians and
Police: Volume IV’ (2012) at 1739, para 2.31, <https://w ww.gov.uk/government/uploads/s ystem/uploads/at-
tachment_data/fi le/270943/0780_ iv.pdf>.
734
The term ‘ISP’ has different legal meaning in different contexts, though it is used
much more often than more legally specific terms. In Europe, a provider of internet
access is an Electronic Communications Network Provider (ECNP), whereas a pro-
vider of content and services is termed an Information Society Service Provider
(ISSP) under the Electronic Commerce Directive (ECD)3 or an Audiovisual Media
Services provider.4 Under the European Framework Directive,
‘electronic communications service’ means a service normally provided for re-
muneration which consists wholly or mainly in the conveyance of signals on elec-
tronic communications networks, including telecommunications services and
transmission services in networks used for broadcasting, but exclude services
providing, or exercising editorial control over, content transmitted using elec-
tronic communications networks and services.5
The definition explicitly excludes ISSPs ‘which do not consist wholly or mainly in
the conveyance of signals on electronic communications networks’.
In the US, the access provider is an Internet Access Provider (IAP), and the ser-
vice provider an Online Service Provider (OSP) under the Digital Millennium
Copyright Act (DMCA),6 though a further distinction lies between access providers
classified under Title I and Title II of the Communications Act 1934.7 In Section
512 an OSP is defined as ‘an entity offering transmission, routing, or providing
2
See Regulation (EU) 2015/2120 of the European Parliament and of the Council of 25 November 2015 laying
down measures concerning open internet access and amending Directive 2002/22/EC on universal service and
users’ rights relating to electronic communications networks and services and Regulation (EU) No 531/2012
on roaming on public mobile communications networks within the Union, OJ L 310, 26 November 2015, 1–18.
This amended Directive 2009/140/EC, OJ L 337/37 18 December 2009; Directive 2009/136/EC, OJ L 337/11
18 December 2009.
3
Directive 2000/31/EC, Art 2(a), OJ L 178/1, 17 July 2000, reiterating Art 1(2) of Directive 98/3 4/EC (OJ L
204/37, 21 July 1998) as amended by Art 1(2)(a) and Annex V of Directive 98/4 8/EC, OJ/L 217/18, 5 August 1998.
4
See Directive 2010/13/E U, Art 1(a)(i) and Art 2 of 10 March 2010 on the coordination of certain provi-
sions laid down by law, regulation or administrative action in Member States concerning the provision of
audiovisual media services (Audiovisual Media Services Directive) (codified version) OJ/L 95/1, 15 April 2010.
5
Directive 2002/21/EC, OJ L 108/33, Art 2(c).
6
Online Copyright Infringement Liability Limitation Act (OCILLA) which amended the 1976 Copyright
Act, passed as a part of the 1998 Digital Millennium Copyright Act (DMCA) and referred to as the ‘Safe Harbor’
(sic) provision because it added Section 512 to Title 17 of the United States Code.
7
47 USC §201(a) and (b). See Chapter 5.
753
It is instructive to note that since the US Supreme Court ruled in ACLU v Reno in
1997,9 it took over a decade for the senior European court (the CJEU) to consider
ISP liability for third party violation of intellectual property rights (IPRs) against
constitutional rights to free expression and privacy of personal data, in Perfect
Communication10 and Scarlet Extended.11
Limiting liability for ISPs as publishers, rather than authors, of material hosted
on behalf of third parties was established via early US caselaw,12 though the
8
Digital Millennium Copyright Act 1998, s 512(k)(1)(A–B).
9
American Civil Liberties Union v Reno (1997) 21 US 844 of 27 June No 96–511.
10
Case C-4 61/10: Bonnier Audio AB and others v Perfect Communication Sweden AB, OJ C 317, 20/11/
2010 P. 0024—0 024 final judgment 19 April 2012 at <http://c uria.europa.eu/juris/document/document.jsf?d
oclang=EN&text=&pageIndex=0&mode=DOC&docid=121743&cid=848081>.
11
Case C-70/10 Scarlet Extended SA v Société Belge des auteurs, compositeurs et éditeurs (SABAM) OJ C 113, 1
May 2010: 20–20. Decided 24 November 2011, OJ C 25/6, 28 January 2012. <http://curia.europa.eu/juris/document/
document.jsf?text=&docid=115202&pageIndex=0&doclang=EN&mode=doc&dir=&occ=first&part=1&cid=574684>.
12
Case law dates to the Bulletin Board Services of the 1980s, see generally references in Marsden, C, Oxford
Bibliography of Internet Law, section ‘Origins of Internet Law’, (New York: OUP, 2012).
763
There are five leading cases, as well as a statutory safe harbour applicable in the
copyright context, which together set out the principles of application of DMCA
13
Section 30, 47 USC §230(c)(1) (Supp. II 1996). This language might shield ISPs from liability for subscriber
copyright infringement as well. However, Section 230(e)(2) specifically states, ‘Nothing in this section shall be
construed to limit or expand any law pertaining to intellectual property.’
14
Yen, Alfred C, ‘Internet Service Provider Liability for Subscriber Copyright Infringement, Enterprise
Liability, and the First Amendment’, (2000) 88 Georgetown Law Journal 1833–1893.
15
Holznagel, B, ‘Responsibility for Harmful and Illegal Content as Well as Free Speech on the Internet in
the United States of America and Germany’, in Engel, C and Keller, H, (eds), Governance of Global Networks in
Light of Differing Local Values (Baden Baden: Nomos, 2000).
16
Frydman, B and Rorive, I, ‘Regulating Internet Content Through Intermediaries in Europe and the USA’,
(2002) Zeitschrift fur Rechtssoziologie Bd.23/H 1 July, Lucius et Lucius.
17
Delle v Worcester Telegram & Gazette Corp, 2011 WL 7090709 (Mass Super Ct 14 September 2011).
73
s 512. In A&M Records v Napster,18 the court refused to extend the safe harbour pro-
visions to the Napster software program and service, leaving open the question of
whether peer-to-peer networks also qualify for safe harbour protection under Section
512. The second is In re Aimster Copyright Litigation19 (involving accessorial copy-
right infringement by an online intermediary based on its use of technology to blind
itself wilfully to infringement by its users); the third Metro-Goldwyn-Mayer Studios,
Inc v Grokster, Ltd20 (involving alleged accessorial copyright liability for online peer-
to-peer intermediaries that facilitated the illegal exchange of copyrighted material);
the fourth Viacom et al v YouTube, Inc et al21 (involving alleged direct and accessory
copyright liability for hosting of video clips where some were illegally posted in vio-
lation of copyright law), which was appealed with oral hearing by the 2nd Circuit on
18 October 2011. The fifth case is Tiffany (NJ), Inc v eBay, Inc22 (involving the hosting
of Tiffany products for auction where the goods in question were largely, though not
exclusively, counterfeits).
The Supreme Court explained in Grokster:23
[o]ne infringes contributorily by intentionally inducing or encouraging direct
infringement,[24] and infringes vicariously by profiting from direct infringement
while declining to exercise a right to stop or limit it[25]. . . [t]he Copyright Act does
not expressly render anyone liable for infringement committed by another,[26]
these doctrines of secondary liability emerged from common law principles and
are well established in the law.[27]
The Supreme Court distinguished the Sony doctrine, which had legalized the use
of video recording equipment to record live television as an ancillary use,28 based
on the fault-based approach, in that Grokster had ‘an actual purpose to cause
infringing use is shown by evidence independent of design or distribution of the
product’29 noting ‘nor would ordinary acts incident to product distribution, such
20
Metro-Goldwyn-Mayer Studios, Inc v Grokster, Ltd 545 US 900 (2005) (‘Grokster’).
21
Viacom et al v YouTube, Inc et al 718 F Supp 2d 514 (SDNY 2010). See also Perfect 10, Inc v Amazon, Inc, 508
F 3d 1146, 1172–73 (9th Cir 2007) (discussing and applying Grokster and Sony), and Perfect 10, Inc v Google Inc
416 F Supp (2nd 828) CD Cal 2006 (Thumbnails in Web searches were fair use. Framed inline images of full size
were not infringing copies. 9th Circuit reversed the District Court’s holding of no Fair Use).
22
Tiffany (NJ), Inc v eBay, Inc 600 F 3d 93 (2nd Cir 2010). 23
Grokster, n 20, at 930.
24
Citing Gershwin Pub Corp v Columbia Artists Management, Inc, 443 F 2d 1159, 1162 (CA 2 1971).
25
Citing Shapiro, Bernstein & Co v HL Green Co, 316 F 2d 304, 307 (CA 2 1963).
26
Citing Sony Corp v Universal City Studios, 464 US [417] 434 (1984).
27
Citing Kalem Co v Harper Brothers, 222 US 55, 62–6 3 (1911); Gershwin n 24, at 1162.
28
Sony v Universal City Studios, 464 US [417,] 434 (1984). For its UK equivalent, see CBS Songs v Amstrad
Consumer Electronics [1988] AC 1013.
29
Grokster n 20, at 933–934.
738
30
Ibid at 937. 31
Napster, n 18, at 1027.
32
See UMG Recordings v Veoh Networks, Inc, 665 F Supp 2d 1099, 1116–18 (CD Cal 2009). Note that Universal
v Reimerdes 273 F 3d 429 (2nd Cir 2001) affirmed DMCA anti-c ircumvention provisions which made it illegal to
reverse engineer or ‘hack’ (a term of art for computer scientists) Digital Rights Management systems.
33
See Varanini, E, ‘National Report: United States, LIDC Congress in Oxford’ 22–2 4 September 2011, at
<http://w ww.ligue.org/documents/2011rapportBUS.pdf>.
34
See Memorandum Of Understanding (2011) dated 7 June 2011, at <http://w ww.copyrightinformation.
org/sites/default/fi les/Momorandum%20of%20Understanding.pdf>, and for a critique see Goldman, E, and
McSherry, C, ‘The “Graduated Response” Deal: What if Users Had Been At the Table?’ Electronic Frontier
Foundation at <https://w ww.eff.org/deeplinks/2011/07/g raduated-response-deal-w hat-i f-u sers-had-been>
18 July 2011.
35
See <http://w ww.copyrightinformation.org/news/releases>, note that there was no update to the website
for the six months (to January 2012) following its first publication.
36
Dowling v United States, 473 US 207, 222 (1985).
793
following the January 2012 arrest of Megaupload executives in New Zealand has
caused some surprise and uncertainty in the application of criminal law,37 as it
follows a 2005 restatement of enforcement policy.38 The ‘wilful’ requirement in
criminal law must be proved beyond reasonable doubt.39 Nevertheless, a more
aggressive prosecution of counterfeiting and other ‘piracy’ (sic) websites was sig-
nalled with the taking down of domain names belonging to suspected overseas
‘rogue sites’.40 The cooperation of several national police forces in the Megaupload
case indicates a more general trend towards aggressive policing of counterfeiting
which continues.
In European debate, the 1999 Commission proposals for the 2002 communica-
tions regulation package41 and the 1997 Bonn Ministerial Conference Declaration
expressed the desire for end-user filtering rather than intermediary liability.42 This
policy led to the E-Commerce Directive (ECD) of 2000, which enshrined this prin-
ciple of the internet host ‘safe harbour’ of non-l iability and leaving much detailed
regulation via codes of conduct to the market actors.43 Benoit and Frydman estab-
lish that it was based on the 1997 German Teleservices Act,44 though with ‘slightly
more burden on the ISPs in comparison with the former German statute’.45
37
See Department of Justice Office of Public Affairs (19 January 2012) Justice Department Charges Leaders
of Megaupload with Widespread Online Copyright Infringement.
38
US ATTORNEY BULLETIN (2005) Novel Criminal Copyright Infringement Issues Related To The Internet
available at <http://w ww.cybercrime.gov/u samay2001_ 5.htm>.
39
See US Attorney Prosecuting IP Crimes Manual, Criminal Copyright Infringement Issues, §II.B.2 (2006),
available at <http://w ww.cybercrime.gov/ipmanual/02ipma.html#II.B.2.a>.
40
See Affidavit in Support of Application for Seizure Warrant Pursuant to 18 USC §§2323, 981, United States
v Domain Names (defendants in rem), (31 January 2011) (No 18 MAG 262).
41
See COM(1999) 539.
42
Bonn Ministerial Declaration (1997) Europe paves the way for rapid growth of Global Information
Networks, 8 July, copy located at <http://web.mclink.it/MC8216/netmark/attach/bonn_en.htm#Heading01>.
43
See Art 1(1)(2), Art 16 and Recitals 32, 49, of 2000/31/EC.
44
German Teleservices Act 1997 (TDG) (‘Federal Act Establishing the General Conditions for Information
and Communication Services’ of 22 July 1997, BGBl. I S. 1870).
45
Frydman and Rorive n 16, at 54.
470
that on trans-f rontier television,46 legal certainty offered by the application of the
‘country of origin’ principle (Internal Market clause) has been introduced, to-
gether with provisions concerning establishment, information and transparency
requirements, provisions concerning commercial communications, electronic
contracting, and the regulation of the liability of intermediary service providers
(which is the primary concern of this chapter). The Internal Market clause is the
core element of the ECD. It takes the form of two complementary provisions:
• Member States must ensure that a provider of information society services es-
tablished on its territory applies with the national provisions applicable which
fall within the coordinated field.47
• In turn Member States may not, for reasons falling within the coordinated
field, restrict the freedom to provide information society services from another
Member State.48
[a] the provider does not have actual knowledge of illegal activity or information
and, as regards claims for damages, is not aware of facts or circumstances from
which the illegal activity is apparent; or
[b] the provider, upon obtaining such knowledge or awareness, acts expeditiously
to remove or to disrupt access of the information.
The key provision here is the establishment of the concept of ‘actual knowledge’.
ECD, though maintaining CDA/DMCA mere conduit principles, limits them sub-
stantially, because when an ISP ‘obtains actual knowledge’ of a site containing in-
fringement it must act ‘expeditiously to remove or disable access to the information
concerned’. Whereas in some cases it might be easy to define what ‘actual know-
ledge’ means, when an ISP receives a notice from a hotline it may simply treat the
complaint as actual knowledge and remove the content. This defers responsibility
for judgment to ‘hotlines’, which might be better trained for such an investigation.
What constitutes actual knowledge remains undefined, including for instance
whether it must be a letter with proof of the identity of the complainant. The term
‘awareness’ seems even vaguer. Article 14 establishes the concept of ‘apparent’ il-
legal content, which the ISP needs to remove expeditiously, if made aware. Article
5(1) of the Copyright Directive drafted a year later establishes an exemption from
liability for ‘Temporary acts of reproduction . . . transient or incidental [and] an in-
tegral and essential part of a technological process and whose sole purpose is to
enable a transmission in a network between third parties by an intermediary’.51
49
Case C-324/0 9 L’Oréal SA & Others v eBay International AG & Others, decided 12 July 2011.
50
SEC(2011) 1641 at p 25.
51
Directive 2001/29/EC of 22 May 2001, OJ L 167, 22 June 2001 P. 0010—0 019.
472
Article 21.2 provides that, when the Directive is re-examined, the issues to be
analysed include NTD procedures and attribution of liability following content
removal.
Rightswatch An IPR hotline and clearing house which would mirror the
European hotlines dealing with some illegal materials, notably child abuse im-
ages (see Section 15.6) was proposed prior to the ECD drafting. RightsWatch was a
multistakeholder project funded by the European Commission in the period from
1999–200252 to standardize the NTD procedure in six steps: location, notification,
verification, information, take down, and confirmation. It failed to gain support
because of fundamental differences of approach between rights holders and ISPs.
To illustrate NTD in action, consider the Code of Practice of the UK ISP
Association (ISPA), which mentions the complaint procedure, but does not dir-
ectly refer to the NTD procedure. All major ISPs in the UK have agreed to ‘use their
reasonable endeavours to resolve a complaint within 10 working days of receipt of
notice be it by email, letter, telephone call or in person’.53 The Code lays out a min-
imum liability NTD guideline:
ISPA UK supports its Members in any independent decision taken by the Member
to proactively limit the accessibility of illegal material via its service, but strongly
states that no greater legal burden, standard of care or obligation should be placed
on the Member who takes such action than is placed upon those Members who do
not take such action.54
The wider European ISSP legal reaction to the failure of hotline initiatives in IPR
cases was similar, and cooperation with rights holders would in general only occur
under court order. Section 15.5 explains the range of enforcement powers in effect
in Europe and the issue of ‘graduated response’ in European law, before Section
15.6 goes on to consider the NTD arrangements which have been introduced for
illegal criminal content, notably child sexual abuse images.
52
Rightswatch Project IST-1999-10639 funded under the 5th Framework Programme (total 1.3 million
euros) at <http://cordis.europa.eu/ist/t rust-security/projects_ f p5.htm>.
53
The ISPA UK Code of Practice Statement of Policy, as amended 2007, at <http://w ww.ispa.org.uk/about_
us/page_16.html>.
54
ISPA (2007) n 53.
473
• It does not provide an exemption from liability, if the ISP acts according to a
clearly defined procedure. This would remove the burden of investigation and
judgment of the ISP, and transfer it to the parties involved, the complainant and
the content provider.
• It does not create an incentive for the ISP to properly investigate whether content
is illegal, but rather to remove the content expeditiously.
The lack of standard European NTD procedure poses several problems. ISPs are
not able to know whether they are properly informed, whether the information
(complaint) received is correct (founded) or not, and whether they can face liability
claims by users when their pages have been shut down,61 and it is established ex
55
Digital Millennium Copyright Act 1998, s 512(i)(1)(A). 56
Ibid, s 512(g). 57
Ibid, s 512(g)(2).
58
Ibid, s 512(g)(2)(c). 59
Ibid, s 512(f).
60
ALS Scan v Remarq Communities, Inc, 239 F 3d 619 (4th Cir 2001).
61
In this regard, the DMCA provides that ‘a service provider shall not be liable to any person for any claim
based on the service provider’s good faith disabling of access to, or removal of, material or activity claimed
to be infringing or based on facts or circumstances from which infringing activity is apparent, regardless of
whether the material or activity is ultimately determined to be infringed.’ Digital Millennium Copyright Act
1998, n 6, at (g)(1).
47
post that the content was neither illegal nor harmful. Consequently there is a po-
tential shortcoming in the protection of freedom of expression, as pointed out by
the United Nations special rapporteur.62
15.3.4 Revision of ECD
The manifest faults with the ECD, not least associated with the scope of ISSPs and
the lack of a uniform NTD process, have not resulted in rapid revision, and its re-
form has been continually postponed. Article 21 ECD requires the Commission
to submit to the European Parliament and the Council every two years a report
on the application of ECD. The first review in 2003 was far too early to produce
any but initial indications of the ECD’s adoption within Member States (and its
non-adoption in France at that point). A 2007 study was charged with the ambi-
tious task of working out how effective ECD was. In November 2010, the European
Commission concluded a thorough consultation on the ECD in which it invited
62
La Rue, F, Report of the Special Rapporteur on the promotion and protection of the right to freedom of
opinion and expression, Human Rights Council Seventeenth session Agenda item 3, A/H RC/17/27 of 17 May
2011 at <http://w ww2.ohchr.org/english/bodies/h rcouncil/docs/17session/A .HRC.17.27_en.pdf>.
63
See on hyperlinking, Shetland Times, Ltd v Jonathan Wills and Another, 1997 FSR (Ct Sess. OH), 24
October 1996. On application of the Defamation Act 1996, s 1, see Godfrey v Demon Internet Service [2001] QB
201, which also provides guidance on take-down. Bunt v Tilley & Ors [2007] 1 WLR 1243, per Eady J at para 37.
64
See the Serious Crime Act 2007, s 57(5–8).
65
Formerly Phone Pay Plus, renamed in 2016, and previously known since 1986 as ICSTIS. See Phonepay
Plus, ‘UK regulator PhonepayPlus to rename as Phone-paid Services Authority’, 12 July 2016 at <http://psauthority.
org.uk/news-and-events/news/2016/july/uk-regulator-phonepayplus-to-rename-as-phone-paid-services-
authority>.
475
comments on, among other matters, the interpretation of the provisions con-
cerning intermediary liability. The Commission published a Communication on
E-Commerce, including the impact of ECD, in January 2012.66 In the subsequent
five years, no further attempts were made through hard law and legislation to
amend the ECD, and the ‘Gordian knot’ of intermediary liability (with the excep-
tion of copyright infringement) remains in place.
Service providers can receive limited liability according to the activities in which
they engage, not the type of service provider their overall business accounts for,
as confirmed by the CJEU in 2010: Google v LVMH.67 The following section of this
chapter considers the enforcement cases, which give guidance to national courts
and legislators that was unfortunately missing in the 2008 CJEU decision in
Promusicae, which simply and very unhelpfully instructed the Spanish court to
observe a ‘fair balance to be struck between the various fundamental rights pro-
tected by the Community legal order’,68 in particular between ECD and various
copyright and privacy rights.
This section proceeds by considering L’Oréal v eBay, and Scarlet Extended, con-
trasted with a series of UK cases of particular interest involving blocking access
to the Newzbin site, which then led to the blocking of live streaming of Premier
League football games in FA Premier League v BT (2017).69
66
COM(2011) 942 A coherent framework for building trust in the Digital Single Market for e-commerce
and online services, at <http://ec.europa.eu/i nternal_ market/e-commerce/docs/communication2012/
COM2011_ 942_en.pdf>. This document was published 11 January 2012. Note also COM(2011) 287 final
Communication from the Commission ‘A Single Market for Intellectual Property Rights—Boosting creativity
and innovation to provide economic growth, high quality jobs and first class products and services in Europe’.
67
Joined cases C-2 36/0 8 to C-2 38/0 8, Google v LVMH, judgment of 23 March 2010, para 113.
68
Case C-275/0 6 Productores de Música de España (Promusicae) v Telefónica de España SAU, judgment of 29
January 2008 [2008] ECR I271 at para 70.
69
Football Association Premier League Ltd v British Telecommunications Plc and others [2017] EWHC
480 (Ch).
70
See further Headdon, Toby ‘Beyond Liability: Injunctions after L’Oreal v eBay, Computers and Law’,
(2011) at <www.blplaw.com/media/pdfs/News%20and%20Views/Headdon.pdf>.
476
• ISSPs have a contractual relationship with the infringing user and can enforce
those terms, which may preclude any use of its service which infringes third-
party rights;
• ISSPs are in a position to ‘take down’ infringing content or even terminate
infringing users;
• It is more cost-effective to pursue a single intermediary rather than all users.
In this case, the CJEU was presented with the question of clarifying how to
balance ISP and user rights in the ECD and data protection law,71 against rights
holders’ need to enforce their IPR under both the 2001 Copyright and 2004
Enforcement Directives. Article 8(3) Copyright Directive72 provides: ‘Member
States shall ensure that rightholders are in a position to apply for an injunction
against intermediaries whose services are used by a third party to infringe a copy-
right or related right.’ Recital 59 Copyright Directive states:
In the digital environment, in particular, the services of intermediaries may in-
creasingly be used by third parties for infringing activities. In many cases such
intermediaries are best placed to bring such infringing activities to an end.
Therefore, without prejudice to any other sanctions and remedies available,
rightholders should have the possibility of applying for an injunction against an
intermediary who carries a third party’s infringement of a protected work or other
subject-matter in a network. This possibility should be available even where the
acts carried out by the intermediary are exempted under Article 5. The conditions
and modalities relating to such injunctions should be left to the national law of the
Member States.
71
Particularly Directive 95/4 6/EC of 24 October 1995, OJ L 281/31 (OJ L 281, 23 November 1995 P. 0031—
00501995); Directive 2002/58/EC, OJ L 201/37 (OJ L 201, 31 July 2002 P. 0037—0 0472002).
72
Directive 2001/29/EC.
73
Directive 2004/4 8/EC of 29 April 2004 on the enforcement of intellectual property rights, OJ L 195, 2 June
2004 P. 0016—0 025.
47
and shall be applied in such a manner as to avoid the creation of barriers to legit-
imate trade and to provide for safeguards against their abuse.’74 Recital 24 goes on
to state that these may include ‘prohibitory measures’ aimed at preventing further
infringements of intellectual property rights and ‘corrective measures’ such as re-
calling, removing from distribution, or destroying infringing goods and materials
used to create them. Article 2(3) of the Directive makes it clear that it ‘shall not af-
fect’ the provisions of ECD.
The CJEU in its L’Oréal ruling explained that injunctions may be available
against an intermediary to prevent not only the continuation of a specific act of
infringement but also ‘the repetition of the same or a similar infringement in the
future, where such injunctions are available under national law’. An appropriate
limit may be to impose a ‘double requirement of identity’: future infringements
would have to be committed by the same third party and in respect of the exact
same IPR (the Commission described such a requirement as a ‘Notice & stay down’
obligation75). However, on what national courts must do, the CJEU stated that ‘con-
ditions and procedures’ are the responsibility of Member States’ courts. It did state
that the scope of the injunction should follow Promusicae but also be determined
by feasibility, cost, proportionality of potential measures.76 Technical measures
that could be taken might include blocking a particular infringing user’s access
to a service or utilizing content filtering applications and web proxy servers to re-
move infringing content.
74
Recital 6 and Article 1 of the Enforcement Directive offer a limited explanation of what is covered.
Paragraph 2.2 of the July 2011 ‘Synthesis of Comments on the Commission Report on the Application of [the
Enforcement Directive]’ reported on feedback regarding the introduction of a pre-defined list of intellectual
property rights covered by the Directive.
75
See Question 57 in European Commission (2010) Public consultation on the future of electronic com-
merce in the internal market and the implementation of the Directive on electronic commerce (2000/31/
EC), at <http://ec.europa.eu/i nternal_ market/consultations/docs/2010/e-commerce/questionnaire_%20e-
commerce_en.pdf>.
76
Article 2(3) of the Enforcement Directive requires that measures are ‘effective, proportionate and dissua-
sive’. Article 3 of the Enforcement Directive requires that measures be fair and equitable, not unnecessarily
complicated or costly, or entail unreasonable time-l imits or unwarranted delays. There can be no general ob-
ligation to monitor, as in Article 15 ECD. Injunctions must not create barriers to legitimate trade. Injunctions
must strike a fair balance following Promusicae which requires Member States to balance competing interests.
478
facts of the case itself: the Belgian court ordered an ISP to filter all traffic for copy-
right infringement and pay for the privilege of enforcing for copyright!
Paragraphs 43–4 4 in Scarlet Extended are critical for general guidance:
The protection of the right to intellectual property is indeed enshrined in Article
17(2) of the Charter of Fundamental Rights of the European Union (‘the Charter’).
There is, however, nothing whatsoever in the wording of that provision or in the
Court’s case-law to suggest that that right is inviolable and must for that reason be
absolutely protected.
[44] As paragraphs 62 to 68 of the judgment in Promusicae make clear, the pro-
tection of the fundamental right to property, which includes the rights linked to
intellectual property, must be balanced against the protection of other funda-
mental rights.
77
Case C-360/10 SABAM v Netlog, decided 16 February 2012, reported at <http://c uria.europa.eu/juris/
document/document.jsf?text&docid=119512&pageIndex=0&doclang=EN&mode=req&dir&occ=first&part=1
&cid=158253>.
78
See Scarlet n 11, at para 52: ‘injunction could potentially undermine freedom of information since that
system might not distinguish adequately between unlawful content and lawful content, with the result that
its introduction could lead to the blocking of lawful communications’. See further SABAM v Netlog n 77, at
paragraphs 36–38.
79
Case 582/14, Patrick Breyer v Germany, 19 October 2016, ECLI:EU:C:2016:779.
479
system used in Newzbin2 was already in place and cheap, and BT’s estimate of the
cost was so low that it was essentially negligible.
