Policy Paradigm-Theory and Pratices

Studies in the Political Economy of Public Policy
The series Studies in the Political Economy of Public Policy presents cutting edge,
innovative research on the origins and impacts of public policy. Going beyond
mainstream public policy debates, the series encourages heterodox and heteroge-
neous studies of sites of contestation, conflict and cooperation that explore pol-
icy processes and their consequences at the local, national, regional or global
levels. Fundamentally pluralist in nature, the series is designed to provide high
quality original research of both a theoretical and an empirical nature that sup-
ports a global network of scholars exploring the implications of policy on
society.
The series is supported by a diverse international advisory board drawn from
Asia, Europe, Australia and North America, and welcomes manuscript submis-
sions from scholars in both the global South and North that pioneer new under-
standings of public policy.
Series editors
Toby Carroll, Department of Asian and International Studies, City University of
Hong Kong
Darryl Jarvis, Department of Asian and Policy Studies, Hong Kong Institute of
Education
Paul Cammack, Department of Asian and International Studies, City University
of Hong Kong
M Ramesh, Lee Kuan Yew School of Public Policy, National University of
Singapore
International Advisory Board

Michael Howlett, Simon Fraser University, Canada
John Hobson, University of Sheffield, UK
Stuart Shields, University of Manchester, UK
Lee Jones, Queen Mary, University of London, UK
Kanishka Jayasuriya, University of Adelaide, Australia
Shaun Breslin, University of Warwick, UK
Kevin Hewison, Murdoch University, Australia
Richard Stubbs, McMaster University, Canada
Dick Bryan, University of Sydney, Australia
Kun-chin Lin, University of Cambridge, UK
Apiwat Ratanawaraha, Chulalongkorn University, Thailand
Wil Hout, Institute of Social Studies, Erasmus University, The Netherlands
Penny Griffin, University of New South Wales, Australia
Philippe Zittoun, Science Po, Grenoble, France
Heng Yee Kuang, National University of Singapore
Heloise Weber, University of Queensland, Australia
Max Lane, Victoria University, Australia
Titles include:
Toby Carroll and Darryl S. L. Jarvis (editors)
THE POLITICS OF MARKETISING ASIA
Pascale Hatcher
REGIMES OF RISK
The World Bank and the Transformation of Mining in Asia
Daniel Novotny and Clara Portela (editors)
EU-ASEAN RELATIONS IN THE 21ST CENTURY
Towards a Stronger Partnership
Philippe Zittoun
POLICY AS POLITICS
Discursive Transformations and Public Policymaking
Philip Mader
THE POLITICAL ECONOMY OF MICROFINANCE
Financialising Poverty
Policy Paradigms in Theory
and Practice
Discourses, Ideas and Anomalies
in Public Policy Dynamics
Edited by
John Hogan
Lecturer, Dublin Institute of Technology, Ireland
Michael Howlett
Professor, Simon Fraser University, Canada
Palgrave
macmillan
Introduction, editorial matter and selection © John Hogan and
Michael Howlett 2015
Individual chapters © Respective authors 2015
Softcover reprint of the hardcover 1st edition 2015 978-1-137-43403-6
All rights reserved. No reproduction, copy or transmission of this

publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted

save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6-10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified as the authors of this
work in accordance with the Copyright, Designs and Patents Act 1988.
First published 2015 by

PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,

registered in England, company number 785998, of Houndmills,
Basingstoke, Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies

and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,

the United Kingdom, Europe and other countries.
ISBN 978-1-349-56900-7 ISBN 978-1-137-43404-3 (eBook)

DOI 10.1057/9781137434043
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
Contents
List of Tables and Figures vii
Acknowledgments ix
Notes on Contributors x
List of Abbreviations xvi
Part I 1
1 Reflections on Our Understanding of Policy Paradigms
and Policy Change 3
John Hogan and Michael Howlett
2 What Is a Policy Paradigm? Overcoming Epistemological

Hurdles in Cross-Disciplinary Conceptual Adaptation 19
Matt Wilder
3 Can You Recognize a Paradigm When You See One?

Defining and Measuring Paradigm Shift 43
Pierre-Marc Daigneault
4 Is There a Fourth Institutionalism? Ideas, Institutions

and the Explanation of Policy Change 61
Jeremy Rayner
Part II 81
5 Comparing and Contrasting Peter Hall’s Paradigms and
Ideas with the Advocacy Coalition Framework 83
Paul Cairney and Christopher M. Weible
6 Paradigm Construction and the Politics of Policy Anomalies 101

Matt Wilder and Michael Howlett
7 From Policy Paradigm to Policy Statement: A New Way to

Grasp the Role of Knowledge in the Policymaking Process 117
Philippe Zittoun
vi Contents
8 Paradigms and Unintended Consequences: New Public

Management Reform and Emergency Planning in Swedish
Local Government 141
Daniel Nohrstedt
Part III 165

9 The Critical Role of Ideas: Understanding Industrial
Policy Changes in Ireland in the 1980s 167
John Hogan and Brendan K. O’Rourke
10 The Bologna Process and the European Qualifications

Framework: A Routines Approach to Understanding the
Emergence of Educational Policy Harmonisation –
From Abstract Ideas to Policy Implementation 189
Sharon Feeney and Conor Horan
11 The Role of Ideas in Evaluating and Addressing Hydraulic

Fracturing Regulations 217
Christopher M. Weible, Tanya Heikkila, and Jonathan J. Pierce
12 Communications Frameworks and the Supply of

Information in Policy Subsystems 239
Samuel Workman and JoBeth S. Shafran
13 How and Why Do Policy Paradigms Change; and

Does It Matter? The Case of UK Energy Policy 269
Florian Kern, Caroline Kuzemko, and Catherine Mitchell
Conclusion 293
14 Bringing Ideational Power into the Paradigm Approach:
Critical Perspectives on Policy Paradigms in Theory and
Practice 295
Martin B. Carstensen
Index 319
List of Tables and Figures
Tables
8.1 Overview of unintended consequences,

operationalization, survey questions, and years covered 152
9.1 Ireland’s main economic indicators, 1974–1989 172
9.2 Identification of macroeconomic crisis in Ireland

in the 1980s 174
9.3 Indication of ideational collapse 176
9.4 Indication of (1) new ideational consolidation and