15.4.3 UK—Newzbin
There were three judgments in the Newzbin saga which confirmed ECSPs must
cooperate in blocking access to infringing websites. First came the decision of
Kitchen J that resulted in an order against Newzbin Ltd directly in 2010,80 then
the judgment of Arnold J in July 2011 (BT/Newzbin No 1) to grant an order against
BT to block Newzbin access to its subscribers,81 followed by the detail of the order
on 21 October 2011 (BT/Newzbin No 2).82 This was the first order in the UK under
Section 97A of the Copyright Designs and Patents Act 1988, an amendment made
in 2003.83 Section 97A of the Copyright Designs and Patents Act 1988 implements
Article 8.3 of Directive 2001/29/EC, though it was already possible under UK law
to seek injunctions against intermediaries, to notify an intermediary of an injunc-
tion served on an infringer, and an intermediary was liable for contempt of court if
it aids and abets an infringer.
The Newzbin website indexed (but did not store) USENET content, with sub-
scribers offered normal (free) and premium membership. Premium members
were able to obtain easy access to films available in binary format on USENET—
in most cases with infringement of copyright. Six film studios sued Newzbin for
damages and an injunction to prevent it ‘from including in its indices or databases
entries identifying any material posted to or distributed through any Usenet group
in infringement of copyright.’ On the basis of a number of factors, Kitchen J found
that Newzbin were secondary copyright infringers. The injunction granted by
Kitchen J required Newzbin to filter any material that matched the studios’ copy-
right works in a database that the studios supplied. The studios wanted a much
wider injunction, that covered all infringing material, whether or not it belonged
to them. The judge refused to permit film studios to use s 97A against ISPs—t hough
using a new power for the courts to grant injunctions against service providers
(like Newzbin) that had actual knowledge of infringement.84 A wider injunction
80
Twentieth Century Fox Film Corporation v Newzbin Ltd [2010] EWHC 608 (Ch) <http://w ww.bailii.org/ew/
cases/E WHC/C h/2010/6 08.html>.
81
Twentieth Century Fox Film Corporation v British Telecommunications Plc (No 1) [2011] EWHC 1981 (Ch)
28 July 2011.
82
Twentieth Century Fox Film Corporation and Others v British Telecommunications Plc (No 2) [2011] EWHC
2714 (Ch) 26 October 2011.
83
As amended by the Copyright and Related Rights Regulations 2003, reg 27(1). This implemented the
Copyright Directive 2001/29/EC, Art 8(3).
84
Newzbin Ltd (2010) at para 135.
570
would be uncertain in scope, and the judge was clear that the injunction should go
no further than he had indicated.
The rights holders pursued an alternative strategy as Newzbin moved out of jur-
isdiction to Sweden and the order could not be satisfactorily enforced, using eBay
v L’Oréal to persuade Arnold J to rule that BT Retail, the UK’s biggest ISP, should
block Newzbin access using its Cleanfeed technology. As explored further in
Section 15.6 below, BT’s Cleanfeed technology is a two-stage system of IP address
re-routing and deep packet inspection (DPI)85 based URL blocking. The relevant
part of the injunction is as follows,
The technology to be adopted is: (i) IP address blocking in respect of each and
every IP address from which the said website operates or is available and which is
notified in writing . . . (ii) DPI based blocking utilising at least summary analysis
in respect of each and every URL available at the said website and its domains and
sub domains and which is notified in writing . . .
2. For the avoidance of doubt paragraph 1(i) and (ii) is complied with if the
Respondent uses the system known as Cleanfeed and does not require the
Respondent to adopt DPI based blocking utilising detailed analysis.
Arnold J limited the injunction to, ‘all BT’s services which incorporate Cleanfeed
whether that is imposed on the customer or taken as an option’. The judge
agreed that it was inappropriate to order BT’s access and upstream divisions (BT
Wholesale, BT OpenReach) to block Newzbin. Arnold J also explained he added
words, ‘any other IP address or URL whose sole or predominant purpose is to en-
able or facilitate access to the Newzbin2 website’ because Newzbin had offered
anti-blocking software to users. To prevent the downloading of such software, film
companies asked for an order to permit them to notify additional IP addresses and
URLs to BT, to block those sites too. The film companies asked for the order to refer
to sites whose predominant purpose is to facilitate access to Newzbin while BT
preferred that an order should be restricted to sites whose sole purpose was to en-
able or facilitate such access. The judge sided with the film companies: the order
could otherwise be circumvented by Newzbin’s posting a weather forecast to the
page from where anti-blocking software could be downloaded.
BT argued that film companies should pay costs of the order, using the prece-
dent of Norwich Pharmacal, in which case the House of Lords held costs should
be borne by the party seeking disclosure.86 The judge rejected BT’s argument at
paragraph 32:
85
See Bendrath, R and Mueller, M, ‘The End of the Net as We Know It? Deep Packet Inspection and Internet
Governance’, (2010) at <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1653259>.
86
Norwich Pharmacal Co v Customs and Excise Commissioners [1973] UKHL 6, [1974] AC 133.
715
The Studios are enforcing their legal and proprietary rights . . . specifically their
right to relief under Article 8(3). BT is a commercial enterprise which makes a profit
from the provision of the services which the operators and users of Newzbin2 use
to infringe the Studios’ copyright. As such, the costs of implementing the order
can be regarded as a cost of carrying on that business.
87
L’Oréal v eBay [2009] EWHC 1094 (Ch), paras 426–431; Case C-324/0 9 L’Oréal v eBay, Advocate General’s
Opinion of 9 December 2010, paras 119–120.
88
Opinion of Advocate General Villalon dated 14 April 2011 in Case C-70/10 Scarlet Extended.
752
infringe copyright. Between 2010 and 2011, judicial opinion shifted from Kitchin J
who did not see fit to order ISPs to block Newzbin because the scope of such an
order would be very uncertain, to Arnold J at paragraph 182 in Newzbin (No 2) who
viewed his Order as ‘very different’ and did not ‘suffer from that vice’.
As a result, at a local level the UK rights holders requested BT to block access to
the file sharing website Pirate Bay,89 and several other major retail ISPs complied
with the Newzbin judgment through the winter of 2011–12, further blocking ac-
cess to The Pirate Bay following judgment on 2 May in Dramatic Entertainment
Limited v British Sky Broadcasting Limited.90 Graham Smith had presciently ar-
gued that ‘Newzbin2 is a first step in a long game, Scarlet has limited how far that
can go’, and that ‘The discussion in Newzbin2 was entirely domestic, predicated
on infringements of UK copyright. It now appears that future cases will have to
take cross-border issues into account, at least where an ISP has customers in more
than one EU country.’91
Other EU countries have also seen successful applications for injunctions under
Article 8(3).92 More generally, the European Commission has recognized that the
position is becoming very uncertain for ISPs, and has agreed in the course of 2012–
13 to review the ECD particularly in the case of NTD procedures. It issued a 2013
working paper on its progress,93 since superseded by the 6 May 2015 proposals on
the Digital Single Market Strategy, which may result in legislative action from 2018
onwards.94
89
’Arts Group tell BT to block access to The Pirate Bay’, BBC, 4 November 2011.
90
Dramatic Entertainment Limited v British Sky Broadcasting Limited [2012] EWHC 1152 (Ch) 2 May
at: <http://w ww.bailii.org/e w/c ases/E WHC/C h/2012/1152.html>.
91
Smith, G, ‘SABAM/ Scarlet meets Newzbin2— but will they play nicely together?’ Cyberleagle, 28
November 2011.
92
Court of Appeal of Antwerp Case 2010/A R/2541 VZW Belgian Anti-Piracy Federation v NV Telenet 26
September 2011.
93
Commission staff working document, Report on the implementation of the e-commerce action plan (23
April 2013).
94
European Commission (2016) Proposal for a Regulation of the European Parliament and of the Council
on addressing geo-blocking and other forms of discrimination based on customers’ nationality, place of resi-
dence or place of establishment within the internal market and amending Regulation (EC) No 2006/2004 and
Directive 2009/22/EC, 25 May.
753
‘Graduated response’ seeks ISP cooperation (through legislation, court order, but
preferably voluntary cooperation) in the following cases:
The backdrop to the 2009 European Parliament net neutrality debates was a
series of laws aggressively aimed at individual internet users and lobbied for by
copyright holders, notably the music and film producers and distributors lobby.
In Sweden, this included the sensational case in which the founders of Pirate Bay,
a website that assisted file sharers, were successfully criminally prosecuted for
aiding and abetting copyright infringement.95 Solutions to the copyright holders’
dilemma were extended nationally to include:
The following sections explain the legislative approach adopted in France and
the United Kingdom.
1. HADOPI, the body created for this purpose, will send the infringer a
warning email.
2. If the infringement is repeated within six months, a new email is sent together
with a warning by registered letter.
95
See Case B13301-0 6 decided 17 April 2009, which was appealed to the Swedish Supreme Court and re-
jected on 1 February 2012, confirming the custodial sentences as reduced in the Court of Appeal. See Torrent
Freak, ‘Pirate Bay Founders’ Prison Sentences Final, Supreme Court Appeal Rejected’, 1 February 2012 at
<http://t orrentfreak.com/p irate-b ay-f ounders-p rison-s entences-f inal-s upreme- c ourt-a ppeal-r ejected-
120201/>.
96
Meyer, T, (2017) ‘HADOPI: 2009 Graduated Response in France’ in Meyer, T (ed), The Politics of Online
Copyright Enforcement in the EU: Access and Control (Springer Verlag, 2017), 165–204 DOI 10.1007/978-3-319-
50974-7_4.
574
3. If the infringement is repeated within a year, the internet user is penalized ac-
cording to the gravity of the act.
The sanction can be the denial of internet access ranging from one month to a
year during which time the internet user must continue to pay the ISP subscrip-
tion and is included on a black list that forbids her to subscribe to any other ISP.
The passage of the ‘HADOPI law’97 on 13 May 2009 was immediately referred by
60 members of the French parliament to the Constitutional Court, which on 10
June struck down the elements of the law which create a punishment prior to ju-
dicial instruction.98 The Court took a rights-based approach, as one would expect
of a constitutional court, forcing a ruling of the judicial authorities, in accordance
with Article 11 of the Charter of Fundamental Rights, now made European law by
the Treaty of Lisbon from 1 December 2009. The HADOPI itself was suspended in
July 2013, as a result of political pressure from the minister from August 2012, and
a consultation which concluded that the state would not allow HADOPI to pursue
suspected infringers in such a manner.99
97
TA No 81 (2009) Act to promote the dissemination and protection of creation on the Internet [‘HADOPI
Law’] creating the High Authority for the dissemination of works and protection of rights on the internet
(HADOPI).
98
Conseil Constitutionnel (2009) Decision No 2009-580 DC of 10 June 2009.
99
Décret n° 2013-596 du 8 juillet 2013 supprimant la peine contraventionnelle complémentaire de suspen-
sion de l’accès à un service de communication au public en ligne et relatif aux modalités de transmission des
informations prévue à l’article L. 331-21 du code de la propriété intellectuelle, at <https://w ww.legifrance.
gouv.fr/a ffichTexte.do;jsessionid=?cidTexte=JORFTEXT000027678782&dateTexte=&oldAction=rechJO&cate
gorieLien=id>.
100
Note the powers for rightsholders compared with CDPA s 97(A): 97A requires that the service provider’s
service ‘is being used’ to infringe copyright (present tense); Section 17 works even if infringement stopped for-
ever and also if there has not yet been any infringement but it is likely to happen. Section 97A requires a service
75
This potential Section 17 power (as well as the overall graduated response
scheme in Sections 9–18) was challenged by BT and TalkTalk. They argued that
under ECD the role of ISPs was ‘passive’ as mere conduits. They also argued that
IP addresses of their users are personal data. The High Court at first instance dis-
missed most of the ISP claims, but chose to move from a 50/50 cost split to 75/25 in
favour of ISPs under the scheme.101 Arguably, this helps make it ‘Scarlet Extended
proof’ as it ensures costs are more reasonably allocated on the rightsholders. BT/
TalkTalk successfully appealed the first instance judgment in October 2011 on all
grounds except Article 15 ECD but lost their appeal in the Court of Appeal on 6
March 2012.102
Ofcom spent several years consulting with the ECSPs and copyright holders over
its Initial Obligations Code, first published in 2010 then revised in 2012. Its costs
and procedures for section 11 made it a lightweight UK version of HADOPI (the
latter had a €12m annual budget and sent a million notices to subscribers warning
of potential infringement and subsequent disconnection, though the latter was
not implemented). A draft revised Code was laid before Parliament in 2011,103 and
extensive consultation took place in 2011–12104. Since 2012, Ofcom—partly funded
by the Intellectual Property Office—has also focused on measuring the changes in
privacy and content consumption, in particular the move towards streaming ra-
ther than peer-to-peer downloading.105 The proposed appeals panel which would
have regulated disconnections was not in fact implemented as the Secretary of
State has not tabled the secondary legislation necessary.106 The perceived failure of
HADOPI in 2012–13 must have contributed to this inactivity.
provider to know about infringement, s 17 does not. Under s 17, a web location blocked may not be guilty of
copyright infringement, but may merely be used to facilitate access to locations that are. Section 97A may not
be used to prevent access to sites that are not themselves infringing. Section 17 DEA is thus much wider than
s 97A—i f implemented.
101
British Telecommunications Plc R (on the application of) v Secretary of State for Business, Innovation and
Skills [2011] EWHC 1021 (Admin).
102
BT Plc and Another, R (on the application of) v Secretary of State for Business, Innovation and Skills [2011]
EWCA Civ 1229 (7 Oct), Court of Appeal decision of 6 March in R v Secretary of State for Culture, Olympics,
Media and Sport [2012] EWCA Civ 232 at <http://w ww.bailii.org/e w/c ases/E WCA/Civ/2012/2 32.html>.
103
Draft Online Infringement of Copyright (Initial Obligations) (Sharing of Costs) Order 2011 (draft Costs
Order) January 2011, <https://w ww.legislation.gov.uk/u kdsi/2011/9780111505779>—it continues to be draft
and may never be implemented.
104
Government would have to notify the Revised Code in draft to the European Commission under the
Technical Standards Directive (98/3 4/EC).
105
IDATE (2015) Online Content Study: Changes in the distribution, discovery and consumption of lawful
and unauthorised online content MC 359 (Lot 3 Label 9 and Label 13) Final report for OFCOM, November 2015,
at <https://w ww.ofcom.org.uk/_ _data/a ssets/pdf_ fi le/0 033/99483/online-content-study-010316.pdf>.
106
Lloyd, I (2014) Information Technology Law (4th edn, Oxford University Press, 2014), 358–362 sets out the
procedure, see also Ofcom (2010) Implementing the online piracy provisions of the Digital Economy Act, pres-
entation to E-Commerce Directive Expert Group, 23 September at <http://ec.europa.eu/i nternal_market/e-
commerce/docs/expert/20100923-presentation-dea_en.pdf>.
576
ISPs and their content hosting partners are the key link between content and end-
users and hence their position as the only feasible point of control for governments
who wish to block or filter content that is considered unsuitable for end-users. ISPs
required to block access to specific websites have therefore relied on three mechan-
isms: IP address filtering, DNS poisoning, and keyword searching.107 A fourth possi-
bility is to use a hybrid which combines two or more of these systems. All rely on one
or a combination of:
107
Brown, I, ‘Internet filtering—be careful what you ask for’, in Kirca, S and Hanson, L, (eds.), Freedom and
Prejudice: Approaches to Media and Culture, (Istanbul: Bahcesehir University Press, 2008) pp 74–91.
108
European Council Recommendation of 24 September 1998 on the development of the competitiveness of the
European audiovisual and information services industry by promoting national frameworks aimed at achieving a
comparable and effective level of protection of minors and human dignity, at OJ L 270, 7 October 1998, p 48.
109
Decision 1151/2003/EC of 16 June 2003, amending Decision 276/1999/EC adopting a Multiannual
Community Action Plan on promoting safer use of the Internet by combating illegal and harmful content
on global networks. See Safer Internet programme, with resources of €55m in the period 2009–13, following
earlier programmes from 1999–2002, 2002–4, and 2005–8, at <http://ec.europa.eu/i nformation_society/activ-
ities/sip/policy/programme/i ndex_en.htm>.
75
110
Nash, Richard, INSAFE Newsletter, September 2007, <http://w ww.saferinternet.org/w w/en/pub/i nsafe/
news/newsletter/newsltr_a rchives/2007_september.htm>.
111
EuroISPA-I NHOPE, ‘EuroISPA and INHOPE lay foundations for cooperation’, 28 January 2004.
112
Brown (2008), above n 107, at p 75. See further van Eijk, NANM, et al, ‘Moving Towards Balance: A study
into duties of care on the Internet’, 2010, University of Amsterdam/WODC (Research and Documentation
Centre of the Ministry of Security and Justice), at <http://w ww.ivir.nl/publications/v aneijk/Moving_
Towards_Balance.pdf>.
113
Norwegian Parliament study group on internet filtering law proposal at NOU-2007-2: <http://w ww.
regjeringen.no/nb/dep/jd/dok/nouer/2007/nou-2007-2/6.html?id=449809>.
114
Callanan, Cormac and Nikos Frydas (2007) INHOPE Global Internet Trend Report at <http://w ww.
saferinternet.org/w w/en/pub/i nsafe/news/a rticles/0 907/2007_ i nhope.htm>.
115
See CETS No: 185 Council of Europe Convention on Cybercrime of 23 November 2001, entered into force
1 July 2004 at <http://conventions.coe.int/Treaty/E N/Treaties/Html/185.htm>.
578
any written material, any image or any other representation of ideas or theories,
which advocates, promotes or incites hatred, discrimination or violence, against
any individual or group of individuals, based on race, colour, descent or na-
tional or ethnic origin, as well as religion if used as pretext for any of these fac-
tors . . . [and] . . . material which denies, minimises, approves of or justifies crimes
of genocide or crimes against humanity.116
15.6.2.1 Creation of IWF
The Internet Watch Foundation was founded in 1996 as the UK ISP industry hotline
for the removal of sexual abuse content. Note that UK filtering systems do not cur-
rently allow for police intervention, but Scandinavian models do. Hotlines are exem-
plars of user reporting and ISP removal of illegal content. It is claimed as a success in
its core remit: increased number of cases of reported illegal content; removing supply
of such material from UK content hosts, through the hotline reporting mechanism,
while the URL database for the filtering systems is designed to control inadvertent
access to such material; involvement in UK Government and police initiatives; and
international support for INHOPE and coordinated police activities.
As internet use developed, it became apparent that in addition to the positive
aspects of internet use, there was a potential for misuse. In particular, although
child pornography has always existed, the internet was perceived as giving greater
116
Article 6.1, Additional Protocol to the Convention on Cybercrime, concerning the criminalisation of acts
of a racist and xenophobic nature committed through computer systems, Strasbourg, 28 January 2003.
117
For comprehensive analyses, see Marsden, C, ‘Internet Co-regulation’, (Cambridge: Cambridge University
Press, 2011) at Chapter 6, pp 164–197; McIntyre, TJ, ‘Blocking child pornography on the Internet: European Union
developments’, (2010) 24(3) International Review of Law, Computers and Technology 209–221; McIntyre, TJ and
Scott, C, ‘Internet Filtering: Rhetoric, Legitimacy, Accountability and Responsibility’ in Brownsword, R, and Yeung,
K (eds), Regulating Technologies, (Oxford: Hart Publishing, 2008) at <http://ssrn.com/abstract=1103030>; Akdeniz,
Yaman, Internet Child Pornography and the Law: National and International Responses, (Aldershot: Ashgate,
2008); Edwards, Lilian, ‘IWF v Wikipedia and the Rest of the World (except OUT-LAW)’ (2008) panGloss, at <http://
blogscript.blogspot.com/2008/12/iwf-v-wikipedia-and-rest-of-world.html>; Clayton, Richard, ‘Technical aspects
of the censoring of Wikipedia, Light Blue Touchpaper’, 11 December 2008, at <http://www.lightbluetouchpaper.
org/2008/12/11/technical-aspects-of-the-censoring-of-wikipedia/>; Akdeniz, Yaman, ‘Who Watches the
Watchmen? The role of filtering software in Internet content regulation’, in The Media Freedom Internet Cookbook,
(Vienna: Organisation for Security and Cooperation in Europe, 2004) at <http://www.osce.org/publications/rfm/
2004/12/12239_89_en.pdf>.
759
118
<http://w ww.iwf.org.uk/public/page.29.htm>—note The Metropolitan Police sent a letter to all ISPs on
9 August 1996 requesting them to censor Usenet news groups, threatening otherwise the prosecution of ISPs
in relation to illegal material therein.
119
Internet Service Providers Association, LINX, and Safety-Net Foundation (1996) R3—R ating, Reporting,
Responsibility for Child Pornography and Illegal Material on the Internet, at <http://w ww.mit.edu/activities/
safe/labeling/r3.htm>.
120
KPMG Peat Marwick and Denton Hall ‘Review of the Internet Watch Foundation’, (1999) London,
KPMG.
121
A full list of donors is listed at: <http://w ww.iwf.org.uk/f unding/page.64.htm>.
122
The Annual Report is available on the Internet Watch Foundation website: <http://w ww.iwf.org.uk/ac-
countability/a nnual-reports>.
123
Including copying images as well as making originals: see R v Fellows and Arnold (1997) 1 CAR 244, the
Court of Appeal held that computer data that represented the original photograph in another form was ‘a copy
of a photograph’ under Section 7(2) of the 1978 Act. Downloading an indecent photograph from the internet is
therefore ‘making a copy of an indecent photograph’.
124
IWF cooperates with the Home Office in relation to Section 63 of the Criminal Justice and Immigration
Act 2008, see <http://w ww.iwf.org.uk/hotline/t he-laws/c riminally-obscene-adult-content/c riminal-justice-
and-i mmigration-act-2008>. It has always cooperated with police to report illegal material under the Obscene
Publications Act 1959.
670
15.6.2.2 Reform of IWF
In the UK, British Telecom and other large ISPs block customer access to sites that
have been identified as containing child pornography by IWF.131 BT in 2003 devel-
oped a system called ‘Cleanfeed’ which blocks end-users’ access to pages containing
illegal child pornography images.132 ‘Cleanfeed’ was implemented in 2004.133 The
125
Online non-photographic visual depictions of the sexual abuse of children, Sections 62–69 of the Coroners
and Justice Act 2009. IWF stated: ‘Following consultation with our Funding Council of industry members, in
October 2009 the IWF Board informed government of our agreement to fulfil this role from 6 April 2010.’ See
<http://w ww.iwf.org.uk/ hotline/t he-l aws/non-photographic- c hild-s exual-a buse-i mages/c oroners-a nd-
justice-act-2009>.
126
Internet Watch Foundation, ‘Incitement to racial hatred removed from IWF’s remit’, 11 April 2011 at
<http://w ww.iwf.org.uk/about-iwf/news/post/302-i ncitement-to-racial-hatred-removed-f rom-iwfs-remit>.
127
The term used for mobile phone-c aptured videos of assaults on victims posted on the internet.
128
Grooming, peer-to-peer services, and such like fall within the responsibility of the Child Exploitation and
Online Prevention Centre, an affiliate of the Serious Organised Crimes Agency: see <http://www.ceop.gov.uk/>.
129
Section 45 amends the Protection of Children Act 1978. The same amendment applies to the offence of
possessing an indecent photograph or pseudo-photograph of a child at Section 160 of the Criminal Justice Act
1988 (Section 160(4) applies the 1978 Act definition of ‘child’). The 2003 amendments were made necessary in
order to prevent law enforcement personnel falling foul of the law, as downloading paedophile images was an
offence in view of the Court of Appeal decision in R v Jayson [2002] EWCA Crim 683, CA.
130
Crown Prosecution Service and Association of Chief Police Officers, ‘Memorandum of Understanding
Between Crown Prosecution Service and the Association of Chief Police Officers concerning Section 46 Sexual
Offences Act 2003’, 6 October 2004, at <http://www.iwf.org.uk/documents/20041015_mou_final_oct_2004.pdf>.
131
Clayton, R, ‘Failures in a Hybrid Content Blocking System, presented at the Workshop on Privacy
Enhancing Technologies’ 2005, Dubrovnik, at <http://w ww.cl.cam.ac.uk/~rnc1/cleanfeed.pdf>.
132
See Internet Watch Foundation 7 June 2004 announcement at <http://w ww.iwf.org.uk/media/news.
archive-2004.39.htm>.
133
Truman, Nick, ‘The Experience of BT in Online Child Protection’, paper presented at the Effective
Strategies for the Prevention of Child Online Trafficking Pornography and Abuse, 2009, Bahrain, at <http://
www.befreecenter.org/Upload/Conference/papers/BT.ppt>.
716
CAIC list is the database used by ISPs following the example of BT’s ‘Cleanfeed’.134
The CAIC list is the Child Abuse Database Service: a list of URLs which have been
reported to the police so that an ISP can block the website (the website owner does
have a right of appeal but this is uncertain in effect as it is often difficult for the re-
tail ISP to determine whether the wholesale ISP has blocked the site).135 In May 2006
a Home Office minister told Parliament that ISPs would be required to implement
CAIC: ‘If it appears that we are not going to meet our target through co-operation,
we will review the options for stopping UK residents accessing websites on the IWF
list.’136 The Home Office has stated that filters could be extended to other topics such
as the ‘glorification of terrorism’:137 ‘our policy is to pursue a self-regulatory approach
wherever possible. However, our legislation as drafted provides the flexibility to ac-
commodate a change in Government policy should the need ever arise.’138
134
Hutty, Malcolm ‘Cleanfeed: the facts’ 2004, LINX Public Affairs, at <https://publicaffairs.linx.net/news/
?p=154>.
135
Internet Watch Foundation (2008) Child Sexual Abuse Content URL List, at <http://w ww.iwf.org.uk/cor-
porate/page.49.233.htm>.
136
Hansard (2006) Vernon Coaker, Hansard HC vol 446 col 715W (15 May 2006) Written Answers.
137
Terrorism Act 2006, s 21 adds a new s 3(5A) to the Terrorism Act 2000, which specifies that the power
to proscribe under s 3(5)(c) shall encompass the ‘unlawful glorification of the commission or preparation
(whether in the past, in the future, or generally) of acts of terrorism’. Glorification requires the reasonable
expectation that the audience will emulate terrorism in present circumstances (s 3(5B)), and it comprises any
form of praise or celebration (s 3(5C)). See further Home Affairs Select Committee, Roots of violent radical-
isation: Written evidence submitted by Professor Clive Walker, School of Law, University of Leeds (RVR10),
at paras 13–14.
138
Hutty, M, ‘Government sets deadline for universal network-level content blocking’ London Internet
Exchange Public Affairs blog, 17 May 2006, at <http://publicaffairs.linx.net/news/?p=497#more-497>.
139
Directive 2011/92/E U of 13 December 2011 on Combating the Sexual Abuse and Sexual Exploitation of
Children and Child Pornography, and Replacing Council Framework Decision 2004/6 8/J HA, L335/1.
140
An earlier draft was politically highly controversial as it included the mandatory ‘shall’ rather than the
final version’s ‘may’.
672
Recital 47, the Directive explains that ‘The measures undertaken by Member
States in accordance with this Directive in order to remove or, where appropriate,
block websites containing child pornography could be based on various types of
public action, such as legislative, non-legislative, judicial or other.’ This leaves
to Member States the decision whether to adopt legal measures, and notes ‘this
Directive is without prejudice to voluntary action taken by the Internet industry
to prevent the misuse of its services or to any support for such action by Member
States.’ Furthermore it states that, ‘Any such developments must take account
of the rights of the end users and comply with existing legal and judicial pro-
cedures and the European Convention for the Protection of Human Rights and
Fundamental Freedoms and the Charter of Fundamental Rights of the European
Union.’
Legal problems will remain in that governments cannot delegate their human
rights responsibilities for regulating the internet,141 in favour of instructing private
companies to control extreme user speech.142 A leading example in the case of hate
and racist speech is the Code of Conduct signed with leading internet platform
companies such as Google and Facebook in May 2016.143
Hotlines may offer IAPs the ability to filter for content with user consent—
expressed in assent to a change in Terms and Conditions of the contract, for ex-
ample. Where the filter list is not supplied or approved by government, as for
instance with the self-regulatory IWF scheme, this may potentially infringe the
Open Internet Regulation described in the next section.
The four largest UK consumer IAPs in 2013 agreed to install the IWF filter by
default as an ‘unavoidable choice’ for new customers under the auspices of the
Department for Culture’s multi-stakeholder UK Council for Child Internet Safety
(UKCCIS). They signed a Code of Conduct that was formally self-regulatory, yet
brokered by government, specifically the Department for Education.144 By 2015,
two of the four (Sky and TalkTalk) had adopted a system whereby adult filtering
would be by default turned on unless customers expressed a wish to opt out. Note
141
Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on
Human Rights or ECHR).
142
See All Party Parliamentary Communications Group, ‘Can we keep our hands off the net?’ 2009, Final
Report, October at <www.apcomms.org.uk/uploads/apComms_Final_Report.pdf>; Callanan, Cormac,
Gercke, M, De Marco, E, Dries-Z iekenheiner, H, Internet blocking: balancing cybercrime responses in demo-
cratic societies, (Dublin: Aconite Internet Solutions, 2009) at <http://w ww.aconite.com/sites/default/fi les/
Internet_blocking_a nd_Democracy.pdf>.
143
IP-16-1937, ‘European Commission and IT Companies announce Code of Conduct on illegal online hate
speech,’ Brussels, 31 May 2016, at <http://europa.eu/rapid/press-release_I P-16-1937_en.htm>.
144
Department for Education, Home Office Press release: ‘Parents asked if adult websites should be
blocked’, Tim Loughton, and The Rt Hon Lynne Featherstone, 28 June 2012.
673
that network-level opt-out blocking has existed in mobile networks since 2004
under the auspices of the Independent Mobile Classification Body, which in 2013
was taken over by the cinematic film censor British Board of Film Classification
(BBFC).145
In 2017, the Digital Economy Act came into force, requiring under Part 3,
sections 14–21, default adult filters on legal pornography with a verifying
authority, likely to be the BBFC. These provisions came into force on 31 July
2017.146
This is a clear example of how the ‘filter default’ became a slippery slope down
which UK adult internet access has slipped.147 Despite trenchant criticism by in-
dependent child protection experts, this law is to be enforced in 2018.148 IAPs
now have to filter for not only potentially illegal material, but also adult content.
Furthermore, Nominet, the UK domain name authority (formally independent
of government though regulated by the Digital Economy Act 2010149), has been
screening domain registrations for potential ‘trigger words’ in a policy change an-
nounced in 2013 resulting from a report by Lord Macdonald, a key change urged
by IWF since at least 2004.150 Other chapters in this book cover the filters imposed
for terrorist and hate materials.
It is clear that filtering of content has developed from a child protection issue to
a wider range of regulatory prior censorship tools.
Mac Síthigh, D, Medium Law (Routledge, 2017), at 60 and Marsden, n 117, at 139–143.
145
147
Schellekens, M, ‘Liability of Internet Intermediaries: A Slippery Slope?’, (2011) 8:2 SCRIPTed 154
DOI: 10.2966/scrip.080211.154.
148
Kelion, L, ‘Porn ID checks set to start in April 2018’, BBC News, 17 July 2017, at <http://w ww.bbc.co.uk/
news/technology-4 0630582>.
149
Sections 19–21 of the Digital Economy Act 2010 give government reserve powers to replace Nominet
under new s 124O, Communications Act 2003.
150
Nominet, ‘Update on registration policy’, 8 August 2013, 14:50, at <http://registrars.nominet.uk/news/
nominet-a nnouncements/update-on-registration-policy>.
674
typically in the last mile of access for end-users.151 ISPs can discriminate against
all content or against the particular content that they compete with when they are
vertically integrated. Conventional US economic arguments have always been
broadly negative to the concept of net neutrality, preferring the introduction of
tariff-based congestion pricing.152 Hahn and Wallsten explain that net neutrality153
‘usually means that broadband service providers charge consumers only once for
Internet access, don’t favor one content provider over another, and don’t charge
content providers for sending information over broadband lines to end users.’ This
is the focus of the problem: Network owners with vertical integration into content
or alliances have enhanced incentives to require content owners (who may also
be consumers) to pay a toll to use the higher speed networks that they offer to end-
users. Note all major consumer ISPs are vertically integrated to some extent, with
proprietary video, voice, portal and other services.
As network neutrality extends to all consumer ISPs symmetrically, it may not be
subject to competition law assessments of dominance, as abuse of dominance is
not necessarily an accurate analysis of the network neutrality problem, at least in
Europe.154 Dominance is neither a necessary nor sufficient condition for abuse of
the termination monopoly to take place, especially under conditions of misleading
advertising and consumer ignorance of abuses perpetrated by their ISP.155
The US regulator FCC has acted on several network neutrality complaints
(notably those against Madison River in 2005 and Comcast in 2008156) as well
as introducing the principle in part through several merger conditions placed
on dominant ISPs, but delayed its report and order on net neutrality until its
eventual publication in the Federal Register in September 2011, whereupon it
was instantly challenged by various interested parties and would be litigated
in 2014.
Development of European legal implementation of the network neutrality
principles was initially slow, with the European Commission referring much
of the detailed work to the new Body of European Regulators of Electronic
151
I have argued that the real problem lies in the ‘middle mile’ of interconnection, in Marsden, C, Network
Neutrality: Towards a Co-regulatory Solution (London: Bloomsbury Academic, 2010). See further Marsden, C,
Network Neutrality: From Policy to Law to Regulation (Manchester: Manchester University Press, 2017)
152
See David, P, ‘The Evolving Accidental Information Super-H ighway’, (2001) 17(2) Oxford Review of
Economic Policy pp 159–187.
153
Hahn, Robert and Wallsten, Scott, ‘The Economics of Net Neutrality’ (2006) AEI Brookings Joint Center
for Regulatory Studies: Washington, DC at <www.aei-brookings.org/publications/abstract.php?pid=1067>.
154
See Marsden (2010), n 151, at 1.
155
Some authors question the distinction between degrading and prioritizing altogether, as they find that
the latter naturally presupposes the former. See eg Chirico, F, Van der Haar, I, and Larouche, P, ‘Network
Neutrality in the EU’, TILEC Discussion Paper (2007), <http://ssrn.com/abstract=1018326>.
156
Comcast v FCC (2010) No 08-1291, delivered 6 April.
765
157
BEREC (16) 127 BEREC Guidelines on the Implementation by National Regulators of European Net
Neutrality Rules, issued 30 August.
158
Netherlands Department of Economic Affairs, ‘Net Neutrality Guidelines May 15th, for the Authority
for Consumers and Markets (ACM) for the enforcement by ACM of Article 7.4a of the Netherlands
Telecommunications Act 2012’ (2015).
159
See further Marsden (2017), n 151, at Chapter 7.
160
See guidelines at Nkom, ‘Net Neutrality’ (2014) at <http://eng.nkom.no/technical/i nternet/net-neu-
trality/net-neutrality>.
161
Olsen, T, ‘Net Neutrality Activities at BEREC and Nkom, Norwegian Communications Authority’ (2015),
slide 5, at <http://berec.europa.eu/fi les/doc/2015-07-13_0 9_56_ 36_ 3.%20Noruega%20Nkom%20net%20neu-
trality%20-%20Summit%20BEREC-EaPeReg-R EGULATEL-E MERG.pdf>.
162
See The Chilean ‘Law 20.453, which enshrines the principle of net neutrality for consumers and Internet
users’ (2010), at <http://w ww.leychile.cl/Navegar?idNorma=1016570&buscar=NEUTRALIDAD+DE+RED>.
163
See <http://w ww.subtel.gob.cl/i mages/stories/a rticles/subtel/a socfile/10d_0368.pdf>.
164
See <https://www.government.nl/documents/policy-notes/2012/06/07/dutch-telecommunications-act>.
165
Netherlands Department of Economic Affairs, Net Neutrality Guidelines May 15th, for the Authority
for Consumers and Markets (ACM) for the enforcement by ACM of Article 7.4a of the Netherlands
Telecommunications Act 2012 (2015).
166
See <https://w ww.finlex.fi/fi/laki/k aannokset/2014/en20140917>.
167
‘No. 003-02-10/2012-32’, at <http://w ww.uradni-l ist.si/1/content?id=11144>.
168
For updates, see ‘Ministry of Justice’, (2016) at <http://pensando.mj.gov.br/marcocivil/>.
67
169
See Western Union Telegraph Co v Call Publishing Co, 181 US 92, 98 (1901).
170
Noam (1994) at p 435, explaining that: ‘When historically they [infrastructure services] were provided
in the past by private firms, English common law courts often imposed some quasi-public obligations, one
of which one was common carriage. It mandated the provision of service of service to willing customers,
bringing common carriage close to a service obligation to all once it was offered to some.’
171
See Marsden, C, ‘Network Neutrality: A Research Guide’ in Brown, Ian, (ed), Handbook Of Internet
Research (Cheltenham: Edward Elgar, 2012).
67
172
The ‘Internet’ is a network of Autonomous Systems, of which about 40,000 are of a scale that is relevant.
See Haddadi, Hamed et al, Analysis of the Internet’s structural evolution, Technical Report Number 756
Computer Laboratory UCAM-CL-T R-756 ISSN 1476-2986 (2009).
173
See Saltzer, JH, Reed, DP, and Clark, DD, End-to-end arguments in system design, 2 ACM Transactions
On Computer Systems (1984) p 288.
174
Crowcroft, Jon, ‘The Affordance of Asymmetry or a Rendezvous with the Random?’, (2011) Convergence
and Communications Review, 12.
175
Clark, David, ‘Network Neutrality: Words of Power and 800-Pound Gorillas’, (2007) 1 International Journal of
Communication 701–708; Peha, JM, ‘The Benefits and Risks of Mandating Network Neutrality, and the Quest for a
Balanced Policy’, (2007) 1 International Journal of Communication 644, 644–659.
678
‘common carriage’ under the Telecommunications Act, and its alter ego ‘informa-
tion services’.
While issues about potential discrimination by ISPs have been current since at
least 1999,176 the term ‘network (net) neutrality’ was coined by Tim Wu in 2003.177 In
the period since, the debate was dismissed as ‘an American problem due to aban-
donment of network unbundling’ and common carriage. Competition is ‘inter-
modal’ between cable and telecoms, not ‘intra-modal’ between different telecoms
companies using the incumbents’ exchanges to access the ‘Last Mile’.178 Instead
of regulated access to both cable and telecoms networks, there are now less regu-
lated ‘information’ not ‘telecommunications’ services. The Regional Bell Operating
Companies (RBOCs) re-emerged in 2006 as two local long-distance internet-wireless
combines, now called AT&T and Verizon.179
176
See Lemley, MA and Lessig, L, ‘The End of the end-to-end: preserving the architecture of the internet in
the broadband era’, (2000) available at UC Berkeley Law & Econ Research Paper No 2000-19; Stanford Law &
Economics Olin Working Paper No 207; UC Berkeley Public Law Research Paper No 37. See further Chapter 1
in Marsden (2010), n 151, and Marsden, ‘Pluralism in the multi-channel market. Suggestions for regulatory
scrutiny’, (1999) 4 International Journal of Communications Law and Policy at: <http://w ww.coe.int/t/dghl/
standardsetting/media/Doc/M M-S -PL(1999)012_en.asp>.
177
Wu, T, ‘Network Neutrality, broadband discrimination’, (2003) 2 Journal on Telecommunications and
High-Tech Law 141.
178
Communications Act of 1934 as amended by Communications (Deregulatory) Act of 1996, 47 USC
§§153(20) (definition of ‘information service’), 153(10) (definition of ‘common carrier’), 153(43) (definition of
‘telecommunications’), and 153(46) (definition of ‘telecommunications service’).
179
See Federal Communications Commission ‘FCC approves SBC/AT&T and VERIZON/MCI mergers’ 31
October 2005 at <http://h raunfoss.fcc.gov/edocs_public/attachmatch/DOC-261936A1.pdf>.
180
Powell (2004) Four Freedoms speech, at <http://h raunfoss.fcc.gov/edocs_public/attachmatch/DOC-
243556A1.pdf>.
181
FCC 05-151, adopted 5 August 2005.
679
FCC then made an Order of 1 August 2008 against Comcast, a major cable broad-
band ISP.184 FCC ordered Comcast to within 30 days:
FCC found that ‘Comcast has an anti-competitive motive to interfere with cus-
tomers’ use of P2P applications.’ This is because P2P offers a rival TV service delivery
182
Madison River Communications, LLC, Order, DA 05-5 43, 20 FCC Rcd 4295 (2005), available at <http://
hraunfoss.fcc.gov/edocs_public/attachmatch/DA-05-5 43A1.pdf>.
183
In AT&T Inc and BellSouth Corp (2007) Application for Transfer of Control, 22 FCC Rcd 5562, note the
dissent at Appendix F, slip op p 154.
184
Federal Communications Commission, Formal Complaint of Free Press and Public Knowledge Against
Comcast Corp for Secretly Degrading Peer-to-Peer Applications; Broadband Industry Practices; Petition of
Free Press et al for Declaratory Ruling that Degrading an Internet Application Violates the FCC’s Internet
Policy Statement and Does Not Meet an Exception for ‘Reasonable Network Management’ 2008, Memorandum
Opinion and Order, 23 FCC Rcd 13028 (‘ComcastOrder’).
70
than cable, which the FCC found ‘poses a potential competitive threat to Comcast’s
video-on-demand (VOD) service.’ FCC concluded that Comcast’s conduct blocked
internet traffic, rejected Comcast’s defence that its practice constitutes reasonable
network management, and ‘also concluded that the anticompetitive harms caused
by Comcast’s conduct have been compounded by the company’s unacceptable
failure to disclose its practices to consumers.’ FCC Chairman Martin was not
condemning ‘metered broadband’. Comcast announced a 250 Gigabyte monthly
limit in early September 2008, replacing its previous discretionary Terms of Use
reasonable caps.
The FCC justified its regulatory authority to issue the order, invoking its Title I
ancillary jurisdiction under the Communications Act to regulate in the name
of ‘national Internet policy’ as described in seven statutory provisions, all of
which speak in general terms about ‘promoting deployment’, ‘promoting acces-
sibility’, ‘reducing market entry barriers’. Comcast appealed to the District of
Columbia Court of Appeals, to overturn the order on these grounds, winning
in 2010 because the FCC failed to tie its assertion of ancillary authority to any
‘statutorily mandated responsibility’.185 This decision occurred as the FCC was
then consulting on a draft NPRM, with the perhaps inevitable result that the
FCC NPRM was immediately appealed in 2010, and finally overturned in 2014
(with the exception of consumer transparency requirements that stayed in force
throughout).
185
Comcast Corp. v FCC, 600 F.3d 642, 2010 United States Court of Appeals for the District of Columbia.
186
See FCC Report and Order (2010) Preserving the Open Internet, 25 FCC Rcd 17905 and FCC Report and
Order, In The Matter Of Preserving The Open Internet And Broadband Industry Practices, GN Docket No 09-
191, WC Docket No 07-52, FCC 10-201 §21–30 (2010).
71
formally make representations to bring the question of the FCC authority under the
Communications Act to court. A case before the DC Appeals Court led to FCC defeat
in 2014187.
187
In Re: Federal Communications Commission, In the Matter of Preserving the Open Internet, Report and
Order, FCC 10-201, 76 Fed Reg 59192 (2011), 23 September 2011, Consolidation Order, 1 (Judicial Panel on
Multidistrict Litigation, 6 October 2011), at <http://commcns.org/sOFyyT>.
188
United States Telecom Association v Federal Communications Commission, Decided June 14, 2016, DC
Cir., No. 15-1063.
189
2015 Open Internet Order, 30 FCC Rcd., at 5706
190
FCC (2010) Report and Order Preserving the Open Internet, 25 FCC Rcd 17905.
191
Kang, C and Tsukayama, H, ‘AT&T to Throttle Data Speeds for Heaviest Wireless Users’ (2011), at <http://
www.washingtonpost.com/ b usiness/t echnology/a tandt-t o-t hrottle-d ata-s peeds-f or-h eaviest-w ireless-
users/2011/0 8/01/g IQAh0HBoI_story.html>.
192
Verizon v Federal Communications Commission, 740 F.3d 623 (D.C. Cir. 2014); 11–1355, 14 January 2014.
193
United States DC Court of Appeals No. 15-1063 United States Telecom Association, et al. v Federal
Communications Commission 14 June 2016.
194
FCC, ‘Internet Policy Statement 05–151’ (2014), at <https://apps.fcc.gov/edocs_public/attachmatch/FCC-
05-151A1.pdf>. FCC, Madison River Communications, LLC, Order, DA 05–543, 20 FCC Rcd 4295 (2005).
195
FCC, ‘Protecting and Promoting the Open Internet’, GN Docket No. 14-2 8, Report and Order on Remand,
Declaratory Ruling, and Order, FCC 15-2 4 (2015), at <https://apps.fcc.gov/edocs_public/attachmatch/FCC-
15-2 4A1.pdf>.
72
196
See Marsden, C, (2016) ‘Better Regulation of Net Neutrality: A Critical Analysis of Zero Rating
Implementation in India and the United States’, Part I, Chapter 4, 52–8 5 in Belli (ed), Net Neutrality
Reloaded. Zero Rating, Specialised Services, Ad Blocking and Traffic Management (2016) United Nations
Internet Governance Forum, Dynamic Coalition on Net Neutrality, at <http://i nternet-g overnance.fgv.
br/s ites/i nternet-g overnance.fgv.br/f iles/p ublicacoes/net_ neutrality_ r eloaded_0 .pdf>. For economic
analysis, see OXERA, ‘Zero rating: free access to content, but at what price?’, July 2016, at <http://w ww.
oxera.com/L atest-T hinking/A genda/2 016/Z ero-r ating-f ree-a ccess-t o-c ontent,-b ut-a t-w hat-p r.aspx>.
197
Northrup, L, ‘T-Mobile Now Exempts 33 Streaming Music Services From Data Limits, Adds Apple Music,
Consumerist’ (2015), at <http://consumerist.com/2015/07/2 8/t-mobile-now-exempts-33-streaming-music-
services-f rom-data-l imits-adds-apple-music/>.
198
Goldstein, P, ‘Net Neutrality Rules won’t Force Carriers to Get FCC Permission for New Plans’ (2015),
at <http://w ww.fiercewireless.com/story/net-neutrality-r ules-wont-force-c arriers-get-fcc-permission-new-
plans-offic/2015-02-26>.
199
Telecom Paper, ‘FCC Set to Approve AT&T’s DirecTV Takeover with Conditions’ (2015), at <http://w ww.
telecompaper.com/news/fcc-set-to-approve-atandts-d irectv-t akeover-w ith-conditions/> (‘If approved by
the commissioners, 12.5 million customer locations will have access to a competitive fibre connection from
AT&T. The additional roll-out is around ten times the size of AT&T’s current FttP deployment and increases
the national residential fibre build by over 40 percent. . . AT&T will not be permitted to exclude affiliated video
services and content from data caps on its fixed broadband connections. It will also be required to submit all
competed interconnection agreements with the FCC.’)
73
FCC ‘may later rescind an advisory opinion, but any such rescission would apply
only to future conduct and would not be retroactive’.204
The new FCC Chair following the election of President Donald Trump, Ajit Pai,
and his Republican colleague (there were only three Commissioners in the first
half of 2017) issued a new NPRM in May 2017 for three months’ consultation.205 It
was finally adopted in December 2017.206 The Order guts the existing rules of their
enforcement power, while staying any investigations into BIAS discrimination in
favour of their zero rated video and audio services (ie not counting to the monthly
data cap).
200
Brodkin, J, ‘Comcast to Stop Blocking HBO Go and Showtime on Roku Streaming Devices’ (2014) Ars
Technica, Condé Nast Digital, at <http://a rstechnica.com/business/2014/12/comcast-to-stop-blocking-hbo-
go-a nd-showtime-on-roku-streaming-devices/>.
201
FCC, In the Matter of Applications of AT&T Inc. and DIRECTV For Consent to Assign or Transfer Control of
Licenses and Authorizations MB Docket No. 14-9 0 (2015), at <http://t ransition.fcc.gov/Daily_Releases/Daily_
Business/2015/db0728/FCC-15-94A1.pdf>.
202
FCC, Open Internet Advisory Opinion Procedures, Protecting and Promoting the Open Internet, GN Docket
No. 14-28 (2015), at <https://w ww.fcc.gov/document/public-notice-open-i nternet-advisory-opinions>.
203
Ibid.
204
Ibid.
205
WC Docket No. 17-108, In the Matter of Restoring Internet Freedom Adopted: May 18, 2017 Reply Comment
Date: 16 August 2017.
206
WC Docket No. 17-108, In the matter of Restoring Internet Freedom, Declaratory Ruling, Report and Order,
14 December 2017 (FCC 17-166).
74
Many lawyers would predict another decade’s litigation and regulation will still
not resolve the issue.
In its initial explanation of its reasons to review the 2002 Directives,214 the
Commission in 2006 noted the US debate but did no more than discuss the
207
An excellent narrative account of the policy and judicial history is Feld, H, ‘Net Neutrality in Court This
Week: The Story of How We Got Here’, Public Knowledge, 2 December 2016, at <https://w ww.publicknowledge.
org/news-blog/blogs/net-neutrality-i n-court-t his-week-t he-story-of-how-we-got-here/>.
208
Complaint of Free Press & Public Knowledge Against Comcast Corp., File No. EB-08-IH-1518 (filed 1
November 2007); Petition of Free Press et al. for Declaratory Ruling, WC Docket No. 07-52 (filed 1 November 2007).
209
FCC, (2009) In the Matter of Preserving the Open Internet, Broadband Industry Practices (Proceeding 09-191).
210
FCC, (2014) In the Matter of Protecting and Promoting the Open Internet GN Docket No. 14-28 Notice of
Proposed Rulemaking Adopted: 15 May 2014 Comment Date: 15 July 2014 Reply Comment Date: 10 September 2014.
211
GN Docket No. 09-191, WC Docket No. 07-52, Report and Order, 25 FCC Rcd 17905, 17910, para 13.
212
Adopted on 26 February 2015, in effect 12 June 2015.
213
Verizon v FCC, 740 F. 3d 623 - 2014.
214
See Directive 2002/21/EC (‘Framework Directive’), Directive 2002/20/EC (‘Authorisation Directive’),
Directive 2002/19/EC (‘Access Directive’), Directive 2002/22/EC (‘Universal Service Directive’), Directive
2002/58 on Privacy and Electronic Communications.
75
There, in summary, are the concerns about ISPs discriminating against content
they dislike, or in favour of affiliated content.219 The Directives became effective in
215
COM(2006) 334 final Review of the EU Regulatory Framework for electronic communications networks
and services, Brussels, 29 June 2006 at section 6.2–6.4.
216
See Scott Marcus, J, ‘Network Neutrality: The Roots of the Debate in the United States’, (2008) 43 Intereconomics
30–37; Jasper P Sluijs, ‘Network Neutrality Between False Positives and False Negatives: Introducing a European
Approach to American Broadband Markets’, (2010) 62 Federal Comm Law Journal 77–117; Cave, M and Crocioni, P,
‘Does Europe Need Network Neutrality Rules?’, (2007) 1 International Journal of Communication 669–679; Valcke,
P, et al, ‘Guardian Night or Hands off? The European Response to Network Neutrality: Legal Considerations on the
Electronic Communications Reform’, (2008) 72 Communications and Strategies 89–112.
217
Leading to a significant emphasis on net neutrality in SEC(2007) 1472 Commission Staff Working
Document: Impact Assessment at 90–102 (2007).
218
European Commission, Declaration on Net Neutrality, appended to Dir 2009/140/EC, O J L 337/37 at
p 69, 18 December 2009 at <http://eur-lex.europa.eu/LexUriServ/LexUriServdo?uri=OJ:L:2009:337:0037:0069:
EN:PDF>.
219
See Jasper P Sluijs, Florian Schuett, and Bastian Henze, Transparency regulation in broadband mar-
kets: Lessons from experimental research, (2011) 35 Telecommunications Policy 592–6 02 for an experimental
analysis of transparency regulation in broadband.
76
Member States in May 2011220 stating that Member States may take action to ensure
particular content is not discriminated against directly (by blocking or slowing
it), or indirectly (by speeding up services only for content affiliated with the ISP).
The 2009 Declaration, and the more legally relevant Directive clauses, relied
heavily on the implementation at national level and proactive monitoring by the
Commission itself, together with national courts, and privacy regulators where
content discrimination contains traffic management practices which collate per-
sonal subscriber data.221 Nevertheless, it laid out the principle of openness and
net neutrality. The Commission itself emphasized in 2010 that it would introduce
‘a particular focus on how the “net freedoms” of European citizens are being
safeguarded in its annual Progress Report to the European Parliament and the
Council’.222 Article 22(3) of the Universal Service Directive, stipulated that regu-
latory authorities should be able to set minimum quality-of-service standards: ‘In
order to prevent the degradation of service and the hindering or slowing down of
traffic over networks, Member States shall ensure that [NRAs] are able to set min-
imum quality of service requirements’.
As with all telecoms licensing conditions, net neutrality depends on the phys-
ical capacity (or its inexact surrogate, the user’s monthly data cap) made available,
and it may be that de facto exclusivity results in some services for a limited time
period as capacity upgrades are developed.223 Regulations passed in licensing can
affect network neutrality at a fundamental level. Interoperability requirements
can form a basis for action where an ISP blocks an application.
220
Directive 2009/136/ EC (the ‘Citizens Rights Directive’) and Directive 2009/ 140/EC (the ‘Better
Regulation Directive’) both of 25 November 2009, which must be implemented within 18 months.
221
See Directive 95/4 6/EC of 24 October 1995, OJ L 281/31 (1995); Directive 2002/58/EC, OJ L 201/37 (2002);
Directive 2006/2 4/EC of 15 March 2006 on the retention of data generated or processed in connection with the
provision of publicly available electronic communications services or of public communications networks
and amending Directive 2002/58/EC OJ L105/5 4 (2006).
222
Ibid.
223
See GN Docket No 09-191 Broadband Industry Practices WC Docket No 07-52 ‘In the Matter of Further
Inquiry into Two Under-Developed Issues in the Open Internet Proceeding Preserving the Open Internet’,
and Andersen et al, Joint Reply Comments Of Various Advocates For The Open Internet, 4 November 2010,
Comments on Advancing Open Internet Policy Through Analysis Distinguishing Open Internet from
Specialized Network Services.
7
<http://stakeholders.ofcom.org.uk/consultations/net-neutrality/?showResponses=true>.
224
Ofcom (2012) Draft Annual Plan 2012/13, at <http://w ww.ofcom.org.uk/about/a nnual-reports-a nd-
225
plans/>.
226
Ofcom (2012) at paragraphs 5.40–5.42. 227
See Marsden (2017), n 151, at Chapter 6.
78
This repaired a lacuna in the law, in that the 2002 framework did not permit formal
complaints to be made by content providers regarding their treatment by ISPs.
They explain that though mobile will always need greater traffic management
than fixed (‘traffic management for mobile accesses is more challenging’232), sym-
metrical regulation must be maintained to ensure technological neutrality: ‘there
are not enough arguments to support having a different approach on network neu-
trality in the fixed and mobile networks. And especially future-oriented approach
for network neutrality should not include differentiation between different types
of the networks.’