(2) level of policy change 180
10.1 The objectives of the Bologna Declaration, 1999 200
10.2 Artifacts that relate to Objective #1 of the Bologna

Declaration: A system of easily readable and
comparable degrees 201
10.3 Qualifications descriptors (known as the

‘Dublin Descriptors’) 204
11.1 Explaining policy actor evaluations of problem

resolution by the disclosure rule and setback rule 226
11.2 Explaining variance in changes in positions on

government regulations related to disclosure
and setbacks 227
11.3 Explaining changes in support for government

regulations on hydraulic fracturing issues 229
11A.1 Means and factor loadings for disclosure and

setback variables and scales 231
11A.2 Measures, means, and factor loadings for changes

in support for government regulations 232
11A.3 Measures, means, and factor loadings for

independent variables 232
vii
viii List of Tables and Figures
12.1 Witnesses at congressional hearings on energy policy,

1995–2010 260
12.2 Bureaucratic composition of witnesses in energy policy,

1995–2010 261
13.1 The pro-market energy policy paradigm in 2000 275
13.2 The energy policy paradigm in 2000 and 2011 283
Figures
2.1 Six-part disaggregation of policy 29
2.2 Normative and technical elements of policy in a

causal map 31
6.1 Oliver and Pemberton’s evolutionary iterative

framework 107
6.2 Contingent strategies and frame extension 109
6.3 Two processes of problem definition and

solution construction 111
8.1 Summary of survey responses indicating unintended

consequences 154
10.1 The ostensive–performative theory of routines 194
10.2 Convergence from policy formation to policy

implementation 209
Acknowledgments
This volume finds its earliest origins in a workshop at the ECPR in

Antwerp, in April 2012, organised by Michael Howeltt and Klaus Goetz
and at panel at APSA in Chicago, in August 2013, organised by John
Hogan. We would like to express our thanks to the series editors Toby
Carroll, M. Ramesh, Darryl Jarvis and Paul Cammack. At Palgrave we
would like to express our deepest appreciation to Ambra Finotello,
Sundardevadoss Dharmendra, Christina M. Brian and all of the edito-
rial team for guiding and supporting us through this process. We would
like to thank the anonymous reviews for all of their advice, guidance
and support. Lastly, but not least, we would like to express our deepest
appreciation to all of the contributors for their time, efforts, patience
and contributions, they were each, and every one, a joy to work with.
ix
Notes on Contributors
Paul Cairney is Professor of Politics and Public Policy, University of

Stirling. His research interests are in comparative public policy, includ-
ing comparisons of policy theories (Understanding Public Policy, 2012)
and methods (Handbook of Complexity and Public Policy, co-edited with
Robert Geyer), policy outcomes in different countries (Global Tobacco
Control, 2012, with Donley Studlar and HadiiMamudu), Scottish politics
(The Scottish Political System since Devolution, 2011, and Scottish Politics
2nd ed., 2013, with Neil McGarvey), comparisons of UK and devolved
policymaking (‘Has Devolution Changed the British Policy Style?’,
British Politics, 3, 3, 350–72), and comparisons of policy outcomes across
the UK (‘Policy Convergence, Transfer and Learning in the UK under
Devolution’, Regional and Federal Studies, 22, 3, 289–307, with Michael
Keating and Eve Hepburn). He is currently funded (October 2013–15) by
the Economic and Social Research Council to research the policymaking
process in Scotland, focusing on areas such as preventative spending.
Martin B. Carstensen is Assistant Professor in the Department of

Business and Politics at Copenhagen Business School. His primary
theoretical interests lie within institutional theory and discursive insti-
tutionalism. More specifically he studies how ideas develop over time –
both gradually and through large ruptures – and how such dynamics
of change are related to various forms of power, including ideational
power. Empirically, Carstensen works on the financial crisis of 2007–
2009, analysing how decision-makers in smaller open economies in the
Western world handled the uncertainty brought on by the crisis. In July
2012 he began a project on ‘Small States in the Perfect Storm’ (funded
by Carlsbergfondet).
Pierre-Marc Daigneault is Assistant Professor of Public Policy and Public

Administration in the Department of Political Science at Université
Laval. His academic interests are in policy theories, program evalua-
tion, research methods, and social policy. Daigneault is currently study-
ing public programs aimed at “activating” long-term social assistance
(welfare) recipients. His research has been published in various peer-
reviewed journals including the American Journal of Evaluation, Canadian
Journal of Political Science, Evaluation Review, Journal of European Public
Policy, Journal of Mixed Methods Research, and Political Studies Review.
x
Notes on Contributors xi
He also co-edits, with Daniel Béland, Welfare Reform in Canada: Provincial

Social Assistance in Comparative Perspective (forthcoming in August 2015).
Sharon Feeney lectures in Business Policy and Corporate Strategy in the

College of Business at the Dublin Institute of Technology. Her research
interests focus on higher education policy at the institutional, national,
and international levels, with particular emphasis on quality systems,
program and institutional evaluation, and qualifications and awards
frameworks. She has consulted on and developed, reviewed, and imple-
mented policies at institutional level in higher education institutions in
Ireland, Russia, Oman, Egypt, and Bulgaria over the past 15 years. She
is the director of programs delivered to the Irish Air Corps and Simon
Community (a non-profit for homelessness). She has been an invited
speaker, as an expert on Quality Systems in Higher Education and
Qualifications Frameworks, at Conferences of University Deans/Rectors
in Nizhny Novgorod, Russia; in Sofia, Bulgaria; and in Beijing, China. In
addition, she has lectured in Leuven, Bologna, Cairo, Alexandria, and
Prague. She has published in numerous journals and edited volumes,
and has presented her work on higher education and policy develop-
ment at conferences around the world.
Tanya Heikkila is Associate Professor and Doctoral Program Director

at the School of Public Affairs at the University of Colorado Denver.
Heikkila’s research focuses on environmental policy and governance
and comparative institutional analysis. She has explored institutions
for coordinating and collaborating across boundaries and for resolving
environmental conflicts. This includes studies of interstate water con-
flicts and cooperation in the US, conflicts and cooperation related to
hydraulic fracturing, the organization and performance of large-scale
ecosystem restoration programs, and the institutions for coordinating
groundwater and surface water management. She holds an MPA and
PhD from the University of Arizona.
John Hogan lectures on Irish Politics and International Political Economy

at the Dublin Institute of Technology. His research interests focus on
developing frameworks for identifying and understanding policy change,
and studying the global regulation of the lobbying industry. He has pub-
lished in a wide range of journals including Policy Studies Journal, Acta
Politica, Politics, the Canadian Journal of Political Science, and The Political
Quarterly. He is co-editor of Irish Business and Society and Approaches to
Qualitative Research and is a co-author of Regulating Lobbying: A Global
Comparison. He has had chapters published in various edited volumes and
has presented his research at numerous conferences around the world.
xii Notes on Contributors
Conor Horan is Lecturer in Marketing at the Dublin Institute of