BEREC in December 2011 published its guidelines on transparency and QoS233
On transparency, ‘BEREC stated that probably no single method will be suffi-
cient’234 and points out the limited role of NRAs.
228
<http://e c.europa.eu/i nformation_ s ociety/p olicy/e comm/ l ibrary/p ublic_ c onsult/n et_ n eutrality/
index_en.htm>.
229
BoR (10) 42 BEREC Response to the European Commission’s consultation on the open Internet and net
neutrality in Europe, at <http://w ww.erg.eu.int/doc/berec/bor_10_42.pdf>.
230
BoR (10) 42 at p 3. 231 BoR (10) 42 at p 11. 232 Ibid.
233
Documents BoR 53(11) Quality of Service and BoR 67(11) Transparency, at <http://erg.eu.int/docu-
ments/berec_docs/i ndex_en.htm>.
234
See BoR 67 [11] at p 5.
79
The Netherlands became the first European nation to formally introduce man-
dated network neutrality in 2012, implementing that law in late 2013, prompting a
new European Regulation on the subject that was negotiated for two years before
being passed in 2015.235
235
See further Marsden, n 151, esp. Chapter 4.
236
Regulation (EU) 2015/2120 of 25 November 2015 laying down measures concerning open internet access
and amending Directive 2002/22/EC on universal service and users’ rights relating to electronic communica-
tions networks and services and Regulation (EU) No 531/2012 on roaming on public mobile communications
networks within the Union, OJ L 301/1.
237
<http://eng.nkom.no/technical/i nternet/net-neutrality/berec-a nd-net-neutrality>.
238
IP-15-5927 (2015) Press Release <http://europa.eu/rapid/press-release_I P-15-5927_en.htm> and
MEMO-15-5275 Roaming charges and open Internet: questions and answers, 27 October 2015, at <http://
europa.eu/rapid/press-release_M EMO-15-5275_en.htm>.
239
Art 2 of Directive 2002/21/EC.
870
‘Virtually all’ may have the meaning ‘substantially all’ or ‘most that are technic-
ally possible’. Is this all end-user points or content providers or both? It is at least a
close approximation of the FCC term broadband IAS:
A mass-market retail service by wire or radio that provides the capability to transmit
data to and receive data from all or substantially all Internet endpoints . . . This term
also encompasses any service that the Commission finds to be providing a func-
tional equivalent of the service described in the previous sentence, or that is used to
evade the protections set forth in this Part.240
The assumption is that creating a new definition captures services otherwise ex-
cluded by Framework Directive 2002/21/EC, Article 2:
The Regulation states that the open internet—not net neutrality, which is not de-
fined or written in the Regulation—is to be preserved in Articles 3–6, which im-
pose duties on IAPs, NRAs, and the Commission itself. Article 3(1) states the 2004
Powell Four Freedoms, Article 3(2) that contracts/practices cannot limit those
rights, and Article 3(3) that traffic must be treated equally:
1. End-users shall have the right to access and distribute information and con-
tent, use and provide applications and services, and use terminal equipment of
their choice, irrespective of the end-user’s or provider’s location or the location,
origin or destination of the information, CAS, via their IAS. This paragraph is
without prejudice to Union law, or national law that complies with Union law,
related to the lawfulness of the CAS.
2. Agreements between PIAS and end-users on commercial and technical condi-
tions and the characteristics of IAS such as price, data volumes or speed, and
any commercial practices conducted by PIAS, shall not limit the exercise of the
rights of end-users laid down in paragraph 1.
240
See FCC (2015) In the Matter of Protecting and Promoting the Open Internet, GN Docket No. 14-2 8 Report
and Order On Remand, Declaratory Ruling, And Order Adopted: 26 February 2015 Released: 12 March 2015 at
10 para 25. Used in both the 2010 and 2015 Open Internet Orders.
781
3. PIAS shall treat all traffic equally, when providing IAS, without discrimination,
restriction or interference, and irrespective of the sender and receiver, the
[CAS], or the terminal equipment used.
Immediately the general condition is detailed, and three justified exceptions are
identified:
The first subparagraph shall not prevent providers of IAS from implementing
reasonable TMM. In order to be deemed to be reasonable, such measures
shall be transparent, non-d iscriminatory and proportionate, and shall not
be based on commercial considerations but on objectively different tech-
nical [QoS] requirements of specific categories of traffic. Such measures
shall not monitor the specific content and shall not be maintained for longer
than necessary.
PIAS in particular shall not block, slow down, alter, restrict, interfere with, de-
grade or discriminate between specific CAS, or specific categories thereof, except
as necessary, and only for as long as necessary, in order to:
(a) comply with Union legislative acts, or national legislation that complies with
Union law, to which the provider of [IAS] is subject, or with measures that
comply with Union law giving effect to such Union legislative acts or national
legislation, including with orders by courts or public authorities vested with
relevant powers;
(b) preserve the integrity and security of the network, of services provided via that
network, and of the terminal equipment of end-users;
(c) prevent impending network congestion and mitigate the effects of excep-
tional or temporary network congestion, provided that equivalent categories
of traffic are treated equally.
Article 3(4) is a basic privacy reminder.241 Article 3(5) sets out a specialized services
(‘fast lane’) exception:
5. PECP, including PIAS, and providers of CAS shall be free to offer services other
than IAS which are optimised for specific CAS, or a combination thereof, where
the optimisation is necessary in order to meet requirements of the CAS for a spe-
cific level of quality. PECP, including PIAS, may offer or facilitate such services
only if the network capacity is sufficient to provide them in addition to any IAS
provided. Such services shall not be usable or offered as a replacement for IAS,
and shall not be to the detriment of the availability or general quality of IAS for
end-users.
241
‘Any TMM may entail processing of personal data only if such processing is necessary and proportionate
to achieve the objectives set out in paragraph 3. Such processing shall be carried out in accordance with
Directive 95/4 6/EC . .. TMM shall also comply with Directive 2002/58/EC.’
782
242
O’Donoghue, R and Pascoe, T, Net Neutrality in the EU: Unresolved Issues Under the New Regulation, 15
March 2015, at SSRN: <http://ssrn.com/abstract=2741173>.
243
Ibid, at 8.
244
Ibid, at 11. 245 Ibid, at 13.
783
There is then a requirement that users can actually enforce measures, but contract
termination is not specified:
2. PIAS shall put in place transparent, simple and efficient procedures to address
complaints of end-users relating to the rights and obligations laid down in
Article 3 and paragraph 1 of this Article.
3. The requirements laid down in paragraphs 1 and 2 are in addition to those
provided for in Directive 2002/22/EC and shall not prevent Member States
from maintaining or introducing additional monitoring, information and tran
sparency requirements, including those concerning the content, form and
manner of the information to be published. Those requirements shall comply
with this Regulation and the relevant provisions of Directives 2002/21/EC and
2002/22/EC.
4. Any significant discrepancy, continuous or regularly recurring, between
the actual performance of the [IAS] regarding speed or other [QoS] param-
eters and the performance indicated by the [PIAS] in accordance with
points (a) to (d) of paragraph 1 shall, where the relevant facts are estab-
lished by a monitoring mechanism certified by the [NRA], be deemed to
constitute non-c onformity of performance for the purposes of triggering
the remedies available to the consumer in accordance with national law.
This paragraph shall apply only to contracts concluded or renewed from 29
November 2015.
Minimal harmonization clauses inside Regulations are very rare, and it will be of
more general interest to analyse the outcome of this process.
874
Article 5(3) sets out BEREC’s duty: ‘By 30 August 2016, in order to contribute to
the consistent application of this Regulation, BEREC shall, after consulting stake-
holders and in close cooperation with the Commission, issue guidelines for the
implementation of the obligations of [NRAs] under this Article.’
Article 6 deals with penalties:
Member States shall lay down the rules on penalties applicable to infringements
of Articles 3, 4 and 5 and shall take all measures necessary to ensure that they are
implemented. The penalties provided for must be effective, proportionate and dis-
suasive. Member States shall notify the Commission of those rules and measures.
Given this short timetable, it may be that the 2019 review recommends codifying
the BEREC Guidelines, which are discussed in the next section.
246
Body of European Regulators of Electronic Communications (2016) BEREC Guidelines on the
Implementation by National Regulators of European Net Neutrality Rules. BoR (16) 127.
785
Regulation, and were primarily drafted by the UK (Ofcom) and Norwegian regu-
lators. BEREC’s guidelines at paragraph 45 expressed agnosticism on zero rating: ‘It
is not the case that every factor affecting end-users’ choices should necessarily be
considered to limit the exercise of end-users’ rights.’ BEREC at paragraph 46 warned
that the combination of the largest mobile operator with largest social network pro-
vider could produce an anti-competitive discriminatory access agreement:
a practice is more likely to limit the exercise of end-user rights in a situation where,
for example, many end-users are concerned and/or there are few alternative offers
and/or competing ISPs for the end-users to choose from.
247
The Open Internet Access (EU Regulation) Regulations 2016, SI 2016/6 07, at <http://w ww.legislation.gov.
uk/u ksi/2016/6 07/pdfs/u ksi_20160607_en.pdf>.
786
the striking down of the provisions in the earlier and broader Netherlands Telecoms
law of 2012. The Rotterdam administrative court of first instance in 2017 struck down
the Netherlands Telecoms Law 2012, Article 7(4) on net neutrality248 :
ACM (Dutch regulator) has ordered T-Mobile to discontinue the provision and
execution of the Data Free Music service under forfeiture of a penalty payment.
At the hearing it appears that Vodafone Ziggo disagrees with the far-reaching sus-
pension by ACM of the burden during a possible preliminary question. The objec-
tion lodged at the sitting—as part of the plea note—indicates that the court should
have been forwarded within the meaning of Article 7:1a.
ACM takes the view that zero rating is contrary to Article 3, second and third para-
graphs, of the Network Neutrality Regulation and with Article 7.4a, third paragraph,
of the Telecoms Law 2012. The court is of the opinion that the neutrality regulation
and in particular Article 3 of the Regulation undoubtedly contains no categorical
prohibition of price discrimination (‘acte clair’). Article 7.4a, third paragraph, is
therefore unequivocally contrary to the network neutrality regulation.
In this connection, the court notes that no other conclusion is possible than
the national legislature has acted against by better understanding by Article 7.4a,
third paragraph, in spite of the establishment history and the text of Article 3 of
the Network Neutrality Regulation.
248
ROT 17/4 68, ROT 17/1160, and ROT 17/1932, T-Mobile v Ziggo BV, Ziggo Services BV, Vodafone Libertel BV,
20 April 2017, at <https://u itspraken.rechtspraak.nl/i nziendocument?id=ECLI:NL:RBROT:2017:2940&showb
utton=true>.
249
European Commission, ‘Digital Single Market Reports and studies: Annual country reports on open
internet from national regulators—2017’, 20 July 2017 at <https://ec.europa.eu/d igital-single-market/en/
news/a nnual-country-reports-open-i nternet-national-regulators-2017>.
250
Ofcom, ‘Monitoring compliance with the EU Net Neutrality regulation: A report to the European
Commission’, 2017, at 8–9.
78
251
Kahn, Robert E, and Vinton G Cerf, ‘What Is The Internet (And What Makes It Work)’ (1999) Internet
Policy Institute, Washington DC at <http://w ww.internetpolicy.org/briefing/12_ 99.html>.
252
Hafner, Katie, and Lyon, Matthew, Where Wizards Stay Up Late, (Washington DC: Free Press, 1996) pp
52–6 6 describes Osterman’s response to Baran’s revolutionary concept.
87
levels of network neutrality, together with the continued development of NTD for
illegal third party content, will be vital to the continued health of the internet ac-
cess market.
The UK Parliament passed a Statutory Instrument in 2016 implementing the
EU Open Internet Regulation, signalling its intention to conform to the European
legal framework for the foreseeable future. Brexit may or may not mean Brexit, but
the UK’s continued tangible interconnection with the European open Internet
suggests that a continued harmonization of regulation in this area will apply.
789
Part VI
16
16 .1 INTRODUC TION
1
See further Chapter 6, Authorization and Licensing.
2
Issues relating to the assignment of frequency spectrum and orbital slots are discussed in Section 16.3.2.
974
3
An idea published by Arthur C Clarke in Wireless World in 1945. Coverage does not really extend to re-
gions above latitudes 75° north or south. The angle of elevation in northern Europe does significantly limit
reception.
4
Inmarsat, Section 16.2.1.2, for example, has three constellations of 11 satellites.
5
eg the O3b MEO system will initially use 12 satellites at an altitude of 8,062 km; Iridium’s LEO system uses
66 satellites at an altitude of 785 km; Globalstar’s LEO system will use 24 satellites at an altitude of 1,414 km.
6
Resolution 1962 (XVIII), adopted at UN General Assembly, 13 December 1963 (GAOR Annexes (XVIII) 28,
p 27). The physical boundaries of outer space are somewhat unclear, although 100 km above sea level, repre-
senting the boundary between the lower and outer atmosphere, is a generally accepted figure: see ‘The legal
regime of airspace and outer space: the boundary problem’ in Cheng, C, Studies in International Space Law
(Oxford: Clarendon Press, 1997) 425–456.
7
Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space,
including the Moon and Other Celestial Bodies (London, Moscow, and Washington, 27 January 1967; TS 10
(1968); Cmnd 3519).
759
also responsible under international law for their activities in outer space, whether
carried out by governmental or non-governmental authorities; the latter requiring
authorization and ongoing supervision (Article VI). Liability for damage caused
by any object placed in space would rest jointly with the State that launches, or
procures the launch of, the object, and the State ‘from whose territory or facility an
object is launched’ (Article VII). Jurisdiction and control over any object in outer
space remains with the State that has registered the object, whilst ownership is un-
affected by the presence of the object in space or its return to Earth outside of the
registering State (Article VIII). To facilitate international cooperation in the use of
outer space, States are required to provide information to the United Nations re-
garding their activities in, and use of, outer space (Article XI).
The liability provisions of the Outer Space Treaty were elaborated further in a
1972 Convention on International Liability for Damage Caused by Space Objects
(‘Liability Convention’).8 Reflecting the terms of the 1967 Treaty, liability lies with
the ‘launching State’; this encompasses both the State that launches or procures
the launch of the space object and the State from where it was launched (Article I).9
In certain circumstances, this definition could result in there being three poten-
tial ‘launching states’; for example, where a satellite supplier based in the UK ar-
ranges for the launch of satellite for a customer based in France, under a ‘delivery
in-orbit’ arrangement, from a launch service provider based in Kazakhstan.10
Where a launch has involved two or more States, then liability is joint and several
(Article V), unless agreed otherwise privately by the parties.11
Liability results from damage caused by a ‘space object’, which includes its ‘com-
ponent parts’ and the ‘launch vehicle and parts thereof’ (Article I(d)). The concept
of ‘damage’ is defined under the Liability Convention in the following terms:
. . . means loss of life, personal injury or other impairment of health; or loss of or
damage to property of States or of persons, natural or juridical, or property of
international intergovernmental organisations (Article I(a)).
8
London, Moscow and Washington 29 March 1972; TS 16 (1974); Cmnd 5551. The Treaty entered into force
for the United Kingdom on 9 October 1973.
9
Launching also includes any attempts.
10
See <http://www.bis.gov.uk/ukspaceagency/what-we-do/space-and-the-growth-agenda/uk-capabilities-for-
overseas-markets/the-outer-space-act-1986/registry-of-space-objects>.
11
eg an agreement between Russia and Khazakstan.
12
See generally Beer, T, ‘The specific risks associated with collisions in outer space and the return to earth
of space objects—t he legal perspective’, (2000) XXV(2) Air and Space Law, pp 42–50.
796
Liability is absolute under the Convention where the damage is caused on the
Earth or to an aircraft (Article II).13 The only formal claim that has been submitted
under Article II was by Canada in 1979, claiming C$6m from the Soviet Union
for damage caused by the radioactive debris from the re-entry of Cosmos 954 in
January 1978. The claim was settled in 1981 for C$3m, but without liability being
acknowledged.14 Liability is fault-based where the damage is to the space object of
another launching state caused elsewhere than on the Earth (Article III), which
raises the potential for complex evidential and causation issues in the event of
a claim. In 2009, a first example of a collision between intact satellites occurred
between the Iridium 33 and a defunct Russian satellite, Kosmos 2251.15 As well
as the direct damage caused to the colliding space objects, the resultant debris
could also have caused damage to other satellites, resulting in a third party claim
against the two states involved in the initial collision (Article IV.1(b)). In another
incident, in 2010, Intelsat announced it had lost control of Galaxy 15, leading to
potential interference with the transmission of AMC 11, a satellite owned by SES
World Skies; although it subsequently regained control in December 2010.16 Such
dangers and potential liabilities can be expected to become more common as the
space segment becomes increasingly crowded.
A State may claim damages either on behalf of itself; its natural or legal per-
sons (ie the State of nationality); or for those sustaining damage whilst in its terri-
tory (Article VIII). Alternatively, such persons may be able to seek remedies under
other rules of international or domestic law. Claims for compensation are subject
to certain time limits and, where diplomatic settlement is not achieved, may be
decided upon by a ‘Claims Commission’ established at the request of either party
(Articles XIV–X X).
Underpinning the 1962 Declaration and the Outer Space Treaty was the con-
cept that each State would maintain a register detailing the space objects for which
the State claimed jurisdiction and control. Such registration procedures were for-
malized under the 1975 Convention on the Registration of Objects Launched into
Outer Space.17 Under the Convention, the launching State accepted an obligation to
maintain a register (Article II), although the contents and conditions of use could
13
Unless it can be shown that the damage is the result of ‘gross negligence or an act or omission done with
intent to cause damage’ by the claimant State (Art VI).
14
Protocol on Settlement of Canada’s Claim for Damages Caused by Cosmos 954 (1981) 20 ILM 689. See also
Brearly, A, ‘Reflections upon the notion of liability: The instances of Kosmos 954 and space debris’, (2008) 34
J Space L 291.
15
New Scientist, ‘Space junk: Hunting zombies in outer space’, 15 September 2010.
16
See <http://w ww.intelsat.com/resources/galaxy-15/operational-status.asp>.
17
New York, 14 January 1975; TS 70 (1978); Cmnd 7271. The Convention entered into force for the UK on 30
May 1978.
79
See n 11.
18
The information to be supplied is: the name of the launching State or States; an appropriate designator or
19
registration number for the space object; the date, territory or location of launch; basic orbital parameters; and
the general function of the space object (see generally <http://w ww.oosa.unvienna.org>).
20
See Section 16.3.2.
21
An example of a typical licence can be obtained at <http://w ww.bis.gov.uk/a ssets/bispartners/
ukspaceagency/docs/osa/osa2008example.pdf>.
22
UK Space Agency, ‘Revised Guidance for Applicants—Outer Space Act 1986’.
23
See Ofcom, ‘Procedures for the Management of Satellite Findings’, 27 March 2007.
24
See ‘David Willetts secures agreement for cheaper access to space’, BIS, 4 July 2011.
25
HM Treasury and BIS, The Plan for Growth, March 2011, at 2.305.
798
will provide the basis for such reforms,26 which was subsequently introduced and
received Royal Assent in March 2018 as the Space Industry Act 2018.27 Under the Act,
an amended OSA will only be applicable to activities carried out overseas (which has
been the reality for UK licensed launches to date), while domestic launches will be
licensed under a new regime. The Secretary of State has been given the power to in-
demnify licensed operators in respect of any liability that exceeds the limit provided
for in the licence, to be specified in regulations (Section 35).
In terms of jurisdiction, a satellite system can be distinguished into two compo-
nents: the ‘earth segment’ and the ‘space segment’. The ‘earth segment’ comprises
those stations that send (‘uplinks’) and receive (‘downlinks’) transmissions from
the satellite and which are subject to the laws of the jurisdiction in which they are
physically located.28 The ‘space segment’ has been defined in the following terms:
. . . the telecommunications satellites, and the tracking, telemetry, command, con-
trol, monitoring and related facilities and equipment required to support the op-
eration of these satellites. (INTELSAT Agreement, Article 1(h))
Jurisdictional responsibility for the ‘space segment’ can be sub-d ivided between
the State that launched the satellite and the State from where the satellite is con-
trolled. If control is distributed between multiple sites, then it is the operator’s
principal place of business.
26
Department for Transport, Draft Spaceflight Bill (Cm 9421), February 2017, at <https://w ww.gov.uk/gov-
ernment/publications/d raft-spaceflight-bill>.
27
See <http://services.parliament.uk/bills/2017-19/spaceindustrybill.html>.
28
The geographical coverage of a satellite’s transmissions is known as its ‘footprint’.
29
Washington, 20 August 1964–20 February 1965; TS 12 (1966); Cmnd 2940.
30
See the Agreement relating to the International Telecommunications Satellite Organization ‘INTELSAT’
(with Operating Agreement), (Washington, 20 August 1971; TS 80 (1973); Cmnd 5416).
97
31
See the Convention on the International Maritime Satellite Organization (INMARSAT) (with the
Operating Agreement), London, 3 September 1976; TS 94 (1979); Cmnd 7722. It changed its name in 1994.
32
See generally Case No IV/35.296—I nmarsat-P, OJ C 304/6, 15 November 1995.
33
See the Convention establishing the European Telecommunications Satellite Organization (EUTELSAT)
(Paris, 15 July 1982; TS 15 (1990); Cmnd 956, as amended by a Protocol of 15 December 1983, Cmnd 9154).
The UK instrument of ratification of the Convention was deposited on 21 February 1985 and the Convention,
Operating Agreement and Protocol entered into force on 1 September 1985.
80
34
See <http://w ww.eutelsat.com/>: ‘Introduction to Eutelsat’.
35
Commission Directive 94/4 6/EC of 13 October 1994 amending Directive 88/301/E EC in particular with
regard to satellite communications, OJ L 268/15, 19 October 1994, at Art 3.
36
Council Resolution on further development of the Community’s satellite communications policy, espe-
cially with regard to the provision of, and access to, space segment capacity; OJ C379/5, 31 December 1994.
37
’Towards Europe-w ide systems and services—Green Paper on a common approach in the field of Satellite
Communications in the European Community’, Communication from the Commission, COM(90)490 final, 20
November 1990. See also the 1991 Guidelines, at paras 122–128.
810
direct access has subsequently been implemented in most of the Member States,
although through separate ancillary agreements rather than amendments to the
provisions of the international agreements.38 However, the Commission did not
consider such developments to be sufficient to ensure a fully liberalized market
in the provision of satellite-based services. Therefore, Member States now have an
obligation to ‘take all appropriate steps to eliminate’ incompatibilities between
the international conventions and the EC treaty.39
In the US, the government took a much more proactive stance towards the
anti-competitive position of the ISOs. In 2000, Congress adopted the Open-
Market Reorganization for the Betterment of International Telecommunications
Act,40 with the express purpose of ensuring that Intelsat and Inmarsat became
independent commercial entities with a pro-competitive ownership structure.
The Federal Communications Commission was required to determine whether
Intelsat, Inmarsat, or any of their successor entities ‘will harm competition in the
telecommunications markets of the United States’ and condition or deny any ap-
plications or authorizations where such harm is found to be present or potential.41
Such a unilateral move was in breach of the US’s international treaty obligations
under the Intelsat agreement,42 but acted as an effective spur to the privatization
process.
With the progressive moves towards full commercialization and privatization,
the treaty-based satellite systems are no longer relevant as a feature of international
telecommunications law. From a competition law perspective, the process of pri-
vatization has raised a number of issues, including the need to ensure that the
private operating entities do not retain any of the legal immunities granted to
international organizations; and opening up the shareholding to non-participant
entities, preferably through a public offering.43 Such operators are now subject to
the scrutiny of competition regulators in the same way as other multinational sat-
ellite ventures.44 However, the ISOs also had a public service remit, both in general
38
See Communication from the Commission, ‘Fifth Report on the Implementation of the
Telecommunications Regulatory Package’, November 1999.
39
Commission Directive 2002/77/EC ‘on competition in the markets for electronic communications net-
works and services’, OJ L 249/21, 17 September 2002 at Art 8(2).
40
The ‘ORBIT Act’, Pub L 106–180, 114 Stat 48 (2000), codified at 47 USC §761 et seq.
41
47 USC § 761(b)(1).
42
Sagar, D, ‘Privatisation of the Intergovernmental Satellite Organisations’, paper presented at the ECSL
Tenth Summer Course on Space Law and Policy, Nice, 27 August–8 September 2001.
43
Ungerer, H, ‘The transformation of the International Satellite Organisations—some aspects from a
European perspective’, 11 April 1999: published on the Competition Directorate-G eneral website. See also
Press Release, ‘Commission gives green light to Inmarsat restructuring’, IP/98/923, 22 October 1998.
44
eg Commission competition decisions: Case IV/3 4.768—International Private Satellite Partners, OJ L
354/75, 31 December 1994 and Case IV/35.518—I ridium, OJ L 16/87, 18 January 1997.
820
16.2.2 Submarine cables
Submarine cables have been a component of the international telecommunica-
tions infrastructure since 1851, when the first submarine cable for telegraphy was
laid between England and France. The first commercially successful transatlantic
telegraph cable was operational in 1866; the first transatlantic coaxial copper tele-
phone cable (TAT-1) in 1956, and the first transatlantic fibre optic cable (TAT-8) in
1988.46 The emergence of satellite technology was widely viewed as signalling the
demise of submarine cable as a transmission medium. However, submarine cable
has continued to prosper and expand as the dominant medium for international
traffic due to its superior transmission quality, reliability, and security, carrying
over 95 per cent of international voice and data traffic.47
The expense of laying submarine cables has meant that, historically, consortia
of operators from different jurisdictions have carried out such projects under pri-
vate agreement, often referred to as ‘cable clubs’. Such ‘clubs’ usually comprised the
monopoly operators from each jurisdiction connected to the cable. In contrast to
the first satellite systems, such consortia were not the subject of international con-
ventions. During the telecommunications boom of the late 1990s, the ‘club’ model
was supplanted by single private ventures, such as Global Crossing and FLAG, who
were able to raise sufficient investment from the capital markets without the need
for consortia. However, with the subsequent downturn in the sector, a number of
these companies experienced financial difficulties and numerous submarine cable
systems have been taken out of service.48 As a consequence, we have seen a return
of cable ‘clubs’ as a financing vehicle for submarine cable systems. Cable laying
projects are driven by the perceived growth in demand for bandwidth to carry data
traffic, which reflects in part general economic activity around the world.
45
Sagar, n 42.
46
See Davenport, T, ‘Submarine Communication Cables and Law of the Sea: Problems in Law and Practice’,
(2012) 43 Ocean Development & International Law, 201–2 42.
47
ICPC Presentation, ‘About submarine telecommunication cables: Communicating via the ocean’, kindly
made available to the author, July 2008.
48
See Burnett, R, ‘The legal status of out-of-service submarine cables’, Maritime Studies, No 137, July/
August 2004.
830
• the process of laying (at sea) and landing (on land) of the cable and its subse-
quent maintenance;49
• the provisioning of capacity in the form of IRUs (Indefeasible Rights of Use) and,
subsequently, as International Private Leased Circuits (IPLCs);50
• the operation of, and access to, the cable landing station; and
• the facilities required to connect the operator’s domestic network to the cable
landing station, commonly referred to as ‘backhaul’.
The issue of cable laying concerns issues of public international law and na-
tional marine law, in respect of landing rights. The establishment of cable landing
stations usually involves a complex array of national and (or) local planning, de-
velopment and environmental laws. The provisioning of capacity and ‘backhaul’
facilities, as well as access to landing stations, has come to the attention of tele-
communications regulatory authorities in terms of competition concerns.