Technology and a PhD candidate at the University of Strathclyde Business
School, Glasgow. He is undertaking research into the generative nature
of routines and practices relating to knowledge creating, exchange,
and development between higher education institutions and compa-
nies across different industry sectors and contexts. His work has been
published in the Journal of Marketing Management, Industrial Marketing
Management, Journal of Business Research, and Journal of Strategic Marketing.
Michael Howlett is Burnaby Mountain Chair in the Department of

Political Science at Simon Fraser University and Yong Pung How Chair
Professor in the Lee Kuan Yew School of Public Policy at the National
University of Singapore. He specializes in public policy analysis, political
economy studies, and resource and environmental policy analysis. He is
both the author and co-author of numerous texts on Canadian public
policy, public policy in general, and political economy. Howlett is the
co-editor of various volumes on public policy and environmental policy.
His articles have been published in numerous professional journals in
Canada, the United States, Europe, Latin America, Asia, and Australia
and New Zealand. He has also edited a wide range of academic jour-
nals including the Canadian Journal of Political Science (2002–06), and
co-edited World Political Science Review (2003–14). He currently sits on
the editorial boards of numerous journals.
Florian Kern is Senior Lecturer at SPRU (Science Policy Research Unit)

at the University of Sussex and Co-Director of the Sussex Energy Group.
Kern’s research combines insights from policy studies and innovation
studies and has been published in journals like Environmental Politics,
Policy & Politics, Energy Policy, Technological Forecasting and Social Change,
and Environment and Planning C. His research interests include the inter-
play of institutional and ideational factors in shaping policy processes,
in particular in the field of energy, environmental and innovation pol-
icy, as well as the politics of governing transitions toward more sustain-
able societal systems.
Caroline Kuzemko is a research fellow working as part of an EPSRC-

funded project, ‘Innovation and Governance (IGov)’, based at the
University of Exeter. She has research interests in international politi-
cal economy, energy and climate governance and institutional change.
Kuzemko is a visiting fellow at the University of Warwick and the (co-)
convenor of two international, inter-disciplinary academic networks: the
PSA-sponsored ‘Anti-politics and Depoliticisation Special Group’ (APDSG)
and the ‘Political Economy of Energy in Europe and Russia’ (PEEER).
Notes on Contributors xiii
She has recently published The Energy Security-Climate Nexus: Institutional

Change in the UK and Beyond (2013, Palgrave Macmillan) and articles in
the Journal of European Public Policy and Policy and Politics (2014). Prior to
her academic career, Kuzemko worked for eight years at UBS investment
bank as Head of Emerging Market Equity Sales (ex-Asia) in London.
Catherine Mitchell is Professor of Energy Policy at the University of

Exeter. She currently holds an Established Career Fellowship with the
EPSRC (2012–2016) called Innovation and Governance for a Sustainable
Economy (http://projects.exeter.ac.uk/igov/). Mitchell has been part of
many international writing teams (e.g. a lead author in the IPCC’s Fifth
Assessment Report in the Policy and Institutions Chapter of WG3) and
research teams (e.g. PI of an ESRC/EPSRC interdisciplinary research clus-
ter into Energy Security in a Multi-Polar World [2008–2013]). She has
worked for governments (e.g. GB in 2001–2002) and has undertaken
numerous advisory positions (e.g. a member of the UK Government’s
Energy Advisory Panel [1998–2003]). She has also been a member of
many boards (e.g. the Board of the Regulatory Assistance Project, http://
www.raponline.org/).
Daniel Nohrstedt is Associate Professor of Political Science at Uppsala

University, Sweden, where he is also Director of Studies in the Center for
Natural Disaster Science (CNDS). His research interests include policy
processes (particularly the Advocacy Coalition Framework), crisis and
disaster management, governance, policy networks, and collaborative
public management. Nohrstedt’s current research centers on interor-
ganizational collaboration in crisis management including adaptation,
learning, and performance in crisis preparedness and response. His recent
articles have appeared in Policy Studies Journal, Public Administration,
Administration & Society, Risks, Hazards, & Crisis in Public Policy and Public
Management Review. Nohrstedt is a member of the editorial board in the
Policy Studies Journal, Risks, Hazards, & Crisis in Public Policy, and Springer
book series Environmental Hazards.
Brendan K. O’Rourke is College of Business Research Fellow at the Dublin

Institute of Technology (DIT), where he focuses on learning in the area
of discourses of the economy. His academic publications include articles
on elite formation, interview methodology, and on enterprise discourses.
Brendan has supervised a range of research on topic such as meetings’
discourse, economists on the radio, and creativity. Currently, he is par-
ticularly interested in discourses of enterprise, elitism, and neoliberalism.
O’Rourke is co-founder of the Discourse Analysis Group within DIT, and
the head of the DIT Business, Society and Sustainability Research Centre.
xiv Notes on Contributors
Jonathan J. Pierce is Assistant Professor at the Institute of Public

Service, Seattle University, where he teaches in the Master of Public
Administration and Bachelor of Public Affairs programs. He completed a
post-doctoral fellowship at the University of Colorado Denver as well as
teaching at the University of Denver. He received a PhD in Public Affairs
from the School of Public Affairs, University of Colorado Denver. His
current research examines the policy process and the politics of natural
resource development and sustainability. He teaches courses in policy
process, policy analysis, public and non-profit administration, research
methods, and environmental policy.
Jeremy Rayner holds a Centennial Research Chair at the Johnson-