In similar fashion to satellites, the international law of the sea governs the laying
of submarine cable and associated liabilities for damage, where such cable lies
outside the territory of a state. The primary international treaty establishing a legal
order for the seas is the United Nations Convention on the Law of the Sea 1982
(UNCLOS), which came into force in November 1994.51 There are 168 parties to the
UNCLOS, which does not include the United States. The Convention divides the
sea into five different zones, each subject to different legal regimes:
• internal waters are ‘on the landward side of the baseline of the territorial sea’
and are part of a state’s sovereign territory (Article 8);
• territorial waters extending 12 nautical miles in breadth and over which the
coastal State has sovereignty (Article 3), subject to the right of ‘innocent passage’
(section 3);
• continental shelf, comprising ‘the sea-bed and subsoil of the submarine areas
that extend beyond its territorial sea’ up to 200 nautical miles (Article 76), and
over which the coastal state exercises ‘sovereign rights for the purpose of ex-
ploring it and exploiting its natural resources’ (Article 77);
49
See generally Burnett, D, Beckman, R, and Davenport, T, Submarine Cables: The Handbook of Law
and Policy (Leiden: Martin Nijhoff, 2014); see also Hogan & Hartson, Submarine Cable Landing Rights and
Existing Practices for the Provision of Transmission Capacity on International Routes, Report to the European
Commission, August 1999.
50
For a consideration of the commercial aspects of IRUs, see Chapter 11, at Section 11.2.
51
See UN General Assembly Resolution A/4 8/263 of 28 July 1994. The Convention came into force in the UK
on 24 August 1997 (TS No 81 (1999), Cm 4524). The European Community has acceded in respect of those mat-
ters for which it has competence (Council Decision 98/392/EC, OJ L 179/1, 23 June 1998).
804
• exclusive economic zone extending over a 200 nautical mile zone, where the
state has the right to declare exclusive economic interests in the resources (Part
V); and
• high seas which are open to all States, both coastal and land-locked (Article 87).
A coastal State is entitled to lay submarine cables in its territorial waters, pro-
vided that they do not obstruct the rights of use of others, such as innocent passage
(Article 21(c)). Any State is entitled to lay cables on the continental shelf, subject
to the rights of other users already present; as well as the right of the coastal State
to take reasonable measures in respect of exploitation, controlling pollution, and
the imposition of conditions on cables entering its territory or territorial waters
(Article 79). States are also free to lay cables in the exclusive economic zone (Article
58) and the high seas (Article 87), subject to an obligation to respect existing cables
and pipelines (Article 112).
The need to protect submarine cables from damage caused by other uses of the
sea, such as fishing, dredging, or anchoring, gave rise to the Convention for the
Protection of Submarine Cables (CPSC) in 1884,52 which is applicable outside of
territorial waters.53 The CPSC was implemented in English law by the Submarine
Telegraph Act 1885, although any incompatible provisions within the UNCLOS
now supersede its provisions (Article 311(2)). Under the Submarine Telegraph Act,
it is an offence to unlawfully and wilfully, or by culpable negligence, break or in-
jure a submarine cable under the high seas, attracting a maximum tariff of five
years’ imprisonment (s 3). Conversely, where a ship owner can prove damage to
his equipment in order to avoid damaging a submarine cable, then the ship owner
may claim compensation from the cable owner, provided that ‘all reasonable pre-
cautionary measures’ were taken.54 In 1958, the International Cable Protection
Committee was established as an industry body comprising owners and operators
of submarine telecommunications cables, including government administrations,
in order to promote the protection of submarine cables against man-made and
natural hazards.55 It has produced a number of recommendations on issues such
as ‘Cable Routing and Reporting Criteria’, concerning the placing of new cables
near existing systems, which members comply with on a self-regulatory basis.56
52
Paris, 14 March 1884 (75 BFSP 356; C 5910). It has 40 state parties.
53
Art 1. Primarily in the continental shelf zone: Wagner, E, ‘Submarine Cables and Protections Provided by
the Law of the Sea’, (1995) 19(2) Marine Policy 127, at 132.
54
UNCLOS, Art 115. See also CPSC, Art VII. Under UK law, see the Continental Shelf Act 1964, s 8(1), refer-
ring to CPSC. Section 8(1A) extends the protection to submarine cables under territorial waters and the exclu-
sive economic zone. See Agincourt Steamship Co Ltd v Eastern Extension, Australasia and China Telegraph Co
Ltd [1907] 2 KB 305, CA.
55
<http://w ww.iscpc.org>. There are also national committees, such as the UK Cable Protection Committee
<http://w ww.ukcpc.org.uk/>.
56
<https://w ww.iscpc.org/publications/recommendations/>.
850
See, for example, ACMA media release 126/2007: ‘Protection zone declared for submarine telecommuni-
58
66
Cityhook Ltd v OFT and ors [2007] CAT 18.
67
See USTR, ‘Results of the 2007 Section 1377 Review of Telecommunications Trade Agreements’, at p 14–
15. Available from <https://u str.gov/sites/default/fi les/Resultsof%20the%202007%201377%20Review.pdf >.
68
For a detailed history of the ITU, see Lyall, F and Larsen, PB, Space Law: A Treatise, (Farnham: Ashgate,
2009) pp 200–206.
69
International Convention on Telecommunications, Atlantic City, 2 October 1947; 1950 UK Treaty Series
No 76, Cm 8124.
70
See the Constitution and Convention of the ITU, Geneva, 22 December 1992 (Treaty Series No 24, 1996,
Cm 3145). The following text is based on the Constitution and Convention as of March 2015.
870
(Constitution, Article 8), the last being held in Busan, Republic of Korea, 2014 and
the next being in Dubai in 2018. Between meetings, a Council, comprising no more
than 25 per cent of the total membership, acts on behalf of the Plenipotentiary
(Constitution, Article 10(3)). The work of the Union is sub-d ivided into three
sectors:
• ‘Member States’, ie national governments, of which there are currently 193, al-
though governments may designate national regulatory authorities as their
representative;71 and
• ‘Sector Members’, representing all the various categories of player within the
telecommunications industry, including regional and international organiza-
tions, such as the GSMA, and the intergovernmental satellite organizations,
such as ARABSAT.72 In total, these entities number over 700.73
Sector members have been involved in the work of the ITU since the Rome
Telegraph Conference in 1871, with the sponsorship of a Member State
(Convention, Article 19(1)(a), (b)). In 1998, the Convention was amended to en-
able Sector Members to apply directly to join the ITU; although the applicant’s
Member State must approve such a procedure (Convention, Article 19(4bis)-
(4ter)). However, despite being eligible for membership, it was not until the
Plenipotentiary in 1994 that Sector Members were able to formally partici-
pate in the decision-m aking processes of the ITU; and only in 1998 that Sector
Members were recognized as having formal rights of participation under the
Constitution:
71
eg Ofcom in the case of the UK, as directed by the Secretary of State under the Communications Act
2003, s 22.
72
Note that the international satellite organizations discussed in section 16.2.1.2, fall under Sector
Members, according to where they are established: Intelsat (US), Inmarsat (UK), Eutelsat (France).
73
See <http://w ww.itu.int/en/membership/Pages/sector-members.aspx>.
80
Under the Convention, the ITU Secretariat has obligations to ‘encourage the en-
hanced participation’ of Sector members (Article 19), while a Sector Member may
also be authorized to act on behalf of a Member State (Convention, Article 19(9)),
which may be the case where an operator continues to be part of the government,
often under a specific ministry, or has been conferred with certain special or ex-
clusive rights within the jurisdiction. Sector Members participate in those sectors
of the ITU for which they apply, eg ITU-R, so participation in one sector does not
confer authorization to participate in another.
Despite the enhanced status of the Sector Members, the fundamental legal
instruments of the ITU, the Constitution, Convention, and Administrative
Regulations,74 continue to be under the exclusive jurisdiction of the Member States.
An industry player may also be invited by a Sector of the ITU to participate as
an ‘Associate’ within a study group (Convention, Article 19(12)), with more limited
rights of participation, although with an obligation to help defray the costs of the
group in which they participate (Convention, Article 33(5)(4bis)). This category of
participants was established within the ITU system in 1988, as a means of enabling
participation by small entities in the work of the ITU.
With the liberalization of the telecommunications industry and the prolifer-
ation of commercial operators, tension has grown within the ITU over the position
of industry members within the ITU structure. On the one hand, governments are
wary of relinquishing their historic rights to control the organization; whilst on
the other hand, they recognize industry’s legitimate interests in the work of the
Union, as well as wanting industry to contribute an ever greater proportion of the
costs associated with its operations and activities.75 The issue of industry involve-
ment dominated the 1998 Plenipotentiary Conference in Minneapolis, where a
single category of industry membership was finally recognized:
74
See Section 16.3.4.
75
See Resolution 110 (Marrakesh, 2002): ‘Review of the contribution of Sector Members towards defraying
the expenses of the International Telecommunication Union’.
809
In terms of financing the work of the ITU, the Constitution was amended to place
Sector Member contributions on an equal footing to those of Member States
(Article 28). In addition, new ‘Advisory Groups’ were established for each Sector,
with a broad remit to review the ‘priorities, programmes, operations, financial
matters and strategies’ of the various bodies within each Sector (Convention,
Article 11A, 14A, 17A). These new bodies have increased the influence of Sector
Members within the ITU as Member States and industry participate on an equal
footing.
As part of a broad review of the ITU’s role and strategy for the future, an ITU
Reform Advisory Panel was established at the end of the last decade, comprising
both governmental and private sector members,76 and made the following rec-
ommendation in 2000 with respect to the balance of influence between Member
States and Sector Members within the ITU:
The decision-making functions of the ITU should reflect the modern, competitive
telecommunications environment in which the private sector plays the lead role
while the regulatory agencies act as an arbitrator for the wider public interest.77
16.3.1 Standards
It was the issue of technical standards that gave rise to the establishment of the
International Telegraph Union in 1865, when governments recognized the need
for standards to extend the telegraph network throughout Europe. Standards
represent the cornerstone of the global telecommunications industry, and the
ITU is one of the leading international institutions for de jure standards-making.
Critically, the ITU’s standards remit extends not only to technical issues, but also
operational and tariff structures for international telecommunication services,
76
For a full list of members, see <http://w ww.itu.int/newsroom/reform/rapmembers.html>.
77
ITU Reform Advisory Panel (RAP), Observations and Recommendations for Reform, 10 March 2000.
801
which has extended its potential to influence or interfere (depending on your per-
spective!) in sectoral developments.78
Over recent years, the ITU’s position in the standards-setting field has dimin-
ished in the face of regulatory ‘competition’ from regional organizations,79 industry
bodies,80 and, most significantly, de facto standards organizations such as the Internet
Engineering Task Force (IETF) which are able to develop standards much more rap-
idly that formal bodies such as the ITU. Recognizing such developments, the ITU in-
stituted an ‘alternative approval process’ (AAP) in 2001,81 to fast-track the adoption
of certain standards; although the process is not available for recommendations that
have ‘policy or regulatory’ implications.82 Standards adopted under the AAP have the
same status as those approved under the traditional process.83
In the standards arena, the ITU has also examined ways to reposition itself:
ITU-T could become a facilitator for collaboration, convening meetings among
different standards bodies and industry forums, in particular on interworking be-
tween the Internet and telecommunications networks, both fixed and mobile.84
78
ie ITU-T D-Series Recommendations: ‘General Tariff Principles’. See further Section 16.3.5.
79
eg the European Telecommunications Standards Institute (ETSI): <http://w ww.etsi.org>.
80
eg the GSM Association: <http://w ww.gsma.com> comprises nearly 800 mobile operators.
81
Recommendation ITU-T A.8 ‘Alternative approval process for new and revised ITU-T recommendations’
(10/2008) .
82
Ibid, at 1.1. See the Convention, Art 20.5bis 4, for guidance as to what may have policy or regulatory
implications.
83
Ibid, at 1.2. 84
RAP n 77, at 3.
85
See Ryan, P, ‘The ITU and the Internet’s Titanic Moment’, (2012) Stan Tech L Rev 8.
81
which resulted in the IETF and the ITU adopting different incompatible
standards.86
16.3.2 Radiocommunications
The development of radiocommunications at the beginning of the twentieth cen-
tury also gave rise to the need for international cooperation to avoid harmful inter-
ference. The International Radiotelegraph Union, established in 1906, adopted
operating principles that have continued to form the basis of the ITU’s regulation
of radiocommunications: Member States were required to notify each other of any
new service utilizing the radio spectrum and were obliged to ensure that such
services did not interfere with other uses of the frequency.87
The Radiocommunications Sector of the ITU, primarily operating through the
Radio Regulations Board, exercises a regulatory function in respect of the use of
two scarce international resources, radio-frequency spectrum and orbital slots,
both of which require management in order to maximize utilization, as well as
prevent interference between services and space objects.88 The ITU is responsible
for the ‘allocation’ of bands of radio-f requency spectrum to particular services (eg
broadcasting) and the ‘allotment’ of a given frequency or channel to a Member
State administration or geographic region. The Member State administration then
grants an ‘assignment’ of a frequency or channel to a specific operator, which is then
registered in the ‘Master International Frequency Register’ (the ‘Master Register’).
The ITU records all satellite filings, both geostationary and non-geostationary, as
well as the earth stations that communicate with those systems.89
Such procedures are designed ‘to eliminate harmful interference . . . and to im-
prove use made of the radio-frequency spectrum’.90 The overriding objective of
the ITU regulatory regime is the efficient use of the spectrum, while ensuring that
public safety and emergency communication services, the only other policy con-
cerns directly addressed in the Radio Regulations, are not adversely affected:
86
See Bennett, R, ‘The gathering storm: WCIT and the global regulation of the Internet’, Information
Technology & Innovation Foundation, November 2012.
87
See Allison, A, ‘Meeting the Challenges of Change: The reform of the International Telecommunications
Union’, (1993) 45(3) Federal Communications Law Journal 498.
88
Constitution, Art 1(2)(a), (b); Chapter II (Arts 12–16) and Convention, Section 5 (Arts 7–12). The ITU’s pro-
cedures cover both geostationary and non-geostationary satellite systems.
89
See the ITU ‘Space Network Systems Online’, at <http://w ww.itu.int/sns/>.
90
Harmful interference is defined as ‘Interference which endangers the functioning of a radionavigation ser-
vice or of other safety services or seriously degrades, obstructs or repeatedly interrupts a radiocommunication
service operating in accordance with the Radio Regulations.’ Constitution, para 1003. See also the Radio
Regulations, Section 16.3.4.2, at Art 1(1.169). ‘Harmful interference’ is distinguished from ‘permissible inter-
ference’ (ie interference which falls within certain parameters) and ‘accepted interference’ (ie interference
greater than certain parameters, but accepted by two or more administrations).
821
91
Article 4, at 4.22. See further Section 16.3.4.
92
Article 44(2). Introduced in the 1973 ITU Convention.
93
ITU, ‘Overview of the Radio Regulations’, available at <http://w ww.itu.int/sns/radreg.html>.
94
RRs (2016), Resolution 4 (REV.WRC-03), ‘Period of validity of frequency assignments to space stations
using the geostationary-satellite and other satellite orbits’.
831
95
Jasentuliyana, N, International Space Law and the United Nations (Leiden: Martinus Nijhoff, 1999), at
p 309–310.
96
Indonesia placed one of its satellites in a slot registered with Tonga on the basis that the ‘assignment’ was
‘wrong in law’. Ibid, at 310.
97
See ITU Press Release, ‘Scrambling for Space in Space’ (Geneva, 16 September 2002), where it is stated that
the backlog of satellites awaiting coordination stood at 1200, with between 400–500 new requests each year.
98
Radio Regulations, Resolution 49, at Annex 2. 99
Radio Regulations, Article 13.6(b).
100
[2010] EWHC 2010 (Admin).
101
See also ECJ Case T-4 41/0 8, ICO Services Limited v European Parliament and Council, 21 May 2010, in
which the ICO sought unsuccessfully to annul Decision 626/2008/EC on the selection and authorization of
systems providing mobile satellite services (MSS).
841
designed for DTH and broadband internet access services across the Asia-Pacific
region. It was eventually granted a slot in September 2008 through Belarus, under
the Intersputnik umbrella. Protostar required an alternative ‘launch’ state sponsor,
which was the Republic of Belarus, the notifying administration for Intersputnik.
Intersputnik ensured that Protostar was compliant with ITU rules, following
complaints of harmful interference by China and orbital concerns expressed by
the United Arab Emirates.102 In April 2009, however, Intersputnik terminated its
concession agreement based on continuing allegations of interference, which re-
quired Protostar to shut down its transponder operations and look for a new spon-
soring administration, which it failed to do and so went in to administration.103
Intelsat has since purchased the satellite asset.
The proposed financial constraints would have included an annual coordin-
ation and registration charge, as well as a refundable deposit. The financial due
diligence proposals were, however, rejected over concerns that this would effect-
ively represent a spectrum usage charge. Instead, it was agreed that the ITU would
be able to recover its full costs for processing such applications.104 Such procedures
have helped reduce the filing backlog; although ongoing wrangles are taking place
between the ITU and satellite operators about the true costs involved and the re-
sulting high fees. This has led to substantial non-payment and arguments over the
consequences, ie the cancellation of the filing, and who bears the liability for the
outstanding invoice, either the operators or the Member States with whom the ITU
has a formal legal relationship.105
In terms of the spectrum bands, the ITU is also the forum for Member States
to debate the allocation or reallocation of newly or prospectively available spec-
trum. In November 2007, for example, at the ITU’s WRC, it was agreed that spec-
trum within the UHF band, which has traditionally been the exclusive preserve of
broadcasters, would be opened up for use by mobile broadband services.106 Such
spectrum is becoming available worldwide as a consequence of terrestrial tele-
vision shifting from analogue to broadband signals, which use considerably less
bandwidth; commonly referred to as the ‘digital dividend’.107 Such spectrum is
highly sought after because of the quality of signal available and their propaga-
tion characteristics, which means the signals travel further and are more capable
102
See ITU Circular Telegram of 21 July 2008 (CTITU A38 11S(SSD)O-2008-0 02171) and of 8 October 2008
(CTITU A45 11S(SSD)O-2008-0 03054).
103
Bloomberg, ‘ProtoStar Ltd., Satellite Operator, Files Bankruptcy’, 29 July 2009.
104
See ITU Resolution 91 ‘on cost recovery for some products and services of ITU’ (Minneapolis, 1998).
105
Sung, L, ‘ITU’s Cost Recovery: The Satellite Factor’, Satellite Today, 1 September 2004.
106
ITU Press Release, WRC- 07, ‘ITU World Radiocommunications Conference concludes after four
weeks: International treaty sets future course for wireless’, 16 November 2007.
107
See further Chapter 6 and Chapter 14.
851
of penetrating buildings. The signal range means the cost of rolling out wireless
broadband networks is considerably reduced, which is obviously beneficial for
developing countries.108 Further spectrum allocations for international mobile
communications were agreed at WRC-12, in February 2012, and have been placed
on the agenda for WRC-15.109
Therefore, since 1982, the ITU has given equal priority to telecommunica-
tions development with standards- setting and radiocommunications. The
Telecommunication Development Sector operates through a Telecommunication
Development Bureau, Telecommunication Development Conferences and associ-
ated Study Groups.
In particular, the ITU has worked with other development agencies, such as the
World Bank and the International Bank of Reconstruction and Development, to
improve the flow of technology, funds, and expertise into developing countries.
The Reform Advisory Panel has proposed that the ITU’s development focus should
be expanded ‘from technical assistance towards helping developing countries es-
tablish pro-market regulatory frameworks’,111 which reflects the influence of the
WTO’s work in the telecommunications sector.
112
ie 27 June 1994 in the case of the UK.
871
Final Acts, at Art 14.1. By virtue of Art 54.5bis of the Constitution, signatories are deemed to have con-
114
sented and become bound to the text if they do not notify the Secretary-G eneral by 1 January 2018.
115
There were 144 member states in Dubai, of which three countries have acceded subsequent to
the signing: Antigua and Barbuda, Belarus, and Kenya (but only the latter two were signatories to the
Melbourne ITRs).
81
three appendices,116 while the 2012 ITRs comprise fourteen articles and two ap-
pendices.117 In the event of conflict, where a party to the 2012 ITRs deals with a
non-signatory member state subject to the 1988 ITRs, the latter should be the ap-
plicable regime.118
Since 1988, there were inevitable calls for the ITRs to be revised, reinterpreted,
or abrogated, with, in the latter case, the provisions of continuing relevance being
transferred into other ITU instruments, such as the Constitution. These calls for
reform were driven, in part, by the considerable changes that have occurred in the
telecommunications sector since 1988, but also by developing country concerns
that the ITRs are too favourable towards richer nations and the dominant global
players they represent. At the 1998 ITU Plenipotentiary, a resolution was adopted
instructing the Secretary-General to establish an Expert Group to advise on the fu-
ture of the ITRs.119 No consensus on the way forward was reached by the following
Plenipotentiary in 2002, or again by the 2006 Plenipotentiary, although the 2006
Resolution finally put a prospective end date on the negotiations, by resolving that
the ITU convene a conference in 2012 to decide on recommendations to amend the
ITRs: The World Conference on International Telecommunications (WCIT) held in
Dubai, UAE, in December 2012.
In the lead up to the WCIT, Member States submitted their proposals for re-
form of the ITRs, representing a broad spectrum of opinion, from no change to
radical expansion.120 It was not possible to reconcile such divergent views at the
WCIT so consensus could not be achieved and a vote was required—a very rare
occurrence within ITU decision-making procedures. The reasons behind this
failure are themselves contested, with accusations of a media campaign based on
misinformation.121
It is beyond the scope of this section to engage in a detailed analysis of the
changes that were made and the differing interpretations of their significance, al-
though the 2012 amendments can be broadly sub-d ivided into updates to existing
provisions to reflect the changing environment, and the insertion of new provi-
sions. In addition, the Final Acts included five non-binding Resolutions.
116
Available at <http://search.itu.int/history/HistoryDigitalCollectionDocLibrary/1.1.48.en.100.pdf>. They en-
tered into force on 1 July 1990.
117
Available at <http://search.itu.int/h istory/H istoryDigitalCollectionDocLibrary/1.42.48.en.101.pdf>.
118
See <http://w ww.itu.int/en/wcit-12/Pages/t reaties-signing.aspx>.
119
Resolution 79 (Minneapolis, 1998): ‘International Telecommunication Regulations’.
120
See ITU CWG-WCIT12/T D-43, ‘Draft compilation of options’, 24 November 2011. See also Bennett, R,
‘The Gathering Storm: WCIT and the Global Regulation of the Internet’, ITIF, November 2012, at <http://
www2.itif.org/2012-gathering-storm-wcit-regulations.pdf>.
121
Hill, R, ‘WCIT: Failure or Success, Impasse or Way Forward’, (2013) International Journal of Law and
Information Technology 1–16, 3.
891
Criticisms primarily revolve around concerns that the ITRs may disturb current
governance arrangements for the internet, facilitating greater (repressive) govern-
mental input, and the threat that rules designed to regulate the provision of tele-
communication services may be used to control content sent over such services.
In terms of the latter, a sentence was specifically inserted into the scope of the 2012
ITRs expressly stating that they ‘do not address the content-related aspects of tele-
communications’ (Article 1.1(a)). Despite this, however, two new provisions ad-
dressing network security (Article 6) and controlling ‘unsolicited bulk electronic
communications’ (Article 7) have been viewed as granting Member States a right
to monitor traffic content for the purpose of ensuring compliance.122
One provision that has been of particular importance since 1988 has concerned
‘Special Arrangements’, which grants administrations the flexibility to enter into
‘special arrangements’ for the provision of international telecommunications
networks and services, either on the basis that they ‘do not concern Members in
general’ or based on ‘special mutual arrangements’ with other Members (1988
ITRs, Article 9, and retained in almost identical terms in the 2012 ITRs, at Article
13). Based on Article 42 of the ITU Constitution, this provision has been used by
Member States to tailor national and regional laws to reflect the evolving policy
of a liberalized market, such as the application of interconnection regulations to
intra-EU traffic, without reference to the other substantive provisions of the 1988
ITRs. The provision has also given ROAs considerable freedom to enter into private
agreements that have effectively established an alternative regulatory environ-
ment, which has been particularly relevant to the explosive growth of the internet.
While the majority of the text in the 1998 and 2012 versions of the ITRs address
similar subject matter, the controversial additions (said to be contained in six of
the seventy-seven paragraphs of the main text123) meant that the ITRs, which had
become increasingly irrelevant over the years, are now a symbolic illustration of a
lack of consensus and lines of tension within the international community in the
age of the internet.
122
eg Internet Society submission to the WCIT, <http://w ww.internetsociety.org/doc/ WCITSubmission>
October 2012.
123
Hill, n 121.
802
due to the ‘law of physics’,124 since non-compliance can result in harmful inter-
ference for all relevant parties. The current edition of the RRs was published in
2016.125
The RRs distinguish between three distinct acts in relation to frequency: ‘allo-
cation’, ‘allotment’, and ‘assignment’ (RRs, Article 1, 1.16–1.18). ‘Allocation’ con-
sists of an entry in the ‘Table of Frequency Allocations’ for use in respect of one
or more terrestrial or space radiocommunication service. Such services may be
categorized as ‘primary’ or ‘secondary’ services, on a regional or global basis; with
the latter being required to comply with the interference rules laid down for the
former, as well as being unable to claim protection from interference from the
former. ‘Allotment’ indicates the use of a designated frequency by administrations
for a service in certain countries or geographical areas and under specified condi-
tions. The ‘assignment’ of frequencies is carried out by Member States, under their
sovereign authority, through an authorization or licensing procedure, such as
under the UK’s Wireless Telegraphy Act 2006.126 Such assignment is then notified
to the ITU for recording in the Master Register.127 When granting an assignment,
Member States are free to derogate from the ITU allocation, but only to the extent
that it does not cause harmful interference to others operating in accordance with
the RRs (Article 4.4).
To ensure compliance with the RRs, particularly the elimination of harmful
interference, an international monitoring system has been established (RRs,
Article 16). The scheme comprises the operation of a network of monitoring sta-
tions, operated by Member States, either alone or in conjunction with others, and
international organizations, such as the ISOs. The system is voluntary in nature.
124
Lyall, F and Larsen, P, Space Law: A Treatise (Ashgate, 2009), at 230.
125
Available free of charge at <http://w ww.itu.int/pub/R-R EG-R R/en>.
126
See further Chapter 7.
127
eg Ofcom, Procedures for the Management of Satellite Filings, 27 March 2007.
128
eg over 2600 ITU-T Recommendations are currently in force.
129
However, see also the opinion of the Advocate-G eneral in Italy v Commission [1985] 2 CMLR 368, 373.
812
the various sectoral ‘Study Groups’ and enter into force either through approval
at the relevant assemblies or conferences, or through direct correspondence with
Member State administrations (Convention, Articles 11(2), 14(1)).
In the event of a dispute regarding the interpretation of any of the legal instru-
ments, Constitution Convention or Administrative Regulations, settlement should
either be achieved through mutually agreed bilateral or multilateral arrange-
ments or, if not settled by such means, via an arbitration procedure (Constitution,
Article 56). The decision of the arbitrator(s) shall be ‘final and binding upon the
parties to the dispute’ (Convention, Article 41), although no enforcement mech-
anism is available in the event of non-compliance. A compulsory arbitration pro-
cedure is also provided for under an Optional Protocol to the Convention, between
Members that are party to the Protocol.130
Constitution, Art 56(3). The UK has ratified the Optional Protocol, 27 June 1994.
130
132
Usually expressed in terms of Special Drawing Rights (SDR), under the International Monetary
Fund: Convention, Art 38; 1988 ITRs, at Art 6.3.1 and 2012 ITRs, at Art 8.2.4.
28
originates the most traffic that is required to make the periodic payments to the
terminating operator.
Although the system is embodied in the International Telecommunications
Regulations and has been elaborated as a series of recommendations from the ITU,
the system operates through a series of bilateral agreements between telecommu-
nication operators in each jurisdiction. Historically, such agreements would be
between public administrations in each country, ie the state incumbent, which
meant the agreements could be considered State measures subject to consider-
ation under public international law, such as the General Agreement on Trade in
Services.133 With liberalization, the overwhelming majority of agreements are now
negotiated privately between commercial entities, effectively taking them outside
the international accounting rate system.134
Whilst the essential elements of the international accounting rate system have
remained the same over many years, the system was in fact designed to operate
under certain conditions, which are no longer present in most telecommunica-
tions markets:
charges applicable in each direction of the same relation’ (Art 6.1.1); which is reiterated in 2012 ITRs, at
Art 8.2.5.
136
Federal Communications Commission, In the matter of International Settlement Rates, Report and
Order, IB Docket No 96–261, 7 August 1997 (‘Benchmark Order’): para 13.
823
137
See Tyler, M, Transforming economic relationships in international telecommunications, Chapter 8,
Briefing Report for ITU Regulatory Colloquium No 7 (1997). Also, Stanley, K, ‘International settlements in
a changing global telecom market’, in Telecom Reform Melody (ed) Technical University of Denmark, 1997.
138
Resolution 22: ‘Apportionment of revenues in providing international telecommunication services’
(Kyoto, 1994).
139
Tarjanne, P, ‘Reforming the International Accounting Rate System’, (1998) 2 ITU News.
140
See ITU Report of the Informal Expert Group on International Telecommunications Settlements,
March 1997.
141
Various forms of ‘call-back’ exist but it essentially involves a reversal in the direction of the call, eg a
call from a country with high originating international tariffs is manipulated to appear to come from the
terminating country which has low originating international tariffs, using features of call signalling systems.
142
’Refile’ involves routing a communication from country A to country B via a third country, C, where the
sum of the tariff rates for calls between A–C and C–B are less than A–B.
842
• ‘by-
pass’ techniques, which completely circumvent the international ac-
counting regime, eg international simple resale services, VSATs,143 internet
telephony.
143
Very Small Aperture Terminals, used for satellite-based telecommunications direct to home.
144
See ITU Resolution 21 of the Plenipotentiary Conference, Kyoto, 1994: ‘Special Measures concerning
Alternative Calling Procedures on International Telecommunication Networks’ (revised at the Minneapolis
Plenipotentiary, 1998), noted that 86 Member States prohibit ‘call-back’ (as of October 1998).
145
Resolution 21, n 144. 146
Similarly under the 2012 ITRs, at Art 8.2.2.
147
ITU Recommendation D.140, 6th edn, ‘Accounting rate principles for the international telephone ser-
vice’ (06/2002).
148
ITU-T Recommendation D.150, ‘New system for accounting in international telephony’ (06/99).
149
See further Chapter 8. 150
Ibid.
852
Reform of the system has also been driven, in part, by decisions made by national
regulatory authorities. In particular, the Federal Communications Commission
(FCC) created considerable consternation in certain countries when it issued
its International Settlement Rates ‘Benchmark’ Order in 1997.151 The FCC recog-
nized that the WTO ‘basic agreement’ had the potential to sharply worsen the
US’s balance of payments deficit on international services, since incumbent op-
erators in non-l iberalized markets would be free to establish US-based operations
subsidized from their monopolistic international revenues. With the slow pace of
reform within the ITU, the FCC decided to take unilateral steps to drive the pace
of change towards cost-based settlement rates. The Benchmark Order laid down
benchmark ‘settlement rates that carriers subject to our [FCC] jurisdiction may
pay for termination of US-originated traffic’ (paragraph 312). Countries were cat-
egorized into three tiers, representing different stages of economic development.
The rates were to be implemented over a transition period, over one to four years,
and operators were able to appeal against a rate determination (paragraph 74). The
regime came into effect on 1 January 1998 and the first targets were to be achieved
by 1 January 1999. All US-licensed carriers were subject to the order, while for
foreign-a ffiliated operators compliance was a condition of obtaining approval for
the provision of long-d istance services to the home jurisdiction (paragraph 207).
The Benchmark Order generated opposition in certain countries, especially in
the Caribbean region, over the potential impact the order would have on domestic
operator revenues. The European Commission and Japan also raised concerns
about the compatibility of the Benchmark Order with the US’s commitments under
the General Agreement on Trade in Services, specifically the principle of ‘most-
favoured-nation’.152 In 1998, Cable & Wireless brought an action before the US
courts challenging the legality of the Benchmark Order. Over 100 other petitioners
and intervenors, comprising national governments, regulators, and operators,
soon joined the case on both sides. The main thrust of the complaint was that the
FCC had exceeded its authority through the extraterritorial nature of the Order’s
provisions.153 The court found overwhelmingly in favour of the FCC, holding that it
had the requisite powers to make decisions regulating the actions of US-l icensed
operators, including the contractual arrangements entered into for international
settlement rates:154 the Commission does not exceed its authority simply because a
regulatory action has extraterritorial consequences. Objections to the FCC’s meth-
odology were dismissed on the grounds that the FCC had acted reasonably, whilst
151
Benchmark Order, n 136. It was reformed in 2004 (FCC 04-53) and 2012 (FCC 12-145).
152
Ibid, at para 109. See also Section 16.4.
153
Cable & Wireless et al v FCC, No 97–1612, DC Cir, 12 January 1999.
154
See 47 USC §205(a), 211(a).
862
the petitioners were criticized for withholding actual cost data which could have
been used as well as failing to propose alternative methodologies.
During the course of the proceedings, the Australian operator Telstra entered a
petition against the Benchmark Order on the grounds that it did not address the
issue of international internet connections. Telstra complained that the Order was
based on a circuit-switched environment, where traditionally each correspondent
operator is responsible for the provision of half of the international circuit. Telstra
argued, however, that in an internet environment non-US operators were effect-
ively forced to purchase a full-circuit in order to connect to the internet exchange
points based primarily in the US.155 As a consequence, US carriers were obtaining
significant financial benefits from the current arrangements for international
internet connections. The court denied Telstra’s petition as constituting insuf-
ficient grounds for overturning the FCC Order, but the issue was subsequently
pursued through the ITU.
In April 2000, ITU-T Study Group 3 approved a draft Recommendation on
‘International Internet Connection’ proposed by Australia. It was presented to
the World Telecommunication Standardization Assembly (WTSA) for adoption in
October 2000, but generated considerable opposition from the US and Europe over
concerns that the asymmetric nature of Web traffic flows would generate new pay-
ment imbalances and outflows. An amended version was eventually adopted at
WTSA, which recommended:
the possible need for compensation between them for the value of elements
such as traffic flow, number of routes, geographical coverage and cost of inter-
national transmission . . . (Recommendation D.50 (10/00) International Internet
Connection)156
This represented a shift from the mandatory wording of the draft, ie ‘will be com-
pensated’, to the possibility of compensation; although the US and Greece still
submitted reservations and stated that the Recommendation would not be applied
in their jurisdictions.
The international accounting rate system is gradually disappearing in its cur-
rent form to be replaced by a multitude of different arrangements reflecting the
state of liberalization in Member States, technological developments, and the
commercial positions of the respective parties. In the US, for example, by 2008
only around 6 per cent of international traffic billed in the US was settled in ac-
cordance with the accounting rate regime detailed in the ITRs, compared to 86 per
cent in 1998.157 Political pressure to accelerate such change has shifted somewhat
in recent years from the ITU to the WTO. A moratorium was agreed between cer-
tain Member States not to pursue a legal action before the WTO on accounting
rates,158 although that has not prevented accounting rate-related issues being ar-
gued before the Dispute Settlement Body.159
The WTO was established in 1994 as part of the final act embodying the results
of the ‘Uruguay Round’ of multilateral trade negotiations.162 The function of the
World Trade Organization is to facilitate the implementation, administration, and
operation of certain multilateral trade agreements (Article III(1)). One unique fea-
ture of the WTO is the establishment of a dispute settlement body to enforce the
158
See WTO Report of the Group on Basic Telecommunications (S/GBT/4), 15 February 1997.
159
See the Telmex case discussed at Section 16.4.5.1.
160
See, Annex to ITU (2005), World Summit on the Information Society Outcome Documents: Geneva 2003–
Tunis 2005, December 2005, Geneva.
161
ITU, ‘ITU Global Cybersecurity Agenda: Framework for International Cooperation in Cybersecurity’,
2007.
162
See the Agreement, Establishing the World Trade Organization with Understanding on Rules and
Procedures Governing the Settlement of Disputes and Trade Policy Review Mechanism (Marrakesh, 15 April
1994; TS 57 (1996) Cm 3277; 33 ILM (1994); OJ L 336/1, 23 December 1994). The Treaties entered into force on
1 January 1995.
82
163
See Section 16.4.5. 164
TS 56 (1996) Cm 3282; 33 ILM 28 (1994).
165
Ministerial Declaration on Trade in Information Technology Products (Singapore, 13 December 1996), at
<https://w ww.wto.org/english/docs_e/legal_e/itadec_e.htm>.
166
Ministerial Declaration on the expansion of trade in information technology products (WT/M IN/(15)/
25), Nairobi, 16 December 2015.
167
Questions from the European Union, Japan and the United States to India regarding Indian Customs
Notification No 11/2014 (G/I T/W/42), 4 April 2016.
168
TS 10 (1996) Cm 3046; 33 ILM 81 (1994).
829
Such a binary distinction and the accompanying definitions seems distinctly ar-
chaic given the nature of modern communications technologies, although they
have not seemingly created problems of interpretation within the WTO system.
Telecommunication services can also be distinguished into a number of categories
on the basis of geographical scope (ie local, long-d istance, and international);
mode of transmission (ie wire and wireless or radio-based); the use and owner-
ship of infrastructure (ie facilities-based or resale); and to whom the services are
provided (ie public or non-public).173 Some 108 Member States have made commit-
ments to liberalize trade in telecommunication services.
The GATS is concerned with four modes of supplying services:
1. from one territory to another, ie cross-border supplies;174
2. the provision to foreign consumers in the service provider’s territory, ie con-
sumption abroad;
3. the establishment of a commercial presence in another State; and
4. through the presence of a natural person in another State.175
169
The Agreement on Government Procurement is a plurilateral agreement under the WTO system, there-
fore only involving some members; currently 47.
170
TS 58 (1996) Cm 3276; 33 ILM 44 (1994). 171
MTN.GNS/W/120, 10 July 1991.
172
<http://w ww.wto.org/english/t ratop_e/serv_e/telecom_e/telecom_coverage_e.htm#basic>.
173
Ibid. 174
This concept was examined in the Telmex case at para 7.25 et seq.
175
GATS, Art I(2).
830
However, a Member may specify that this principle shall not be applicable to cer-
tain measures listed in an Annex on Article II Exemptions.176 Such MFN exemp-
tions are subject to review after a five-year period and should not exceed a period
of ten years.177
There has been some debate whether the MFN principle should operate in re-
spect of the international accounting rate regime (see Section 16.3.5), since
in non- competitive markets the amount an incumbent operator charges for
GATS, Art II(2).
176
GATS, Annex on Art II Exemptions, paras 5–7.
177
813
See GATT (1947), Art III, ‘National Treatment on Internal Taxation and Regulation’.
178
eg Asian countries.
179
832
The other key specific commitment under the GATS concerns ‘market access’
(Article XVI), under which Members detail those service sectors into which ser-
vice suppliers from other Members may enter.
The GATS permits members to derogate from these obligations, particularly the
non-d iscrimination provisions, on certain grounds, provided they are ‘necessary’
and are not applied in a manner that would constitute an arbitrary or unjustifi-
able discrimination or disguised restriction (Article XIV). The grounds include the
protection of public morals and public order, which could be used to justify the
imposition of network blocking, as well as the protection of personal data, which
could be relevant to data localization requirements or restrictions on transborder
data flows.180
As an instrument of public international law, the obligations and disciplines
contained within the GATS are strong, substantial, and impactful. However,
they are only triggered in respect of those service sectors that members choose to
commit to in their schedules, which remain relatively shallow, except in a few key
areas, such as telecommunications.
16.4.2 Telecommunications Annex
At the time of the GATS, Members also adopted a supplementary Annex on
Telecommunications. Its objective was to clarify the position of Members ‘with re-
spect to measures affecting access to and use of public telecommunications trans-
port networks and services’ (paragraph 1). The Annex is concerned with the supply
of any service over such public networks and services, including the basic tele-
communication services of another Member State,181 rather than any right to pro-
vide the networks and services. These obligations are incurred, therefore, whether
or not the Member has liberalized the provision of basic networks and services.
The Annex imposes obligations of transparency of conditions of access and
use, including tariffs, terms and conditions, and specifications of technical inter-
faces with the public networks and services (paragraph 4). The first draft of the
Annex stated that access and use should be on cost-orientated terms, but this
was removed in the face of opposition.182 Access should be ‘non-d iscriminatory’,
a term which embraces both the MFN and national treatment principles. Service
providers should be permitted to attach terminal equipment to the public net-
work; interconnect private circuits and utilize any operating protocols that do not
interfere with the availability of the public network (paragraph 5(b)). In terms of
restrictions, Members may only impose conditions that are necessary:
Such conditions may include restrictions on the resale of such services, compli-
ance with any ‘type-approval’ regime,183 or licensing and notification obligations. In
addition, developing countries may impose conditions ‘necessary to strengthen its
domestic telecommunications infrastructure and service capacity and to increase
its participation in international trade in telecommunications services’ (paragraph
5(g)). To assist the growth of telecommunications in developing countries, devel-
oped Members are encouraged to make available information and opportunities
concerning the transfer of telecommunications technology and training to the least-
developed countries.
183
See Chapter 4, at Section 4.4.3. 184
33 ILM 144 (1994).
185
GATS, Annex on Negotiations on Basic Telecommunications.
186
For a detailed history of the negotiations, see Sherman, L, ‘ “Wildly Enthusiastic” about the first multilat-
eral agreement on trade in telecommunications services’, (1999) 5(1)1 Federal Communications Law Journal,
pp 61–110.
187
As of 15 May 2017, this number had risen to 99 members, see <https://w ww.wto.org/english/t ratop_
e/serv_e/telecom_e/telecom_commit_exempt_l ist_e.htm>. The then 15 EU Member States submitted one
384
existing submissions made by Members and are annexed to the existing sched-
ules through a device referred to as a Protocol, which becomes an integral part of
the GATS (Article XX). As such, these submissions constitute the fourth Protocol to
have been entered into by certain Members of the WTO. The Fourth Protocol was
intended to enter into force on 1 January 1998; however, further delays meant that
it became effective on 5 February 1998.
Supplementary to the Schedules, the Chairman of the Group on Basic
Telecommunications issued two explanatory notes clarifying certain issues ap-
plicable to the scheduling of commitments. First, a ‘basic telecom service’ was de-
fined in the following terms:
(a) encompasses local, long-d istance and international services for public and
non-public use;
(b) may be provided on a facilities-basis or by resale; and
(c) may be provided through any means of technology (eg, cable, wireless,
satellites).188
Second, any qualifications referring to market access being limited due to the
availability of spectrum/frequency were compatible with the GATS and did not
need to be specifically noted.189
The ‘Basic Agreement’ has been seen as the most significant development in
the global liberalization of the telecommunications market. It has been estimated
that the Member countries represent over 90 per cent of global revenues in tele-
communications.190 The commitments made by Members encompassed market
access, foreign direct investment and, for the majority of Members, adherence to
a set of pro-competitive regulatory principles. The Protocol addressed the intro-
duction of competition into the four biggest bottleneck markets within telecom-
munications: satellite services, international public voice telephony, domestic
long-d istance, and the provision of the local loop.
In respect of the MFN exemptions, a number of countries specified accounting
rates as outside the scope the ‘Basic Agreement’, including India, Pakistan, Sri
Lanka, and Turkey. The US maintained a MFN exemption for DTH and DBS satel-
lite services to enable the continuation of existing ‘reciprocity’ regulations.
Schedule: see Annex to Council Decision (97/8 38/EC) of 28 November 1997 concerning the conclusion on be-
half of the European Community, as regards matters within its competence, of the results of the WTO negoti-
ations on basic telecommunications services; OJ L 347/45, 18 December 1997.
188
Note by Chairman, S/GBT/W/2/Rev.1, 16 January 1997.
189
Note by Chairman, S/GBT/W/3, 3 February 1997.
190
See Spector, PL, ‘The World Trade Organization Agreement on Telecommunications’, (1988) 32(2)
Summer The International Lawyer, pp 217–222.
853
16.4.3.1 Reference paper
One unique feature of the Fourth Protocol was the adoption of a ‘Reference Paper’
by 57 of the 69 Member signatories as an ‘additional commitment’ under GATS
(Article XVIII) and incorporated into the Schedules.191 The Reference Paper com-
prises a set of definitions and principles on the regulatory framework governing
the provision of basic telecommunications.192 The principles address particular
objectives for the establishment of a pro-competitive regulatory regime, rather
than the mechanisms or processes for their achievement. As such, the Reference
Paper represents an important body of international legal principles for the tele-
communications sector, of considerably greater significance than the ITU con-
stitutional principles.193 In addition, where a Member State has incorporated the
Reference Paper into its Schedule of Commitments, the principles are enforceable
before the WTO Dispute Settlement Body.
In terms of competition law, the Reference Paper firstly defines two key con-
cepts, ‘essential facilities’ and ‘major supplier’:
Essential facilities mean facilities of a public telecommunications transport net-
work or service that
(a) a re exclusively or predominantly provided by a single or limited number of
suppliers; and
(b) cannot feasibly be economically or technically substituted in order to provide
a service.
A major supplier is a supplier which has the ability to materially affect the terms of
participation (having regard to price and supply) in the relevant market for basic
telecommunications services as a result of:
(a) control over essential facilities; or
(b) use of its position in the market.
194
For US law, see MCI Communications v AT&T, 708 F 2d 1081 (7th Cir 1983), 464 US 891 (1983); for EU law,
see Case C-7/97 Oscar Bronner GmbH & Co KG v Mediaprint Zeitungs-und Zeitschriftenverlag GmbH & Co KG
and Others [1998] ECR I-7791. See further Chapter 10.
195
See further Chapter 5.
836
The first two substantive issues addressed in the Reference Paper concern
controls to be placed upon the ability of a ‘major supplier’ to be able to restrict
competition. First, a supplier who, alone or with others, constitutes a ‘major sup-
plier’ must be subject to ‘appropriate measures’ to prevent anti-competitive prac-
tices, whether current or future. Three specific anti-competitive practices are then
listed:
• cross-subsidization;
• the use of ‘information obtained from competitors with anti-competitive re-
sults’, such as the forecast traffic volumes in interconnection arrangements; and
• ‘not making available to other services suppliers on a timely basis technical
information about essential facilities and commercially relevant information
which are necessary for them to provide services’ (paragraph 1.2).
• although the need for, and form, of any regulator is not addressed, the Reference
Paper imposes an obligation upon a Member State to ensure that any such
regulator(s) are ‘separate from, and not accountable to, any supplier of basic
telecommunications services’ (paragraph 5);
• the allocation and use of scarce resources, ‘including frequencies, numbers and
rights of way’, should be carried out in an objective, timely, transparent, and
non-d iscriminatory way (paragraph 6).
Whilst the Reference Paper addresses ‘ends’ rather than ‘means’, its influence is
likely to be considerable at both a national and international level. First, as part of
the Schedules of Commitments, the Reference Paper represents a Member State
commitment to which foreign service providers may refer. Second, over time na-
tional legislators are likely to reflect and incorporate such principles into domestic
law. Third, the Reference Paper represents a baseline from which future multilat-
eral negotiations depart.
• whether the WTO agreements, and in particular the Reference Paper, may be
used in the interpretation and application of national or regional (eg EU) tele-
communications regulations; and
• whether the Reference Paper could be used as the basis for initiating proceed-
ings before a court in the event of a conflict with existing regulations, ie have
direct effect?
Within the European legal order, the Court of Justice has addressed the first
issue, that of interpretation, on a number of occasions. In Commission v Germany
(International Dairy Agreement)196, it was held that where the Community has en-
tered into an international agreement, the provisions of secondary Community
legislation ‘must, as far as possible, be interpreted in a manner that is consistent
with those agreements’ (paragraph 52). Further, in Hermès International v FHT
Marketing,197 the Court held that national courts, when interpreting a Community
measure that falls within the scope of a WTO agreement, must apply national le-
gislation ‘as far as possible, in the light of the wording and purpose’ of the agree-
ment (paragraph 28). Therefore, a court should consider the principles contained
in the Reference Paper when interpreting the application of European telecommu-
nications laws implemented in national law.
With regard to the second issue, that of WTO law having direct effect, all the
major trading nations have denied such an outcome,198 of which the EU is one ex-
ample, the final recital in the Community Decision adopting the WTO agreements
stating:
. . . by its nature, the Agreement establishing the World Trade Organization,
including the Annexes thereto, is not susceptible to being directly invoked in
Community or Member State courts.199
Despite this, the European Court of Justice has been required to consider the issue
of the status of WTO agreements on a number of occasions, most significantly in
Portugal v Council.200 First, the Court addressed the status of the WTO agreements
in the legal order of the Member States, concluding that:
. . . the WTO agreements, interpreted in the light of their subject-m atter and
purpose, do not determine the appropriate legal means of ensuring that
they are applied in good faith in the legal order of the contracting parties.
(paragraph 41)
Second, their status within the Community legal order was examined. The Court
considered that the WTO agreements were based on the ‘principle of negotiation’
which distinguished them from other international agreements that were recog-
nized as having direct effect (paragraph 42). The Court also noted that the EC’s
major trading partners did not give direct effect to the agreements, which would
effectively disadvantage the Community in future negotiations. Therefore, the
Court concluded that:
197
[1998] ECR I-3603.
198
Ruiz-Fabri, H, ‘Is there a Case—L egally and Politically—for Direct Effect of WTO Obligations’, (2014)
25(1) Eur J Int Law 151–173.
199
Final Recital in Council Decision 94/8 00/EC, of 22 December 1994, concerning the conclusion on be-
half of the European Community, as regards matters within its competence, of the agreements reached in the
Uruguay Round multilateral negotiations (1986–1994) OJ L 336/1, 23 December 1994.
200
[1999] ECR I-8 395. See also Case C-93/02 Biret International v Council [2006] 1 CMLR 17, where the court
confirmed the existing position, but did leave open the possibility of private claims against EU institutions
based on EU measures that are found to violate WTO law by the Dispute Settlement Body, a position which had
been suggested by Advocate General Alber [2003] ECR 10, at para 24.
839
the WTO agreements are not in principle among the rules in the light of which
the Court is to review the legality of measures adopted by the Community institu-
tions. (paragraph 47)
The Court’s reasoning in this case has been heavily criticized for undermining the
status of the WTO agreements.201 However, the Court did confirm its previous juris-
prudence that the GATT rules could have direct effect where either the adoption of
the measures implementing obligations assumed within the context of the GATT
is at issue; or a Community measure refers expressly to specific provisions of the
general agreement (paragraph 111).202 In this regard, it is interesting to note that
the European Commission’s 2002 package of measures in the telecommunications
sector, make explicit reference to the commitments made by the Community and
its Member States in the context of the Fourth Protocol to the GATS.203
In terms of UK law, the general applicability of the WTO agreements has been
somewhat uncertain due to a lack of clarity as to which aspects of the ’mixed agree-
ments’ fall within the competence of the Community, as opposed to the individual
Member States.204 The problems raised by such joint competence were examined
inconclusively in a dispute brought by the US against the Community, the UK, and
Ireland, in 1997, in respect of the tariff classification of Local Area Network equip-
ment.205 Post-L isbon, the EU’s competence in the area of trade in services (TFEU,
Article 207(1)), seems sufficiently extensive to address all GATS-related matter,
including the provision of telecommunications services and networks.206
In the absence of direct effect, either under European or national law, the only
mechanism under which a party could seek enforcement against a Member State
for failure to comply with their obligations in respect of the telecommunications
sector is through the WTO Dispute Settlement Body.
The UK’s intended departure from the EU places the status of the WTO agree-
ments back into the limelight, as the UK will be required to submit its own ‘sched-
ules’ once it is no longer part of the EU’s. The UK government has announced its
201
See generally Zonnekeyn, G, ‘The status of WTO Law in the EC Legal Order’, (2000) 34(3) Journal of World
Trade Law pp 111–125; and Griller, S, ‘Judicial Enforceability of WTO law in the European Union: Annotation
to Case C-149/96, Portugal v Council’, (2000) 3(3) Journal of International Economic Law pp 441–472.
202
See Case C-2 80/93 Germany v Council [1994] ECR I-4973, paras 103–112.
203
eg Directive 02/21/EC on a common regulatory framework for electronic communications networks and
services, OJ L 108/33, 24 April 2002 at Recital 29.
204
See Opinion 1/94 of the Court of Justice [1994] ECR I-5267.
205
Customs Classification of Certain Computer Equipment, WTO doc. series WT/DS62, WT/DS67 and
WT/DS68. See also Heliskoski, J, ‘Joint Competence of the European Community and its Member States and
the Dispute Settlement Practice of the World Trade Organization’ in (1999) 2 The Cambridge Yearbook of
European Legal Studies, pp 61–85.
206
See Opinion 2/15 (C-376), 16 May 2017 re: Singapore FTA. See also Klamert, K, Services Liberalisation in
the EU and the WTO (Cambridge University Press, 2015).
804
intention ‘to replicate our existing trade regime as far as possible’,207 although this
is dependent on ‘certification’ by the other 163 members. While objections are un-
likely to arise in respect of the telecommunications sector per se, disagreements in
other areas may cause substantial delay in the whole process.
However, it would not seem appropriate to characterize the DSB as a judicial body.
The Panel shall comprise three individuals chosen by the DSB secretariat with the
consent of the parties. In the absence of agreement, the Director-General may ap-
point the panellists. After an investigation, the Panel submits a report to the DSB
for consideration, detailing the Panel’s findings and conclusions. The DSB will
usually adopt the panel report unless one of the parties notifies the DSB of its in-
tention to lodge an appeal to the Appellate Body (Article 17). The Panel or Appellate
Body will decide whether a particular Member State measure is inconsistent with
the terms of the relevant agreement and may recommend ways of overcoming the
207
Statement by Julian Braithwaite, FCO, ‘Ensuring a smooth transition in the WTO as we leave the EU’, 23
January 2017, <https:// blogs.fco.gov.uk/julianbraithwaite/2017/01/2 3/ensuring-a-smooth-t ransition-i n-t he-
wto-a s-we-leave-t he-eu/>.
208
Understanding, n 162. See generally, Merrills, JG, International Dispute Settlement (3rd edn)
(Cambridge: Cambridge University Press, 1998).
209
The dispute settlement system under GATT 1947 was essentially a conciliation procedure.
210
Ibid, at Appendix 1.
814
issue. A Member, against whom a decision has been reached, is obliged to imple-
ment the recommendations and rulings of the DSB within a reasonable period of
time (Article 21).
In the event that a Member fails to comply, the Understanding allows for the
payment of compensation or the suspension of concessions (Article 22). The ability
to suspend trade concessions granted to an infringing Member is the real stick
within the dispute settlement procedure under the WTO. A complaining party
may be able to suspend concessions or obligations not only in the sector of dispute
(eg telecommunications), but also, where appropriate, in other sectors under the
same agreement (eg GATS), or even under another covered agreement. Any such
concession must be authorized by the DSB and should be ‘equivalent to the level of
the nullification or impairment’ (Article 22.4).