Shoyama Graduate School of Public Policy, University of Saskatchewan.
He has a longstanding interest in the role of ideas in explanations of
stability and change, dating back to a dissertation on political metaphor.
More recently, his work has focused on natural resource and energy poli-
cies. As Chair of the Global Forest Expert Panel on the International
Forest Regime, he has promoted the use of international forest govern-
ance mechanisms to improve policy learning. In Canada, his research
has investigated the impact of public engagement on complex policy
mixes in the energy sector.
JoBeth S. Shafran is currently a doctoral candidate studying American

Politics and Public Policy in the Department of Government at the
University of Texas at Austin. Her research centers on the policy process,
subsystem policy making, the federal bureaucracy, and congressional-
bureaucratic interactions. Her current project looks at information shar-
ing in the policy process, who supplies information to Congress and the
effects of information on the problem definitions used in legislation.
She is also associated with the Policy Agendas Project at UT-Austin. Prior
to attending the University of Texas, Shafran completed a BA and an MA
in Political Science at West Virginia University.
Christopher M. Weible holds a PhD in Ecology from the University of

California Davis. He earned a Master’s in Public Administration and a
BSc in Mathematics and Statistics from the University of Washington.
Weible is co-editor of the Policy Studies Journal. He teaches courses in
environmental politics, policy processes, policy analysis, and research
methods and design. He studies politics and policy in contentious
environmental issues. Recent and current research include studying
multi-stakeholder collaboration processes in aquaculture partnerships,
assessing policy designs and improving outcomes in organic farming,
Notes on Contributors xv
and analyzing the politics of unconventional oil and gas development

using hydraulic fracturing (fracking).
Matt Wilder is a PhD candidate at the Department of Political Science,

University of Toronto, Canada. His dissertation, ‘Constraint and
Necessity: The Politics of Canadian Industrial Policy in the Age of
Globalization,’ explains how and why state intervention in the econ-
omy has persisted despite the emergence of institutions and global
norms that discourage it. This work accompanies his broader interest in
Canadian politics and government, Canadian and comparative politi-
cal economy, comparative public policy, and mixed-methods analysis.
Wilder’s research has been published in a range of international journals
and he has presented his findings at conferences across North America
and Europe.
Samuel Workman is Assistant Professor in the Department of Political

Science at the University of Oklahoma, where he is also a faculty asso-
ciate of the Center for Risk and Crisis Management and the Center
for Intelligence and National Security. His new book, The Dynamics of
Bureaucracy in the U.S. Government: How Congress and Federal Agencies
Process Information and Solve Problems (2015), focuses on the regulatory
process as generating information that shapes the congressional agenda
in a two-way flow of influence between the bureaucracy and Congress.
His current projects examine the influence of governmental policy anal-
ysis on agenda setting, the impact of regulatory uncertainty and frag-
mentation on business investment and employment, and the politics of
food security and food systems.
Philippe Zittoun is Research Professor of Political Science at LET-ENTPE,

University of Lyon, France, and Research Fellow at PACTE (Science
Po Grenoble). He is the General Secretary of the International Public
Policy Association (IPPA), Vice-Chair of the IPSA ‘Public Policy and
Administration’ Research Committee and coordinator of the interna-
tional Conference on Public Policy. He is associated editor of the Journal
of Comparative Policy Analysis and also member of the editorial board of
several public policy journals such as Critical Policy Studies, Policy Studies
Journal, Policy and Society, and the Journal of Policy Research. His research
focuses on understanding the policy process and, in particular, on the
importance of knowledge, definitional struggles, and argumentative
strategies of the actors and on political aspects. He is the author of The
Political Process of Policymaking (Palgrave, 2014).
List of Abbreviations
ACF advocacy coalitions framework

BETTA British Electricity Trading and Transmission Arrangements
BFUG Bologna Follow-Up Group
CCC Committee on Climate Change
CJT critical junctures theory
COGCC Colorado Oil and Gas Conservation Commission
CSO Central Statistics Office
DECC Department for Energy and Climate Change
DI discursive institutionalism
DTI Department of Trade and Industry
EHEA European Higher Education Area
EIA Energy Information Administration
ENIC European Network of Information Centres
ENQA European Network for Quality Assurance
EQF European Qualifications Framework
ESRI Economic and Social Research Institute
EU European Union
FDI foreign direct investment
FiT feed-in-tariff
GDP gross domestic product
GNP gross national product
HE higher education
HEA higher education area
HI historical institutionalism
IDA Industrial Development Authority
IMF International Monetary Fund
NESC National Economic and Social Council
NPM new public management
OECD Organization for Economic Cooperation and Development
Ofgem Office of the Gas and Electricity Markets
ORED Office for Renewable Energy Deployment
NARIC National Academic Recognition Information Centres
PE punctuated equilibrium
PEPP pro-market energy policy paradigm
PIU Performance and Innovation Unit
xvi
List of Abbreviations xvii
RI rational choice institutionalsim

RO renewable obligation
SI sociological institutionalism
UK United Kingdom
US United States
Part I
1
Reflections on Our Understanding
of Policy Paradigms and Policy
Change
John Hogan and Michael Howlett
Introduction
This volume seeks to contribute to the ongoing conversation among

policy scholars on the subject of policy paradigms. It provides a window
into the research frontier of policy dynamics and a re-evaluation of the
precision and utility of existing policy paradigm orthodoxy. A ‘policy
paradigm’ constitutes a theoretical tool to specify and understand the
guiding principles, or ideas, for creating public policy, why the various
actors involved are involved, and why they pursue the strategies they
do. The book provides unique and varied insights into the current state
of the art regarding how a range of scholars understand such paradigms,
and public policy ideas, both conceptually and empirically. It does this
by drawing together contributions from leading political science and
social science researchers, to provide a multidimensional set of perspec-
tives on how paradigm-related elements such as policy ideas, coalitions,
discourses, interests, crises, anomalies and routines contribute to policy
development and our understanding of that process. As academics, we
are conscious that, by presenting a variety of perspectives in one book,
we and our readers can learn from each other.
Although a variety of books look at the topic of ideas and their impact
on policy making, we have placed policy paradigms at the centre of
our focus in this volume, and use the policy making/policy change pro-
cess lenses to perfect our understanding of these phenomena. We pro-
vide a critical analysis of what is currently taking place at the nexus of
discourses, ideas and discussions of policy anomalies that has resulted
in the extant iterations of the policy paradigm concept and its appli-
cation to policy studies. Readers will see both the commonalities and
3
4 Policy Paradigms in Theory and Practice
differences across the concepts of policy paradigms used in the book and
how these concepts are evolving and changing.
The chapters are grouped into related sections, but each contribution
is also a self-contained unit. In this way, readers can, by examining
just one chapter, gain an insight into specific aspects of contemporary
thinking about policy paradigms, how that thinking has evolved and
how it is likely to develop in future. Students of public policy are also
provided with the opportunity to familiarize themselves with a more
comprehensive appreciation of policy paradigms, as the book presents
multiple examples of the application and critique of paradigms in the-
ory and in their practical application to policy developments.
In summation, our hope is that readers will find this a useful volume
in assisting them gain a more comprehensive appreciation of the con-
cept and application of the policy paradigm notion in contemporary
policy studies. In providing the readers with a varied set of studies we
hope to encourage them to investigate further those aspects of the para-
digm idea that interest them, or that they find useful in comprehend-
ing aspects of policy-making, and in so doing help push forward the
research frontier on this subject.
This chapter seeks to place the volume and its aims in context
and provide the reader with a guide to the wide-ranging, diverse
and thought-provoking contributions on policy paradigm made by
its contributors. Here we will show how the other chapters in the
book link together and emphasize the importance of better defin-
ing paradigms, their origins and diversity. As the book looks at how
paradigm-based theoretical frameworks tie cognitive ideas, discourses
and coalitions to policies and to the norms and values of the wider
society, the chapter aims to show how understanding policy in such
a paradigmatic manner can provide a better appreciation of why and
how policies change and evolve as they do. Such insights can help us
appreciate how polities develop their unique and sometimes confus-
ing characteristics and policies.
The chapter is structured in four main sections. The first section
looks at the broader academic context within which the book is set
and at the origins and development of the policy paradigm concept.
The second section looks at how policy paradigms, ideas and discourses
are intertwined. The other two sections provide the reader with an
introduction to several outstanding research questions in the field and
an overview of the book’s structure and objectives, highlighting how
each of the chapters help us to better understand policy paradigms and
policy change.
John Hogan and Michael Howlett 5
Policy paradigms and the study of policy change
According to Carson (2004, p. 38) ‘a policy paradigm is a cognitive

model shared by a particular community of actors, and which facili-
tates problem solving’. Similarly, Baumgartner (2013, p. 252) states that
‘when ideas are widely shared by an entire policy community, they can
be called a paradigm’. ‘Ideas on steroids’ is how Baumgartner (2014,
p. 476) has referred to policy paradigms’ significance in the context of
all policy ideas. It is ‘a set of coherent and well-established policy ideas’
capable of having an impact on the content of public policy (Daigneault,
2014, p. 482).
In talking about policy paradigms we are talking about policy dynam-
ics, as the idea of a policy paradigm is one of an ideational construct
that provides some continuity to policy content and discourse over
time. But, although the term policy change is something that is often
spoken of, written about and considered, our understanding of policy
change processes remains limited. As Capano (2009, p. 7) noted, due
to the multidimensional nature of policy dynamics, ‘policy change is
a very ambiguous area of academic study, and one full of pitfalls’. This
is because recognizing how public policy develops is complex and dif-
ficult. Policy change is a multifaceted process that must be understood
in the context of larger societal/political change, but is not limited to it
(Feldstein, 1994). The complex dynamics of policy change constitute a
significant obstacle to further progress in our understanding of policy
making. However, the study, development and evolution of the idea
of a policy paradigms is one which has promised to help resolve this
problem and enable us to better comprehend policy change by under-
standing how ideational factors can structure, or limit, policy debate
and action. This is important because comprehending the policy change
process, how policies evolve and develop over time, is vitally important
in gaining deeper insights into how societies develop.
Not surprisingly, efforts to comprehend policy change have fostered
a variety of theoretical frameworks. The theoretical, and accompany-
ing empirical, literature has developed through the efforts of researchers
to examine the spectrum of public policies from a variety of perspec-
tives, which has provided a rich diversity of comparative and single-case
examinations at various levels of governance – international, national
and subnational. The frameworks used in policy studies are myriad and
have included, but are not limited to, the multiple stream approach, the
punctuated equilibrium framework, the advocacy coalition framework,
the path dependency framework, notions of layering, exogenous shocks,
focusing events, displacement, critical junctures, and drift, as well as

the study of epistemic communities, policy entrepreneurs and barri-
ers to change (Birkland, 1997; Cohen, March, & Olsen, 1972; Garrett &
Lange, 1995; Haas, 1992, 2004; Howlett, 2009; Jenkins-Smith, 1990;
Jones & Baumgartner, 2005; Jones & Jenkins-Smith, 2009; Kleistra &
Mayer, 2001; Legro, 2000; Meijerink, 2005; Mintrom & Norman, 2009;
Nohrstedt, 2011; Sabatier, 1988; Streeck & Thelen, 2005; True, Jones, &
Baumgartner, 2007; Weible et al., 2011; Zahariadis, 1999). These frame-
works each seek to provide glimpses into the mechanics of policy devel-
opment, windows of understanding, but approach the problem from
slightly different angles.
The result has seen the creation of a variety of competing ontologi-
cal perspectives on the subject of policy change. For instance, policy
change has been studied from the perspective of exogenous shocks,
incremental changes and uncertainties as well as the perspective of
groups and coalitions who have a vision of how policies can change. As
Howlett (2009, p. 241) pointed out, ‘most attention to date has focused
upon homeostatic models in which exogenously-driven shocks under-
mine institutionally entrenched policy equilibria’. But, as Blyth (2011,
p. 86) noted, ‘what’s actually exogenous and what is endogenous to
the social world is oftentimes analytically, not empirically, adjudicated’
and, moreover, such institutionally based conceptions fail to take into
account the key role played by policy ideas in affecting the substance
of policy content and change. As Blyth (1997) also argued, ideas are
important objects in the investigation of the context and content of
policy dynamics which cannot be ignored.
The study of policy paradigms begins from the recognition that ideas
are important, underlying processes of policy change and stability and a
key to appreciating patterns and processes of policy dynamics (Lewis &
Steinmo, 2010). To attempt to understand policy change also serves to
shine another, and much needed, light into the black box of policy mak-
ing which Heclo (1974, pp. 305–6) described as a form of collective puz-
zlement on society’s behalf, providing greater transparency for the wider
society as to how choices are made and decision arrived at. Investigating
the complexities and nuances of policy ideas is thus expected to provide
us with a window into our own society, how it seeks to solve problems
and how the solutions it generates often have unforeseen consequences,
as they are frequently constrained by earlier policy decisions as well as
exogenous factors. Béland (2009) has encouraged this growing interest
in ideas, as he regards them as central to our understanding of public
policy.
Some of the most significant works of the last century developed