Whilst the WTO dispute procedures are between governments, industry ob-
viously plays an important role in bringing such matters to the attention of gov-
ernments. Under European law, complaints may be submitted in writing to the
Commission and a formal examination procedure may be invoked prior to the
decision to pursue a dispute.211 In the US, the Office of the United States Trade
Representative (USTR) is required to solicit comments from industry when con-
ducting its annual analysis of the operation and effectiveness of any trade agree-
ment regarding telecommunications products or services and determining any
action.212
The dispute settlement procedures have so far been invoked in respect of very
few disputes in the telecommunications sector. Formal proceedings before the
DSB have been pursued by the European Commission against Korea213 and Japan
in respect of preferential trade practices in favour of US suppliers of telecommu-
nications equipment, both of which were resolved by agreement.214 Proceedings
have also been brought by the US against Belgium, regarding telephone directory
services,215 which was settled. The only case to reach a Dispute Panel and a formal
decision was a claim made by the US against Mexico, the so-called ‘Telmex case’,
discussed at Section 16.4.5.1.
211
See Council Regulation (EC) No 3286/94 of 22 December 1994 laying down Community procedures in
the field of the common commercial policy in order to ensure the exercise of the Community’s rights under
international trade rules, in particular those established under the auspices of the World Trade Organization;
OJ L 349/71, 31 December 1994 (as amended by Regulation (EU) No 654/2014). To date, some 24 ‘trade barrier
regulation’ complaint procedures have been initiated.
212
19 USC § 3106 and 3108.
213
WT/DS40 ‘Korea—L aws, regulations and practices in the telecommunications procurement sector’, 5
May 1996. See also Agreement on telecommunications procurement between the European Community and
the Republic of Korea; OJ L 321/32, 22 November 1997.
214
WT/DS15 ‘Japan—Measures affecting the purchase of telecommunications equipment’, 18 August 1995.
215
WT/DS80 ‘Belgium—Measure affecting commercial telephone directory services’, 13 May 1997.
824
16.4.5.1 Telmex
The Telmex case concerned a preferential arrangement between Telmex, the
Mexican incumbent, and the US operator Sprint. Other US operators, such as AT&T
and MCI, complained to the US Government that this arrangement was discrim-
inatory, and therefore in breach of Mexico’s commitments under the GATS, the
Telecommunications Annex, and the Reference Paper. Following the lodging of a
formal complaint before the WTO, the Mexican regulator, Cofetel, issued new re-
gulations requiring Telmex to terminate the preferential arrangement and provide
non-d iscriminatory treatment to all foreign long-d istance operators. Despite this,
the US decided to proceed with its request to the DSB for the establishment of a
panel, which was duly formed in August 2002. The Panel was required to make de-
terminations on a number of issues, both of fact and law, interpreting the various
WTO agreements, as well as broader issues of international telecommunications
law.220
In terms of findings of fact, the ‘relevant market’ was disputed, with Mexico ar-
guing that the operation of a traditional accounting rate regime for international
calls meant that the ‘relevant market’ had to be two-way traffic, not just the termin-
ation of communications into Mexico, as argued by the US.221 The Panel accepted
US evidence that demand substitution was essential to the market definition
216
See 1998 Annual Report of the President of the United States on the Trade Agreements Program, at 257.
217
See ‘US warns on German telecoms’, Financial Times, 12 August 1999. See also 1999 Annual Report,
at 293.
218
eg ‘US uses WTO threat to challenge Japanese pricing’ (20 September 1999): <http://w ww.totaltele.com>.
219
See USTR Press Release: ‘United States and Japan agree on interconnection rates’, 18 July 2000.
220
See ‘Mexico—Measures affecting Telecommunication Services’, Report of the Panel, WT/DS204/R, 2
April 2004.
221
Ibid, at paras 4.151–4.158.
834
process and that an outgoing call was not a substitute for an incoming call.222 In
terms of market power, the Panel concluded that Telmex was a ‘major supplier’ on
the basis of its position under applicable domestic rules, which granted Telmex the
right ‘to negotiate settlement rates’ for the entire Mexican market.223
On matters of law, one fundamental issue to be determined was whether conduct
of a major supplier could be considered ‘anti-competitive’ if such conduct was re-
quired by law. Surprisingly, the European Commission, as a third party to the pro-
ceedings, supported Mexico’s position that State rules could not be considered an
anti-competitive practice. However, the Panel held that ‘a requirement imposed by a
Member State under its internal law on a major supplier cannot unilaterally erode its
international commitments’ made under GATS and related measures.224
The Panel concluded that Mexico had failed to meet its commitments under both
the Annex on Telecommunications and the Reference Paper. Under the Annex,
Mexico had failed to comply with Articles 5(a) and (b) in respect of access to and use
of the ‘public telecommunications transport networks’, on a facilities basis, on rea-
sonable and non-discriminatory terms. Under the Reference Paper, Mexico’s obliga-
tions to maintain ‘appropriate measures’ preventing anti-competitive practices (at
1.1) were held to have not been met, as well as its obligations to ensure that Telmex
provided interconnection at ‘cost-orientated rates’ (at 2.2(b)). However, since Mexico
had not made commitments for non-facilities based services, it was found not to have
violated any of its obligations in respect of such services.
Both sides in the dispute had reason to be unhappy with aspects of the
Panel’s conclusions, but neither party chose to appeal and, in June 2004, the par-
ties reached an agreement resolving the dispute;225 with Mexico subsequently
amending its resale regulations in August 2005 in full compliance with the DSB’s
recommendations.
222
Ibid, at paras 7.149–7.152. 223
Ibid, at paras 7.153–7.155. 224
Ibid, at para 7.244.
225
WT/DS204/7, S/L/162, 2 June 2004.
84
greater significance for the legal and regulatory frameworks of developed coun-
tries.226 Third, while developing nations have adopted GATS-compliant regulatory
frameworks ‘on the books’, often with expert input from developed nations funded
by development organizations, regulatory performance ‘on the ground’ remains
poor.227
As already noted, the process of trade liberalization under the WTO regime is an
ongoing one, with multinational negotiations attempting to broaden and deepen
the commitment of Member States to free trade. The current round of negotiations
formally commenced at Doha, Qatar, in November 2001.228 In parallel with these
multilateral negotiations, Member States are negotiating and entering into re-
gional and bilateral trade agreements with trading partners, at a level that gener-
ally goes beyond that which States are prepared to commit at a multinational level.
Telecommunications forms a component of the current round, with the major
industrialized countries calling upon other countries to make commitments to
fully liberalize and the ‘elimination of MFN exemptions for telecommunication
services’.229 Currently, proposals either comprise offers to improve existing com-
mitments or to make an initial commitment to telecommunications liberaliza-
tion.230 In the current international political climate, further progress on trade
liberalization has largely stalled, while the ‘Doha Round’ has effectively come to
an end. However, the telecommunications sector has already made substantial
progress towards full liberalization and the current agreements have fundamen-
tally altered national and international telecommunications law.
226
Henderson, A, Gentle, I, and Ball, E, ‘WTO Principles and Telecommunications in Developing
Nations: Challenges and Consequences of Accession’, (2009) 29 Telecommunications Policy 205.
227
Djiofack- Z ebaze, C and Keck, A, ‘Telecommunications Services in Africa: The Impact of WTO
Commitments and Unilateral Reform on Sector Performance and Economic Growth’, (2009) 37(5) World
Development 919.
228
WTO Ministerial Declaration, 14 November 2001 (WT/M IN(01)/DEC/1). See also Chapter 17, at Section
17.4.1 for a discussion of competition policy within the Doha Round.
229
TN/S/W/50, ‘Communications from Australia, Canada, the European Communities, Japan, Hong Kong
China, Korea, Norway, Singapore, the Separate Customs Territory of Taiwan, Penghu, Kinmen, and Matsu and
the United States’, 1 July 2005.
230
As of July 2008, some thirty-n ine governments had made such offers; see <https://w ww.wto.org/english/
tratop_e/serv_e/telecom_e/telecom_e.htm>.
854
17
TELECOMMUNIC ATIONS R EF OR M IN
EMERGING MAR KE T S
Ann Buckingham, Camilla Bustani, David Satola,1 and Cameron Whittfield1a
17.1 INTRODUC TION
For much of the latter half of the twentieth century, the provision of telecom-
munications services in developing countries was the responsibility of the state.
Penetration rates were low, service quality was poor, and the incumbent operator
was often unprofitable. Government attempts at improvement focused on securing
investment, technical assistance, and financial support. Commencing in the late
1990s, the telecommunications sector in much of the developing world underwent
a profound and lasting transformation: governments established new regulatory
regimes and institutions, corporatized and privatized their state-owned telecom-
munications operations, and liberalized markets as part of wide-ranging sector
reforms.
As a result of this initial wave of sector reform, global teledensity and pene-
tration materially increased. Global mobile subscriptions alone are now esti-
mated to number in excess of seven billion (with mobile broadband subscriptions
1
The author is Lead Counsel, the World Bank. The views expressed are those of the author and not neces-
sarily those of the World Bank, its Board of Directors, or the countries they represent.
1a
The authors wish to acknowledge Keong Min Yoon, World Bank Legal Department, for his assistance in
preparing this chapter.
84
continuing to grow at double digit rates), and penetration rates are at 99.7 per cent
globally and 94.1 per cent in the developing world.2
With the primary objectives of these initial reforms largely met, the attention
of regulators has increasingly turned to second-generation reforms that address
twenty-fi rst century issues of connectivity, broadband access, and convergence.
Moreover, convergence, compounded by the increasing reliance on broadband
capacity, has focused regulatory attention on issues such as digital authentica-
tion, cloud services and storage, data security, privacy, access to (and freedom of)
information and cybercrime, which were previously viewed as ancillary to, ra-
ther than part of, the core telecommunications legal and regulatory framework.
In developing countries, the International Telecommunication Union (ITU) esti-
mates that more than 50 per cent of households worldwide have internet access
(including 41 per cent of households in the developing world).3 The pre-eminence
of the internet as a central feature of commerce, education, and social interaction
is increasingly changing the drivers of regulatory reform. Additionally, the rise of
the ‘smart phone’, and associated bandwidth heavy applications, is also impacting
regulatory design.
This chapter will address the various factors providing the impetus for reform,
the foundational components of reform, the impact on regulation and market
reforms—i ncluding an evolving role of the state—i n the transition to broadband-
enabled networks and services and, finally, the evolving second-generation re-
forms being undertaken in the developing world to address the increasing role
and importance of internet access and broadband communications to economic
development and growth.
2
ITU World Telecommunication/ICT Indicators Database, at <http://w ww.itu.int/en/I TU-D/Statistics/
Pages/facts/default.aspx>. In contrast, in 1995 the number of global mobile subscriptions was 90.7 million
and the penetration rate was 1.585%. <https://data.worldbank.org/i ndicator/I T.CEL.SETS>.
3
ITU World Telecommunication/ICT Indicators Database, at <http://w ww.itu.int/en/I TU-D/Statistics/
Pages/facts/default.aspx>.
894
4
Exceptions include the Caribbean, where in most of the English-speaking islands, Cable & Wireless (until
the early part of the twenty-fi rst century) enjoyed a private monopoly.
5
See eg Smith, W, ‘Utility Regulators—Roles and Responsibilities’, Viewpoint Note No 128, the World Bank
(October 1997).
580
6
Defined as including fixed voice telephony and other core telecommunications services.
7
eg the schedule of commitments of Lao People’s Democratic Republic and Kazakhstan (which became
WTO members in 2013 and 2015, respectively) required the liberalization of certain segments of their tele-
communications sector within 5 years and 2.5 years, respectively. Cape Verde (2008), Tonga (2007), Ukraine
(2008), and Viet Nam (2007) also submitted specific commitments in telecommunications. These vary in
terms of scope of market opening, and two (Ukraine and Viet Nam) have incorporated the Reference Paper
in their commitments.
8
The Reference Paper is available at: <http://w ww.wto.org/english/t ratop_e/serv_e/telecom_e/tel23_
e.htm>.
9
In this section, we use the term ‘multilateral development institution’ to cover both international finan-
cial institutions (IFIs) and national bilateral aid agencies or programmes providing grant funding for tech-
nical assistance and other forms of support. IFIs include members of the World Bank Group (eg International
Bank for Reconstruction and Development (IBRD)), the International Development Association (IDA), and
the African Development Bank (AfDB). When referred to in this chapter, the World Bank means the IBRD and
the IDA.
815
10
Corporatization and commercialization of the state-owned incumbent (involving the transformation of
the legal structure and operations of the incumbent from a government department to a corporate enterprise)
are usually critical steps to ensure the incumbent is able to compete in a liberalized environment and to en-
able its subsequent privatization.
852
and political objectives of the reform process, as well as the specific characteristics
of the subject country.
In most countries, reform objectives have broadly fallen into three
categories: generating the social benefits inherent in telecommunications
services, encouraging economic growth and financial investment, and developing
industry-specific expertise. Improving access to, and the quality of, telecommu-
nications services is at the heart of the first objective, reflecting the nature of tele-
communications services as a public good. If a country’s telecommunications
infrastructure is inadequate to secure that access, or where services are inef-
ficient, low-quality or costly, sector policy will often require private investors to
make commitments regarding network roll-out, service coverage, increases in
subscriber lines, and quality of service.
Secondly, governments also see a robust and competitive telecommunications
sector as a driver of economic growth. A healthy telecommunications sector also
stimulates technology investment in other industries that rely on telecommuni-
cations, including banking, outsourced services, and call centres. Governments
usually also wish to encourage foreign investment, typically as investors in (or
managers of) both the incumbent PTT and new entrants. Governments often
benefit significantly from privatizations and liberalization (with both licence and
radio spectrum fees generating substantial revenues).11 Governments may seek to
strengthen local investment institutions by requiring local initial public offerings
(IPOs) or domestic shareholdings as conditions to market entry.
Finally, in addition to capital, local capacity building is often a key government
objective. Foreign investors generally bring know-how and technology, and their
presence can be pivotal in management and service improvements. Bids for li-
cences can be won or lost on an investor’s ability or willingness to make these in-
tangible contributions.
In addition to the market reform objectives above, the design of regulatory
interventions must be tailored to the circumstances of the country undertaking
reform. In particular, common ‘prioritization’ or functional differences between
developed and developing countries often mean that the regulatory approach
undertaken in developed countries needs significant modification to reflect local
realities.
A key difference is the relative importance of mobile networks. In many
developing countries, the lack of capacity, low penetration, and poor quality of
fixed networks mean that mobile networks are the primary means of access to
11
The total investment from telecommunication with private participation in IDA countries from 2000 to
2014 was $55,967,485,000. See, World Development Indicators 2017, World Bank, 2017 <https://openknowledge.
worldbank.org/handle/10986/26447>.
853
12
In 2016, the average worldwide total number of mobile subscriptions per 100 inhabitants was 99.7 and
fixed telephone subscriptions was 14.3, whereas for developing countries these figures were 93 and 9.3 re-
spectively. See ITU World Telecommunication/ ICT Indicators database: <http://w ww.itu.int/en/I TU-D/
Statistics/Pages/facts/default.aspx>.
13
Some developing countries, such as Timor-L este, emphasize ‘access’ in a technology-neutral way, recog-
nizing that access is more important than the type of network providing service. See, eg, Timor-L este, Decree
Law No. 15/2012 of 28 March 2012, On the Regulation of the Telecommunications Sector, §57.4.
584
14
ITU, ‘Trends in Telecommunication Reform 2010–11: Enabling Tomorrow’s Digital World Summary’,
Geneva, 2011.
15
See eg country level data in The Little Data Book on Information and Communication Technology 2017,
World Bank, 2017. <https://openknowledge.worldbank.org/bitstream/handle/10986/25737/9781464810282.
pdf>. While this is also true in some developed countries, usually institutional separation is less clear and
political involvement in operational decision-making more pervasive in the developing world.
16
In Lebanon, for example, a ‘new’ sector law (Law 431/2002 of July 2002) was adopted but only partially
implemented. Provisions regarding corporatizing the telecommunications arm of the Ministry and transfer
of full regulatory powers to the regulatory agency created under the law are not fully in place.
85
17
The US model has been emulated in some countries, particularly in Latin America. However, the com-
plexities of the interplay of federal and state regulation, the impact of US constitutional principles, and the
lack of legacy state ownership in the sector mean that US regulatory structures are generally less relevant to
the design of regulatory frameworks in developing countries.
18
The directives which together form the EU regulatory telecommunications framework are currently
being reviewed and recast into a single directive, the European Electronic Communications Code, currently
expected to be adopted in mid-2018 and transposed into the national law of EU Member States over the sub-
sequent 18 months.
586
package’), was designed, among other things, to respond to the challenges posed
by the growth in broadband internet usage, next-generation networks, and net-
work security, including strengthening provisions on infrastructure sharing and
consumer protection. The EU framework is now once again under review, the
guiding principles of which are the promotion of ‘very high-capacity’ broadband
connectivity, attempts at centralizing spectrum management at EU level and po-
tentially bringing into the scope of consumer regulation certain ‘over the top’ or
online communications services that are currently outside the scope of telecom-
munications regulation.
Central to the 1998 package were the requirements to establish a national regu-
latory authority, to adopt a new regulatory framework designed to control anti-
competitive conduct by the incumbent, and to manage the transition from monopoly
to full competition. The reforms introduced by the 1998 package were based on the
principle of ‘open network provision’, which emphasized that access to and use of
public telecommunications networks and services should be unrestricted, except
where limited by non-economic reasons in the general interest such as network in-
tegrity and security. The principles of objectivity, proportionality, transparency, non-
discrimination, and regulatory independence were required to underpin Member
States’ regulatory frameworks.
The 1998 package is most germane to developing countries designing first-gen-
eration reforms. The 2002 and 2009 packages are often less relevant, as many of the
‘lighter touch’ regulatory principles or regulatory exit strategies assume a backdrop
of robust competition law to control anti-competitive market behaviour. However,
elements of both the 2002 and 2009 packages might be relevant to address spe-
cific second-generation issues, such as convergence, broadband access, network
security, and consumer protection, as discussed more fully in Section 17.5 below.
In addition, certain aspects of the current EU framework review might also be
relevant, particularly with respect to ensuring appropriate consumer protections
in relation to online communications services and promoting broadband access
(even though the scope and level of broadband access sought might be lower in
developing countries).
An effective legal and regulatory framework is essential to regulate the sector once
state control (through ownership of a monopolist incumbent) is relinquished, and
to attract private investment into the telecommunications sector. However, there
is no ‘one-size-fits-a ll’ approach. Rather, as described above in Section 17.2, each
individual country must carefully analyse its specific market characteristics, legal
foundations, regulatory and institutional capacity, and any political realities or
tensions that might influence or shape a particular regulatory approach. The key
to successful implementation of a new regulatory framework in developing coun-
tries is to combine the lessons learned from international experience with a deep
understanding of local circumstances and priorities. Yet, despite the best of in-
tentions and well-structured regulatory parameters, true regulator independence
remains a challenge in many developing countries.
19
ECOWAS Supplementary Acts of Telecommunications, Information and Communication Technology
(ICT) Sector are available at <http://w ww.comm.ecowas.int/dept/stand.php?id=f_ f 1_act_add&lang=en>;
SADC Protocol on Transport, Communications and Meteorology is available at <http://w ww.sadc.int/key-
documents/protocols/protocol-on-t ransport-communications-a nd-meteorology/>.
20
For a thorough review of regional telecommunications initiatives in Africa, see Report on ICT Initiatives
and Research Capacity in IST-A frica Partner Countries, IST-A frica (2016).
21
See, Schwarz, T and Satola, D, ‘Telecommunications Legislation in Transitional and Developing
Economies’ World Bank Technical Paper No 489 (Washington DC, the World Bank, 2000), at 13–17 (Schwarz
and Satola).
58
22
See eg the experience of Niger, where Telecel’s ‘unchangeable’ comprehensive licence was repeatedly
revoked for political reasons in the late 1990s.
859
23
See Ungerer, H, ‘Access Issues under EU Regulation and Anti-t rust Law: The Case of Telecommunications
and internet Markets’, Incidental Paper, The Program on Information Resources Policy, Harvard University and
the Center for Information Policy Research (2000) (available at <http://w ww.pirp.harvard.edu>) for a com-
prehensive discussion of the increasingly important interplay between sector-specific legislation and general
competition legislation.
24
The principle of ‘regulatory forbearance’ directs the FCC to forbear from applying any provision of the
Telecommunications Act of 1996 where analysis of the relevant market leads the FCC to conclude that forbear-
ance would not harm consumers and is generally in the public interest.
25
See eg the Nigerian Communications Act 2003, Chapter VI, Part I.
26
See the telecommunications law of the Solomon Islands (Telecommunications Act 2009 (No 20 of 2009)),
which adapts the principles familiar in the EU framework to local circumstances (see Section 60 et seq).
860
27
Benchmarking, while imperfect, can enable a nascent regulatory authority to provide some guidance
to market participants, and is similar to the rebuttable presumption of dominance for operators with greater
than 25% market share that was adopted in the EU 1997 Interconnection Directive.
28
See n 26, at Section 70, et seq.
816
29
<http://w ww.itu.int/en/I TU-D/Statistics/Pages/l inks/nta.aspx>.
30
The concept of ‘independence’ is mutable, being affected by heterogenous factors apart from formal in-
stitutional and legal considerations, including type of underlying political system, maturity of institutions in
developing countries, the history of institutions acting autonomously from ‘government’, etc. For example,
the UK’s Ofcom reports to the Secretary of State for Industry, but in reality is fully independent. See, also,
Northfield, D, ‘Global Trends in Communications Regulatory Structures’, (2001) 1 Global Regulatory Strategies
No 5–6, the Yankee Group (September–October).
31
In this chapter, independence refers to the collection of attributes about a regulator that allow it to func-
tion independently.
862
the telecommunications line ministry, such as the ministry responsible for the
economy or finance. By separating the policy function from ownership of the
incumbent, sector policy is more likely to be formulated in the wider national
interest, even where this might conflict with the interests of the state-owned
incumbent.
The remit of the new regulatory authority must also be clearly defined. Most com-
monly, the regulatory authority is given responsibility solely for the telecommuni-
cations sector.32 However, a number of countries have established multi-sectoral
regulators. An assessment of the advantages and disadvantages of a sector-specific
versus multi-sectoral regulator must be made on a case-by-case basis, taking into
account which model best protects against industry and political capture, which
is most advantageous for leveraging economies of scale, as well as which sectors
are to be covered.33
Multi-sectoral models are more common in smaller countries, particularly
where human and financial resources are scarce. Economies of scale and scope
can be achieved through sharing a common pool of legal, regulatory, financial, and
economic expertise, and administrative overhead. Multi-sectoral models can also
be useful where industries are dominated by a single player, as the incumbent’s
influence might be diminished where the regulatory authority has jurisdiction
over multiple industries. Moreover, where the regulated sectors fall within the re-
sponsibility of different line ministries, the use of a multi-sectoral model can re-
duce the risk of political capture.
A multi-s ectoral regulator might be responsible for a range of unrelated
utility sectors that share significant commonalities, such as the Public
Utilities Commission of Latvia, which has jurisdiction over the telecommu-
nications, postal, energy, water management, and waste disposal sectors. 34
Conversely, creating a regulator of related ‘converged’ sectors can facilitate
the implementation of a technology-neutral regulatory framework, avoid
disputes over the delineation of regulatory authority in related sectors, and
better position the regulator to adapt to technological change. A good ex-
ample is the Malaysian Communications Multimedia Commission (MCMC),
which is responsible for regulating the telecommunications, broadcasting,
multimedia, e-c ommerce, and postal sectors, and is also the certification au-
thority for digital signatures.
The legislation creating a regulatory authority should specify the scope of its au-
thority and powers. Sector-specific powers should cover licensing, interconnection
32
ICT Regulation Toolkit, <http://w ww.ictregulationtoolkit.org/en/P racticeNotes.html#1254> (Toolkit).
33
See eg Schwarz and Satola, n 21.
34
‘Driving Performance at Latvia’s Public Utilities Commission’ (Paris: OECD Publishing, 2016).
863
35
Conversely, in the Pacific Islands, recently formed regulators (eg Samoa, Solomon Islands, and Vanuatu)
have relied on the single regulator model, in part due to human and economic resource constraints.
36
ICTEYE Database, <http://w ww.itu.int/net4/itu-d/icteye/Default.aspx>.
37
Telecommunications Regulation Handbook, 21 (The World Bank, infoDev and the ITU, 2011).
684
authority. Commissioners’ terms should preferably be fixed and end dates stag-
gered so as to reduce the influence of any one government over their appointment,
to encourage policy continuity, and to preserve institutional memory. Generally,
full-t ime members are preferable to part-t ime members, to ensure that individual
commissioners are able to give sufficient time and focus to their regulatory duties.
However, staffing difficulties or poor remuneration may might make a part-t ime
structure more practicable.
Individual commissioners or board members should have no actual or per-
ceived conflict of interest. The telecommunications law should include controls
on conflicts of interest (applicable to both commissioners and their direct fam-
ilies), and bar the appointment and authorize the removal of commissioners
with a clear conflict of interest (eg due to political involvement or material in-
vestments or involvement in market participants). Often commissioners are re-
quired to file a statement on conflicts of interest on an annual basis, so that
compliance can be monitored. However, conflict of interest provisions should
not be so broad as to prevent skilled and experienced individuals from being
selected. This can sometimes be a difficult balance to achieve in a country
lacking a pool of experts in the field of telecommunications, where the initial
staff of the regulator might come directly from the incumbent and/or the line
ministry. Conflicts might also arise on a day-to-day basis, where individual
matters before the regulatory authority raise a potential conflict of interest. In
these circumstances, the procedures of the regulatory authority should require
affected commissioners and staff members to disclose this conflict and to ab-
stain from any related decision-m aking.
Independence is also enhanced if the regulator is financially autonomous
from the telecommunications ministry and from government budget alloca-
tions. Therefore, regulatory authorities are most commonly financed through
an industry-w ide levy, typically imposed on sector participants through an
annual licence fee (see Section 17.3.5). At the same time, efficiency and trans-
parency are fostered by requiring the regulator to prepare formal budgets
and audited accounts, which should be filed before Parliament or another
overseeing authority, and by requiring the regulator to limit industry levies
and licence fees to no more than the level needed to finance its budgeted
running costs.
Transparency, predictability, and impartiality of regulatory decision-making
are further enhanced where the regulatory authority conducts broad public con-
sultation prior to making regulatory decisions or publishes guidelines in advance
setting out the regulatory authority’s proposed approach, preferably based on
objective criteria, and where regulatory decisions are subject to a clear right of
appeal.
856
38
The Toolkit is available at <http://icttoolkit.infodev.org/en/i ndex.html>. The Regulators’ decisions data-
base can be accessed at: <http://web.archive.org/web/20130606011411/http://w ww.ictdec.org:80/ictdec-web/
en/i ndex.html>.
39
ECTEL has actual regional (ie supranational) regulatory powers (<http://w ww.ectel.int/>).
40
Pacific ICT Regulatory Resource Center <https://pirrc.org.fj/>.
86
17.3.5 Licensing
Regulatory frameworks typically require operators of telecommunications net-
works and/or providers of telecommunications services to obtain some form of
licence or authorization, or alternatively to register with or notify the regulatory
authority, prior to commercial service launch (which, for simplicity, we collect-
ively refer to as ‘licences’ in this section). As licences control market entry, good
regulatory practice dictates that both the licensing procedures and licence docu-
ments be straightforward, clear, and streamlined and be issued on the basis of ob-
jective licensing criteria. Licensing mechanisms that lack these features can deter
new entrants, hamper sector development, encourage corruption, or increase the
risk premium attached to investment in a country’s telecommunications sector.
The regulatory regime must specify the issuing authority for licences. In some
jurisdictions, the constitution might require licences to be issued by a minister,
or some form of ministerial control might be sought for constitutional or political
reasons. Ministerial (ie ‘political’) discretion over licensing is undesirable and cre-
ates another layer in the licensing process which is likely to delay or complicate
licence grant. Where this additional layer is unavoidable, the scope of the minis-
terial power to veto a licence grant should be specified explicitly and be limited to
matters of genuine governmental concern (eg national security), with time limits
for ministerial intervention. Regulatory discretion in licence allocation should
also be kept to a minimum, both to help protect the licensing system from corrup-
tion, and also to minimize licensing delay.