the concept of a paradigm while trying to understand large-scale pol-
icy change. This includes Weir and Skocpol’s (1985) examination of
Keynesian responses to the great depression and Hall’s (1989) seminal
work on the spread of Keynesian ideas. For Hall (1993) the spread of the
Keynesian paradigm relied on extant arrangements aligning with new
ideas. But, empirical failure was also necessary for Hall (1993) as were
sociological and discursive factors (Blyth, 2013, p. 204). In this case,
paradigm change was episodic, as in the Kuhn (1962) model.
There have been many definitions of ‘paradigms’ put forward in this
work, from Kuhn (1962) onwards, but we are particularly interested in
the policy paradigms concept as developed by Hall (1990, 1993). In his
path-breaking work, Hall (1993) sought to overcome the problems that
led to questioning of how the basic ideas behind extant and new poli-
cies are related, change and transform. In this respect, Hall’s 1993 article
pushed the concept of policy paradigms to the centre of our understand-
ings of public policy (Baumgartner, 2014, p. 475); despite the fact that
the empirical basis of Hall’s (1993) work was subsequently challenged by
Oliver and Pemberton (2004).
While useful, in and of itself, as a means of describing the ideational
components of policy (Campbell, 1998; Hall, 1993; Kern, Kuzemko, &
Mitchell, 2014), in Hall’s work the concept of a policy paradigm pro-
vides a normative and cognitive reading as to how policy should change
and develop – as a solution to the extant problem (Campbell, 1998). As
Princen and ’t Hart (2014, p. 472) argue this is very useful in the assess-
ments of policy dynamics since ‘the presence or absence of policy para-
digms can, we believe, predict the pattern of policy change that a policy
sector is likely to display’.
A better appreciation of paradigms, and their capacity to explain the
policy process, is thus expected to enable us to better explain and com-
prehend public policy dynamics in the world around us. This volume
shows how a selection of theoretical frameworks, at the research fron-
tier, are employed in understanding extant public policies, policy mak-
ing and historical and ongoing policy change. In addition, it provides
practical examples of the application of these paradigms and what we
can learn from them. Although the contributing authors look at policies
from a variety of frameworks, each with somewhat different perspec-
tives on the role of ideas, beliefs, coalitions, discourses, interests and
crises in affecting policy ideas and policy-making, the book shows that,
across the paradigms employed and topics investigated, there are com-
monalities and consistencies in relation to how the various theoretical
frameworks examine public policy and its development. These common-

alities derive from the shared lineage contained within the paradigms –
effectively their shared intellectual DNA. It is to paradigms, ideas and
the discourse surrounding them that this volume is focused.
The context of the book
The last half decade has witnessed the onset of what is being referred
to as the ‘Great Recession’. After three decades marked by the apparent
triumphant march of free market economic ideas and policies across
the world, we have seen private companies, particularly in the form
of banks and other financial institutions, rescued by states and their
citizens (Hogan, Donnelly, & O’Rourke, 2010). For whatever reason,
these institutions have been deemed to be ‘too big to fail’. As a con-
sequence, the ideas underlying the doctrine of free market capitalism
are being questioned; a similar phenomenon to that of the late 1970s
examined by Hall when then orthodox ideas underlying communism
and Keynesianism struggled in the face of efforts to maintain price sta-
bility and low unemployment (Rutland, 1994, p. xi). This suggests the
free market paradigm and ideas that underlie many current policies,
financial and otherwise, may be failing. However, it is by no means
obvious that this will lead to policy change. We are left asking if many
current policies and ideas will change and evolve to respond to the
situation, or if states will continue to cling to them in the hope that
the problems with the political economy will come right? A better
understanding of the role of policy paradigms is required to answer
these questions.
This is because paradigmatic change is not an automatic process put
into motion by crisis, as some would have it. Just because there is a cri-
sis, or policy failure, does not mean change will result. As Baumgartner
(2013, p. 243) points out, ‘where the status quo policy can be demon-
strated to be functioning reasonably well, or where there is no widely
accepted alternative policy available, significant policy change is
unlikely’. Or, as Blyth (2013, p. 209) noted, ‘It is entirely possible that
the dominant paradigm is seen to fail and that nothing in particular
comes along to replace it.’
The policy reality is that ‘occasionally people come up with new ideas
for policy solutions, but for the most part they work with old ideas,
thinking about ways to reformulate them or combine them with others’
(Mintrom, 2000, p. 43). Governments typically react quite slowly in
response to new ideas (Mintrom, 2000). Even when greater degrees of
change are required, or desired, any change of this type can have unin-
tended and unanticipated consequences – making it risky for politi-
cians to enact (Hood, 2010). There can be reluctance to engage in policy
change even when extant policy has failed – such a window of oppor-
tunity is a necessary, but not a sufficient, condition for policy change
(Keeler, 1993; Kingdon, 1984).
Recently, Hall (2013) pointed to policy paradigms as the means by
which policy change occurs, but also noted that on their own are
insufficient for transformative change. As Hogan (2006) and Donnelly
and Hogan (2012) argue, the presence of policy and political entre-
preneurs, championing ideas, is crucial in driving the policy change.
But, there is no guarantee that even in a crisis a group of policy entre-
preneurs is on hand to implement a paradigmatic shift (Baumgartner,
2013; Birkland, 2004).
As Blyth (2001, p. 4) noted, once a policy has become institutionally
embedded, ‘policy making becomes possible only in terms of these
ideas’. The greater the level of consensus surrounding a policy and the
ideas it is based upon, the greater its level of protection – ensuring its
continuity. Yet, despite the fact that there has been no radical over-
throw of the current economic policy framework, we will continue to
see policy change and evolve incrementally. This is because the ideas
underlying extant policy can be undermined by alternative ideas over a
longer period than just the crisis phase – meaning that a crisis is not the
only time when extant ideas and policy are discredited.
Re-evaluating policy paradigms: Their relationship

to ideas and discourses
To better understand the role of policy paradigms, it is important to begin

at the beginning; with the role of ideas in policy-making more generally.
For Hall (1993), as well as Blyth (2013) and Baumgartner (2014, p. 475)
and others, the power of ideas is critically important to our understand-
ing of policy change, as reflected in their centrality to frameworks such as
advocacy coalitions framework (ACF) (Sabatier, 1988; Sabatier & Jenkins-
Smith, 1993) and punctuated equilibrium (Baumgartner & Jones, 1991,
1993) approaches.1 When it comes to paradigms, ideas matter, ‘particu-
larly in the form of structured complexes of ideas’ (Carson, 2004, p. 16).
Ideas are the lifeblood of paradigms.
This stresses, as Campbell (2002) noted, that it is vital to the better
understanding of policy change that we understand how ideas affect
policy making. Ideas reduce uncertainty, allow coalitions to form
around them, can be used to challenge existing policies, build new poli-
cies to replace older one, and produce policy and institutional stability.
Thus, ‘ideas are more powerful than commonly understood and can
have a definite and long lasting impact on policy’ (Pérez-Caldentey &
Vernengo, 2007, p. 1). However, exactly how this happens remains
a matter of some controversy in policy studies, as each framework
of analysis understands the role of ideas in policy making slightly
differently.
According to Daigneault (2014), for example, there is now wide-
spread recognition of the power and role of ideas in the field of public
policy. Ideas and paradigms are seen as crucial in determining policy
choices due to uncertainty over the basic workings of policy, the dif-
ficulties of interpreting policy effectiveness and the lack of agreement
over what constitutes ‘correct’ policy (Baumgartner, 2013; McNamara,
1998, p. 57). The idea of a policy paradigm offers a conceptual frame-
work to help us understand events and their causes, as well as prob-
lems and their definition, and what criteria can resolve these problems
(Carson, 2004).
Ideas are themselves influenced by, and embedded in, political dis-
course (be that descriptive or metaphoric), being used to justify and
articulate a particular view of reality that feeds back into the theoreti-
cal frameworks (Berman, 2001). In this respect Cox and Béland (2013)
argue that it is essential we appreciate such factors as the valence of
policy ideas, by which they mean the emotional quality of those ideas,
be they positive or negative, which can serve to differentiate one idea,
or set of ideas, from another. One set of ideas can become embedded in
the political discourse and become cognitive locks while another set can
become the means to breaking these locks (Schmidt, 2008; Skogstad &
Schmidt, 2011, p. 3).
A central precept of the notion of a policy paradigm is that ideas exist-
ing in the absence of a policy paradigm have less influence on policy
than ones bundled together into more or less interlocking sets of ideas
(Princen & ’t Hart, 2014). For Daigneault (2014, p. 482) policy para-
digms, in effectively simplifying reality by referring only to policy ideas
that are coherent and powerful, convey information better than indi-
vidual ideas.
Consequently, how policy paradigm research deals with the topic of
ideas is very important, as any time an extant policy is found wanting,
its underlying ideas, and the validity of those ideas, can be called into
question. This results in discourse that may serve to legitimize, or chal-
lenge, the extant order (Carson, 2004).
Outstanding issues in policy paradigm research
This is a different conception of policy paradigms, than often found in

the Hall-inspired literature on the subject, which assumes a more auto-
matic and ‘rational’ process of policy change. Determining the exact
contours of a paradigm and the forces which lead it to be influential,
and to change, hence represents a new and ongoing area of research in
this field.
In what follows below, several key questions in this research agenda,
ones which motivate the essays in this volume, are set out.
Technical analysis or policy bricolage?

As Howlett and Ramesh (2003) point out, policy making is a problem
solving activity. But, a policy paradigm, in the sense in which it is com-
monly used, constitutes both the model used to construct the solutions
for the problems it is designed to address as well as the definition of
those problems themselves (Sabatier & Jenkins-Smith, 1993). As Carson
(2004) argues, conceptual models thus structure and constrain how
policy is developed, as policy paradigms influence policy making and
vice versa. The policy paradigm constitutes a shared model of reality
that guides our understanding of public policy and public policy making
(Andersen, 1999; Hall, 1993).
This raises several questions about the extent to which paradigmatic
change and development occurs as a result of technical, or objective,
inquiry into the causes of policy failures and how much results from less
precise efforts at problem solving, what Carstensen (2011a) has termed
‘policy bricolage’: a more trial-and-error-filled discursive process of para-
digm construction. As a shared conceptual model for problem solving,
it is important that we better understand how paradigms change, how
policy ideas are developed though discourse, and the different forms of
policy change that result.
Speed and pattern of change?

Such an analysis also raises issues about the speed, or pattern of change,
expected as paradigms shift. An alternative to the Kuhn/Hall model
of episodic shifts, to a new paradigm, is a model that views paradigm
change as much more incremental, gradual and layered in nature
(Carstensen, 2011a; Mahoney & Thelen, 2010; Sabatier & Jenkins-Smith,
1993; Skogstad, 2011; Streeck & Thelen, 2005). Small changes over time
to the objectives and ideas underlying the policy paradigm may accu-
mulate and result in transformation (Carstensen, 2011b; Howlett &
Cashore, 2009; Skogstad & Schmidt, 2011), a phenomenon which, if

accurate, presents a very different model of paradigms and paradigmatic
change than is often found in the existing literature on the subject.
(In)Commensurability?
Carson (2004, p. 39), like Hall and Kuhn before him, has argued that
‘the policy paradigm is an important cognitive-normative concept that
permits the analysis of distinctly different, sometimes incommensura-
ble ways of conceptualizing the issues problems, interests, goals, and
remedies involved in policy making’. That is, ideas in one paradigm are
expected to be incompatible with those found in another.
As Princen and ’t Hart (2014, p. 471) note, this is an important assump-
tion, and ‘incommensurability is arguably a key element in Hall’s [1993]
explanation of radical change, because it precludes gradual change’.
However, if policy ideas are not incommensurable but rather, for exam-
ple, supplement or reinforce each other, then shifts between ideas can be
incremental, or take some other path (Princen & ’t Hart, 2014; Wilder &
Howlett, 2014). This point, like the previous two, is a subject of some
interest and current research in the field.
The structure and objectives of the book
The main aim of this book is to provide readers with a wide-ranging