Licence availability should reflect government liberalization policy. As a gen-
eral principle, licence numbers should be unlimited and available upon request,
except where there are real constraints, such as the availability of radio frequency
spectrum. To the extent possible, temporary bottlenecks due to numbering or
spectrum issues should be resolved by a revision of numbering or spectrum plans,
rather than by curtailing market entry.
Regulatory bottlenecks can be reduced by limiting formal licences to those re-
quired to operate key public networks (eg fixed and mobile network infrastructure
licences, international gateways, and frequency usage). Entities wishing to pro-
vide value-added, internet, or data services or to operate private networks could be
under an obligation simply to register their activities with the regulatory authority,
41
Pacific Islands Legal Information Institute <http://w ww.paclii.org/>.
867
42
See NCC, ‘Unified Access Service’, <http://w ww.ncc.gov.ng/component/docman/doc_download/45-
unified-access-service.html>.
43
In practice, some countries which purport to have adopted a ‘unified’ licensing regime still restrict the
types of networks or services operated and provided, as the actual licence may itself place limitations on the
licensee’s field of use, commonly in the form of licence annexes which define the types of service that may be
provided or networks or technology that may be deployed.
68
total revenue, though ideally this should be specified as a cap with the regula-
tory authority adjusting the annual fee according to the costs of regulating the
sector.44 In countries with a multi-sectoral regulatory authority, regulatory costs
might need to be allocated across regulated industries. One option is to allocate
the regulatory budget between participants in all regulated industries on the basis
of their respective revenues. This approach has the advantage of simplicity and
transparency. However, it can also create imbalances, since both profit margins
and the cost of regulation might vary significantly between industries. An alter-
native approach is to separate the regulatory cost attributable to each industry,
and set licence fees for each industry accordingly. This process can become un-
necessarily complex where the majority of regulatory costs are common across in-
dustries. Where a multi-sectoral regulator is responsible for information services,
data protection, and e-commerce, these areas of regulation might in practice be fi-
nanced by the licence fees generated in the telecommunications sector because of
the difficulties in identifying and taxing the other regulated activities separately.
44
The level of and method for calculating annual fees should in this case be set out in a regulation or order
of general application.
896
45
This approach was adopted in Bahrain, where licensed operators with significant power are subject to
tariff controls in relation to any telecommunications service for which the regulatory authority determines
that insufficient competition exists.
807
these factors can materially distort retail tariff regimes. This is one reason why
many developing countries resort to international benchmarks to set some re-
tail price parameters.
The incumbent’s tariffs are typically highly unbalanced prior to undertaking
sector reforms: that is, charges for line rental and local calls are often historically
set well below cost, with charges for national and international calls set well above
cost. Incumbents have often retained monopolies over international gateway
services, which has allowed them to use international services to cross-subsidize
losses made in other services. Yet rises in rental or local call charges dispropor-
tionately affect low-income or low-volume users, and therefore unsurprisingly
tariff control is usually very politically sensitive. In practice, regulators have often
found it difficult to raise and rebalance the incumbent’s tariffs, even where they
have the statutory power to do so. At the same time, innovations such as VoIP have
(lawfully or otherwise) reduced incumbents’ market shares in their more profit-
able markets and undermined the benefits of cross-subsidization. Accordingly,
governments and regulators have been forced to reconsider their policies on man-
datory tariff rebalancing.
17.3.7 Interconnection
New operators will not be able to enter the market unless their subscribers are able
to call subscribers on other existing networks. Interconnection is also required for
the provision of indirect access, which allows customers to receive telecommu-
nications services, most commonly long-d istance and international calls, from a
provider other than the operator that provides the access line. The terms and con-
ditions of interconnection (technical, commercial, and legal) have an important
role in promoting competition in the sector.
The mechanisms for regulating interconnection are established through four
key regulatory elements—the law governing the sector, subsidiary regulations,
operator licences, and interconnection agreements. The law typically sets out the
basic obligations to interconnect, the ‘two-t ier’ regulatory structure (if used) that
distinguishes dominant from non-dominant operators, the powers of the regu-
lator with regard to interconnection, and the basic principles for setting inter-
connection charges. Licences might contain similar provisions but often include
greater detail. The regulator might also issue detailed interconnection regulations
that apply to all operators, typically with more onerous provisions attaching to
those that are dominant. Regulations tend to be a better tool than licences for this
purpose because of the need to ensure that all operators are subject to the same
regulatory controls, and to allow regulatory structures to be updated as markets
evolve.
871
46
ITU World Telecommunication/ ICT Indicators Database, <http://w ww.itu.int/en/I TU-D/Statistics/
Documents/statistics/2017/Mobile_cellular_ 2000-2016.xls>.
47
Oman Mobile, Reference Interconnection Offer, <https://www.omantel.om/wps/wcm/connect/0cfe943a-
fda0- 4 ccb- b ed1- 9 1740d124743/ 1 .+Main+Agreement_ 2 0160614.pdf?MOD=AJPERES&CONVERT_ T O=
URL&CACHEID=0cfe943a-fda0-4ccb-bed1-91740d124743>.
48
National Information and Communication Technology Authority of Papua New Guinea, ‘A public con-
sultation document on a draft rule specifying the acceptable form for reference interconnection offers’, 15
November 2011.
782
are approximately symmetrical (and where the applicable termination rates are
at broadly comparable levels), the network operators might decide to waive the
charges on the theory that net payments would be too insignificant to justify the
administrative cost of an interconnection payment system.49 Where no intercon-
nection payments are paid, this is known as a ‘bill and keep’ system. Although
the EU framework is still based on the CPP model,50 ‘bill and keep’ systems are
increasingly popular elsewhere. For example, the FCC adopted bill and keep as
the national interconnection framework 51 arguing that this model imposes fewer
regulatory burdens and eliminates a carrier’s ability to shift network costs to com-
petitors and their customers.52 However, bill and keep might not always be the
optimal approach to pricing IP interconnection given, eg, higher network costs
particularly where quality of service is guaranteed.53
Developing countries have largely adopted the European CPP model. In these
countries, where markets are developing rapidly, it is rare for traffic flows to be suf-
ficiently stable and symmetrical to justify a bill and keep system, and therefore, in
practice, systems of interconnection payment are generally in operation.
As noted above, where networks charge for the provision of interconnection
services, it is generally accepted that the applicable tariffs should be based on cost.
This ensures that the provider of interconnection receives some compensation for
the costs that it incurs and is given an incentive to invest in interconnection cap-
acity. A number of different ways of calculating the cost of interconnection have
emerged, the two most common methods being fully allocated costs and Long-
Run Incremental Cost (LRIC). While LRIC has emerged as the global regulatory
standard, its calculation can be very complex (often requiring significant network
engineering, financial, and management accounting information) and, conse-
quently, regulators in developing countries may instead rely on international
49
It is also worth bearing in mind the different termination rates applicable on fixed and mobile networks
(largely the result of different cost bases). Symmetrical traffic between fixed and mobile networks might still
result in net payments (likely to the mobile operators) and ‘bill and keep’ would therefore be unlikely to be
suitable.
50
This is not the result of an explicit policy preference, but because European termination rates (which are
required to be cost-based) still exceed the transaction costs of the payment system. While operators are free to
enter into ‘bill and keep’ agreements commercially, ‘bill and keep’ cannot be imposed by regulation because
the regulatory framework requires regulators to ensure cost recovery when setting regulated prices.
51
FCC Press Release, ‘FCC Releases Connect America Fund Order’, 18 November 2011, <http://w ww.fcc.
gov/document/press-release-fcc-releases-connect-a merica-f und-order>.
52
FCC, ‘Order and Further Notice of Proposed Rulemaking’, 27 October 2011 [738], <http://h raunfoss.fcc.
gov/edocs_public/attachmatch/FCC-11-161A1.pdf>.
53
See GSMA publication: Economic Study on IP Interworking: White Paper Prepared for GSMA by CRA
International and Gilbert + Tobin, February 2007, <http://w ww.itu.int/I TU-D/t reg/Events/Seminars/
GSR/G SR07/ D ocuments_ p resentations/ I P%20Interconnection%20-% 20GSMA%20White%20paper%20-
%20FINAL.pdf>.
783
54
eg in Botswana, the regulator (the BTA) issued two determinations on interconnection disputes in 2003
that covered the setting of termination charges. Although the BTA expressly recognized LRIC as the best
methodology for setting interconnection charges, benchmarking was used due to the lack of available cost
data. See ‘Ruling on Interconnection Charges Dispute between Botswana Telecommunications Corporation
and Mascom Wireless (Pty) Limited’, BTA Ruling No 1 of 2003, 26 February 2003.
847
a high level, carried out according to standards set by the ITU.55 In practice, spec-
trum grants have tended to be ad hoc, with first-comers often granted large blocks
of spectrum in an attempt to maximize interest and licence value. Today, many
developing countries are looking to deploy broadband access, typically using
wireless technologies. If prior spectrum grants were inefficient or over-extensive,
regulators might find that new broadband wireless technologies require the re-
location of existing spectrum users to other bands. Additionally, balancing the use
of licensed and unlicensed spectrum remains a challenge everywhere, increasing
the importance of monitoring the use of spectrum. Typically, spectrum plans will
need regular updating, and in some cases a (often politically contentious) spec-
trum re-farming exercise will be essential to efficiently manage allocations and
use of spectrum.
There are three main models of spectrum regulation: (1) a ‘command and control’
model, where the government or licensing agency determines the strict operating
parameters and rules defining spectrum rights; (2) a ‘market/usage rights’ model,
where the licensee is given more flexible rights to use specified spectrum within a
defined area during a fixed period of time (with spectrum use rules largely limited
to technical parameters in order to manage interference issues), and often the
ability to transfer those rights to other spectrum users (through trades or leases,
for instance); and (3) a ‘commons’ model, where spectrum is unlicensed and users
share a frequency block subject only to limited technical requirements, eg with
respect to power emissions (a well-k nown example of the latter being the use of
WiFi technologies in WLANs).56 The ‘command and control’ model is the model
most commonly used for spectrum blocks around the world, particularly given
its usefulness to meet military, emergency services, radio astronomy, and other
public needs. However, it creates an administrative burden as operators need to
acquire specific radio frequency licences for specific uses governed by bureaucrat-
ically defined spectrum allocations, thereby discouraging innovation and techno-
logical evolution. A ‘market/usage rights’ model provides more flexibility, but can
encourage spectrum ‘hoarding’ as operators might seek to acquire large blocks of
spectrum for unspecified uses in order to deny spectrum opportunities to com-
petitors, or profit from a subsequent trade. A ‘commons’ model has the benefit of
fostering innovation, but is not appropriate for most spectrum uses as it does not
enable sufficient management of over-crowding and interference, particularly
55
However, specific allocations between different uses and spectrum coordination with neighbouring
countries will need to be undertaken at both a regional coordination level and at a national level. In practice,
spectrum usage allocation might be negotiated informally between the relevant responsible ministries; alter-
natively, a formal inter-m inisterial body might be established for this purpose.
56
ITU, ‘Trends in Telecommunication Reform 2006: Regulating in the Broadband World’, 2006.
867
where the spectrum use is not highly localized. In practice, spectrum manage-
ment is a hybrid of these three approaches, providing flexibility to allow the rapid
deployment of new radio technologies, while safeguarding the interests of other
spectrum users and minimizing the regulatory burden on spectrum regulators.
Licences to use frequencies might be separate from or incorporated in network
operating licences and authorizations. Rights to use frequencies that are an essen-
tial adjunct to the operating licence (such as the uplink and downlink frequencies
for satellite and mobile networks) should ideally be incorporated in the oper-
ational licence.
Frequency fees should encourage efficient frequency use, to reflect the fact that
radio spectrum is a valuable, limited public resource. Thus, fees should be high
enough to prevent operators from ‘hoarding’ unused spectrum, but low enough
so as not to discourage smaller operators from entering the market or using wire-
less transmission media. There are various market-based mechanisms, including
spectrum auctions, for ensuring that spectrum fees are appropriately set, and the
approach taken should reflect the availability and characteristics of the particular
frequency range, current and likely future demand, and the need to promote
efficient use.
Given the scarcity of the resource and increased spectrum demands (particu-
larly to support mobile applications) there are recent trends towards allocation
initiatives that seek to maximize the use of unlicensed spectrum for spectrum re-
served for specific use, eg, in the case of ITU allocated ISM (Industrial, Scientific,
Medical) bands. Use of these bands differs internationally, but, regardless of the
legacy model of spectrum allocation, many countries have recognized the bene-
fits of permitting unregulated use (or class licensed use) of low-powered devices,
including wireless LANs, Bluetooth, and cordless phones in the 2.4 GHz bands.
The idea is to allow many services to coexist (increasing the efficient use of the
scarce resource) and increasing opportunities for experimentation and innov-
ation (by reducing financial and administrative barriers to entry).
One current spectrum management issue that is not confined to developing
countries is the realization of the ‘digital dividend’, ie the spectrum freed up upon
the conversion from analogue to digital (largely broadcasting) transmission. In
developing countries, the digital dividend might be used for mobile broadband,
a cost-effective means of improving broadband availability particularly where
fixed infrastructure might be lacking. In markets which already have high mo-
bile penetration, allocating the digital dividend spectrum for the deployment of
4G networks has the potential to lower the cost of provisioning additional cap-
acity on existing mobile networks. The growth in demand for mobile broadband
services and associated spectrum requirements mean that governments can po-
tentially receive significant financial gains, especially if done through spectrum
78
auctions (as was recently the case in India), from repurposing broadcast transmis-
sion spectrum.
the absence of retail price regulation) could result in high prices and be politically
contentious.
In recent times, universal access and service initiatives that were originally
conceived for fixed telecommunications, have had to become increasingly sophis-
ticated, firstly, to accommodate multiple and regionally based universal service
providers (often selected via auction-based systems to select providers) and sec-
ondly to accommodate a changing technology landscape that involves conver-
gence and broadband access.
17.3.13 Property rights
New entrants in developing countries will require property rights to both public
and private land in order to construct, maintain, and operate their networks, such
as rights of way, compulsory purchase powers, and rights to cut trees, fly lines, or
erect network infrastructure. Fast-track zoning procedures might also be desirable to
allow rapid network roll-out. The extent to which an operator will need such property
rights will depend on its existing rights, its network roll-out plans and the technology
deployed. In addition, the ease with which an operator can access public and pri-
vate land and install infrastructure will have a material impact on whether it decides
to roll out duplicating network infrastructure (which might be both economically
inefficient and environmentally disruptive) or simply to lease the necessary infra-
structure from existing market players. In some cases it might also be necessary to
coordinate the property rights exercised by different utilities, particularly in respect
of network construction, so as to minimize disruption. The design of appropriate
property rights will depend heavily on the property law system and land ownership
structures in the relevant country. Often, regulatory objectives will dictate the nature
of property rights, in particular whether the regime is intended to foster facilities-
based competition or facility sharing among operators.
57
For a more thorough discussion of innovations in telecommunication dispute resolution, on which this
section is based, see eg Dispute Resolution in the Telecommunications Sector: Current Practices and Future
Directions, World Bank/I TU, 2005, currently available at: <http://w ww.itu.int/I TU-D/t reg/publications/I TU_
WB_Dispute_Res-E .pdf>.
80
markets through the incentive structures it creates. Where the incentives are for
operators and service providers to seek resolution of disputes, rather than seeking
to create disputes or prolong them, stakeholders in the sector should benefit from
resulting efficiencies. In particular, sector development will benefit from a regu-
latory environment that encourages the prevention, early identification, and reso-
lution of disputes. Earlier editions of this chapter went into greater detail on the
causes of and responses to disputes in the sector. Those basic dispute resolution
policy principles continue to apply, even in a broadband-dominated world. The
years immediately following the first wave of post-liberalization privatizations
and new entrant mobile licensing saw corresponding waves of disputes involving
the nature and duration of exclusive rights granted to incumbents, the licensing of
new entrants, tariff arrangements, interconnection arrangements, and spectrum
matters.58 As a response, telecommunications laws increasingly included sector-
specific mechanisms (so-called ‘alternative dispute resolution’ techniques such as
arbitration) to deal with those disputes. The attempt to ensure some degree of cer-
tainty about the resolution of those disputes finds new purchase in the increasing
complexity of the market, as both the number of service providers and the range of
services (and technologies used to provide them) grow and diversify. For instance,
we are seeing growing tensions in the market between infrastructure and service
providers on the one hand and so-called ‘over-the-top’ providers on the other,
which are playing out in regulatory debates over customer access to OTT content
(in the context of ‘net-neutrality’).
Investing and operating in the developing world typically carry a higher degree
of political, economic, and security risk, which might not be adequately addressed
(and indeed might be exacerbated) by local avenues for dispute resolution. For
example, if there is a change in government policy, such as the abolition of ex-
isting rights or a nationalization of core telecommunications infrastructure, the
affected foreign investors might find local courts or regulatory bodies unsym-
pathetic. Accordingly foreign investors might seek to protect their investments
in high-r isk markets by requiring disputes relating to their investments to be re-
solved through international arbitration in a neutral venue (eg through the World
Bank’s International Centre for Settlement of Investment Disputes (ICSID),59
the International Chamber of Commerce,60 or according to the United Nations
Commission on International Trade Law).61 These solutions can have an adverse
effect on the local regulatory framework, by hampering the development of local
jurisprudence or creating inconsistent regulatory outcomes that do not reflect
58
Ibid. 59
For further information cf <https://icsid.worldbank.org/en/>.
60
For further information cf <http://w ww.iccwbo.org/policy/a rbitration/id2882/i ndex.html>.
61
For further information re cf <http://w ww.uncitral.org/u ncitral/en/about_u s.html>.
81
62
Currently there are more than 1480 Bilateral Investment Treaties between developing and developed
countries. See <http://i nvestmentpolicyhub.unctad.org/I IA>.
63
See Bronckers, M and Larouche, P, ‘A Review of the WTO Regime for Telecommunications Services’, in
The World Trade Organization and Trade in Services (Leiden: Martinus Nijhoff, 2008).
82
clear that a country that had submitted its schedule of commitments was not ex-
cused from performance of those commitments merely because it had not yet
adopted the necessary implementing regulations. Following the decision, Mexico
agreed to promulgate the necessary regulations to give effect to its commitments.
64
See, World Bank, Montenegro— Policy note on broadband: achieving universality of high- speed
broadband—review and application experience of the EU State aid framework. (Washington, DC: World Bank
Group, 2017), <http://documents.worldbank.org/c urated/en/556361495708776351/Montenegro-Policy-note-
on-broadband-achieving-u niversality- of-h igh-s peed-broadband-r eview-a nd-application- e xperience- of-
the-E U-State-a id-f ramework>.
83
65
‘Sub-
Saharan Africa: Broadband, Regulation and International Bandwidth Pricing Drive National
Backbone Roll-Out’, Global Insight, 29 January 2008.
84
networks were in place.66 Contracts for such networks have been commissioned
by numerous governments in sub-Saharan Africa. Another approach is the one
taken by the government of South Africa, which in 2006 combined the telecommu-
nications assets of electricity provider Eskom and transport parastatal Transnet
to create a new broadband infrastructure company, Infraco, to provide national
backbone transmission capacity on a cost-plus basis.67
These government-backed networks pose interesting policy questions in pri-
vatized and liberalized markets. In some ways, they turn the clock back ten years,
with governments again investing in network deployment where previously they
had divested their interests in the telecommunications sector. One question
that arises is whether governments will be able to utilize these networks effect-
ively and fairly without distorting the market. Operators could, for example, find
themselves under increasing political pressure to purchase backbone transmis-
sion capacity from the new government-owned backbone networks in order to
provide financial support to the state-owned entity. On the other hand, where
government-owned backbone networks exist and are offering adequate quality
of service at the right price, operators could be willing to use these networks to
increase the capacity of their own networks. Access to such backbone services
could significantly increase operators’ ability to roll out 4G (or even 5G) and other
high-speed technologies more quickly and more widely. However, to secure real
benefits from the substantial government investments in fibre-optic backbone
infrastructure, these initiatives should be accompanied by complementary steps
ensuring ‘open access’, including liberalizing the market for services (if not already
liberalized), ensuring access to networks (through, for example, making available
spectrum and licences to provide broadband wireless access), and ensuring ac-
cess to international connectivity. Such interventions might come at a cost to the
public purse given the cost of financing network construction. Government in-
volvement in rolling out these networks might also have implications on public
procurement rules.
Increasing government involvement and the potential for new national monop-
olies also have implications for regulatory reform and design. In addition, growing
penetration and reliance on broadband networks raise new regulatory challenges
(including around cybercrime, data security, privacy, access to information,
freedom of information, and online content regulation), which were not generally
included in the wave of first-generation reforms.
66
<http://w ww.macrolan.co.za/blog/fibre-optic-networks-reach-4 4-of-t he-a frican-population/>.
67
‘South Africa: Infraco Plans New Submarine Cable’, Global Insight, 3 August 2007.
85
at the time by Portugal Telecom) in East Timor, was renegotiated and converted
to a licence in 2012.68
BOT-t ype arrangements were previously thought to encourage investment by re-
ducing investor risk: the investor’s ‘exit strategy’ being guaranteed at the date of ex-
piry or termination of the BOT. However, in practice, BOTs in the communications
sector can be much more problematic than BOTs in other infrastructure projects
(such as toll roads and power plants), and the investor’s exit can be exceedingly com-
plex, acrimonious, and protracted. This is in part because, at the end of the BOT con-
tract, it is not just the physical network infrastructure that needs to vest in the state.
Without the complex back-office systems and software to manage and monitor the
network, significant service disruptions can occur, and therefore comprehensive
transitional arrangements might need to be put in place with the exiting operator
(which might not have been adequately provided for in the original BOT contract).
Less common are ‘build-t ransfer-operate’ (BTO) and ‘build-operate-own’ (BOO)
contracts. Under BTO contracts the investors build the network and transfer title
to the state, but continue to operate the network and share revenues from its oper-
ation with the state. This allows the state to have ownership and control over the
network from the start, typically reflecting a reluctance of the state to relinquish
control of key infrastructure, but is detrimental to the investors, who bear most of
the costs without the benefit of ownership. A BOO is, essentially, the contractual
equivalent of a licence, in that it simply authorizes an entity to build and operate
its own network. However, unlike many licences,69 it typically includes a revenue-
sharing obligation similar to those found in BOTs and BTOs.
In recent times, significant government-sponsored fibre initiatives materially
changed the investment landscape and PPPs became more widespread. Care is
needed to ensure that the resurgence of concession-t ype arrangements through PPPs,
driven by broadband roll-out imperatives, can co-exist with the pro-competitive
licensing regimes established in most countries over the last two decades. For ex-
ample, it is important to ensure that a PPP does not result in the creation of new mon-
opolies in relation to infrastructure or the provision of broadband services.
As a new generation of state- owned or partially state- owned enterprises
emerges in the broadband space, it becomes essential to ensure that proper sector
and competition regulation is applied to these state-owned market segments, that
all PPPs are properly licensed and have entered into interconnection agreements,
68
Bray, J, ‘International Companies and Post- Conflict Reconstruction Cross- Sectoral Comparisons’,
at 19, <http://w w w-w ds.worldbank.org/s ervlet/ W DSContentServer/I W3P/I B/2 005/0 3/3 0/0 00012009_
20050330161732/Rendered/I NDEX/31819.txt>.
69
Although licences may also contain revenue sharing obligations: see eg the 20% revenue share obligation
in the mobile licence won by a consortium led by Turkcell in Iran. See ‘Turkcell Enters Iranian Mobile Market’,
Telecom Worldwide, 19 February 2001.
87
and that PPPs are subject to the same regulatory controls as other market partici-
pants. Additional questions about on-going investment in the underlying infra-
structure, including the decision making and financing of upgrades, will have to
be addressed in these PPP structures.
advertising revenues that come with the content (and, by extension, in the cus-
tomer data that enables the targeting of that advertising). Online content stand-
ards, questions around freedom of information and expression (including ‘net
neutrality’), and data protection have therefore appeared on the agendas of tele-
communications regulators as part of a ‘second wave’ of regulatory reforms. As
broadband has enabled commercial transactions and government services to
move online, network and information security have also become increasingly
important, often necessitating new regulatory reforms to address issues such
as network and cyber-security. In developing countries, despite the fact that
certain of these changes are outside what might typically be considered tele-
communications regulation, the responsibility to address these non-telecom
specific areas of regulation of modern communications often falls to the tele-
communications regulator.
70
See, World Development Report 2016: Digital Dividends, (Washington, DC: World Bank, 2016), at
224 et seq., at <http://documents.worldbank.org/c urated/en/961621467994698644/pdf/102724-W DR-
WDR2016Overview-E NGLISH-WebResBox-394840B-OUO-9.pdf> (‘WDR’).
71
See, ‘Combatting Cybercrime: Tools and Capacity Building for Developing economies’: <http://w ww.
combattingcybercrime.org> (Combatting Cybercrime).
72
Case C-131/12, Google Spain v Agencia de Protección de Datos, 2014 EUR-L ex 13 May 2014. European Court
of Justice. For a discussion of the case, see, Kelly and Satola, ‘The Right to be Forgotten’, (2017) University of
Illinois Law Review 1, at 1.
73
Case C-362/14, Maximilian Schrems v Data Protection Commissioner 6 October 2015, ECJ.
74
Regulation (EU) 2016/679 of the European Parliament and of the Council, <http://eur-lex.europa.eu/
legal-content/E N/T XT/PDF/?uri=CELEX:32016R0679&from=EN> (GDPR).
890
75
Combatting Cybercrime, n 71, at 184.
76
See World Bank, World Development Report 2016: Digital Dividends, (Washington, DC: World Bank,
2016), at 221 et seq., at <http://documents.worldbank.org/curated/en/961621467994698644/pdf/102724-WDR-
WDR2016Overview-ENGLISH-WebResBox-394840B-OUO-9.pdf> (‘WDR’).
77
WDR, at 227.
819
infrastructure and the communications (ie the data) flowing over them. Whether
these issues are undertaken by telecommunications regulators or not, today’s net-
works and services demand a more holistic, comprehensive, and coherent regula-
tory approach across a variety of disciplines.
Developing countries can become, even if unwittingly, ‘safe-havens’ for cyber-
criminals due to outdated regulatory structures or a lack of enforcement cap-
ability. Governments will, accordingly, need to examine their legal frameworks to
ensure an adequate regulatory regime is in place (including in relation to digital
authentication, electronic transactions, information security, critical infrastruc-
ture protection and data and privacy protection), enabling them to remain good
‘citizens’ in the global community. Increasingly, elements of cybercrime legisla-
tion (eg penalties for unauthorized access to networks or data, or interference with
networks, data or communications) will need to be included in telecommunica-
tions or electronic transactions (e-commerce) laws or as a package of legislation
coordinated with telecommunications legal reforms.78
78
For an example of how cybercrime elements are included in e-t ransactions laws, see eg Arts 34 et seq.,
of the Electronic Transactions Law (No 5 of 2004) of Union of Myanmar; and the Kingdom of Tonga’s reforms
included updating its telecommunications law together with a new cybercrime law.
829
Reform of the telecommunications sector can lay the foundations for the reform of
other sectors of the economy, attracting new (foreign and domestic) investment,
839
Index
896 Index
Index 897
898 Index
Index 899
900 Index
Index 901
902 Index
Index 903
904 Index
Index 905
906 Index
Index 907
908 Index
Index 909
910 Index
Index 911
912 Index
Index 913
914 Index
Index 915
916 Index
Index 917
918 Index
Index 919
920 Index