introduction to the debates surrounding policy paradigms and how
these help us to understand policy making and policy dynamics at the
beginning of the 21st century, while addressing the issues raised above
with respect to current orthodox uses and applications of the concept.
Understanding how policies develop and evolve is vitally important,
as polices have an impact on all of our lives. While there have been a
number of books on this topic, or at least tangential to this topic, in
recent times (see, for example, Béland & Cox, 2010), along with innu-
merable journal articles over the years, what this volume adds to the
discussion on policy paradigms is the presentation of a variety of disci-
plinary perspectives, from some of the leading contributors in the field,
on the critical aspects of the concept raised above. Within the volume
are 14 contributions from 20 authors, based in a wide range of institu-
tions of higher learning across the world. This collection represents a
significant resource from which readers can draw.
The academics whose combined chapters constitute this volume spe-
cialize in the development and utilization of a variety of approaches to
the study of policy paradigms. Although their publications and research
interests are broad, the contributors have been brought together with
the objective of better understanding the role of policy paradigms in
policy making.
One objective of the volume is theoretical and practical continuity
across the chapters. As such, there is an element of overlap between
chapters, so that readers will be able to see how the various approaches
used to study public policy contained herein share key similarities. We
will see that although a wide range of policies are studied, the concepts
of paradigms employed to find answers to the nature of policy making
and policy change are in ways very similar. This allows theory building
and concept formation to be cumulative.
As already noted, this volume presents a series of unique insights
into various aspects of the policy paradigm concept. To make manag-
ing these contributions easier, the volume has been divided into three
parts, each containing a number of interrelated chapters that exam-
ine an overarching theme, or set of themes, through the use of the
idea of a policy paradigm, providing readers with a variety of lenses
on the paradigm concept. These chapters are relatively short, but cap-
tured within each is the insight of a specialist’s expertise on the policy
paradigm they use to understand the policy area of interest to them.
In this respect, the reader is presented with a wealth of knowledge and
learning, courtesy of some of the world’s leading political and social
scientists.
The first part, consisting of chapters 1 through 4, provides the reader
with an overview of policy paradigms and insights into the theory of
what constitutes a paradigm and their use in policy analysis. Some of
the issues covered here include gaining an appreciation of what is a
broadly acceptable definition of a policy paradigm, in light of the fact
that recent adaptations have yielded a significantly softer image of para-
digms and policy paradigms than originally espoused by either Kuhn
(1962) or Hall (1990, 1993); how to recognize a paradigm through its
ideational nature and how to define and measure it using specific guide-
lines and paradigm shifts; and an examination of how successful the
new concept of discursive institutionalism is in marrying ideas and
institutions into a comprehensive theory of policy change. These chap-
ters present a range of perspectives on our current and evolving under-
standing of policy paradigms.
The second part, chapters 5 through 8, looks at how policy paradigms
can help us understand the processes of policy change. The chapters
provide a review of the similarities and differences between Hall’s (1993)
focus on policy making within paradigms and the other attempts to
focus on translating beliefs into policy; the development and testing of a

new theory for comprehending the role of ideas in the policy process by
focusing on actors’ treatment of policy anomalies; the use of Foucault’s
concepts of a dispositif and episteme for building a new concept, the
policy statement, to better grasp the role of ideas and discourse in the
policy process; and, finally, an examination of how emergent paradigms
are received by public managers in their everyday work. These contribu-
tors argue that although much has been done over the past two decades
in terms of improving our understanding of policy paradigms, and what
they can do for us, much still remains to be done in terms of utilizing
the concept. They also point out that this progress in our understanding
of policy paradigms raises new challenges to the very idea of a paradigm
that have to be considered.
The third part, chapters 9 through 13, focuses on a range of policy
paradigms employed in a series of case studies of policy change pro-
cesses. The issues examined include industrial policy change; the use
the ostensive–performative routines framework to understand the
move from ideas to policy implementation in educational policy har-
monization across the European Union; how consensus was reached
between environmentalists and the oil and gas industry in a US state
on new disclosure rules for “fracking”; the use of a communications
framework in developing a new perspective on information generation,
information supply, and its influence on policy dynamics in the US fed-
eral government; and the setting out of a new framework for measuring
and explaining how and why policy paradigms change by looking at
the United Kingdom’s energy policy. The authors here provide chapters
that look at both micro (country/state specific) and macro (pan state/
federal) instances of policy change using the paradigm concept. They
provide insights into how policy paradigms can help us understand
how policy change differs across jurisdictions, time and the size of the
jurisdictions involved.
The final chapter of the volume contains additional reflections on the
state of the art of policy paradigm studies, along with a discussion of the
future research agenda in this field.
In all, we feel this volume encompasses a great range of knowledge
about the role of ideas in policy-making that is of critical importance to
understanding policy paradigms. The chapters highlight how modern
societies, and their policies, are highly complex and integrated entities.
A paradigm for comprehending policy development in one context is
also of use in another context, highlighting the value of an interdiscipli-
nary volume such as this.
Note
1 This point can also be extended to Kingdon’s (1984) multiple streams
approach (Real-Dato, 2009).
References
Andersen, S. S. (1999). European Integration and the Changing Paradigm of Energy
Policy: The Case of Natural Gas Liberalization. ARENA Working Papers WP 99/12.
Baumgartner, F. (2013). Ideas and Policy Change. Governance, 26(2), 239–58.
Baumgartner, F. (2014). Ideas, Paradigms, and Confusions. Journal of European
Public Policy, 21(3), 475–80.
Baumgartner, F., & Jones, B. (1991). Agenda Dynamics and Policy Subsystems.
Journal of Politics, 53(4), 1044–74.
Baumgartner, F., & Jones, B. (1993). Agendas and Instability in American Politics.
Chicago, IL: University of Chicago Press.
Béland, D. (2009). Ideas, Institutions, and Policy Change. Journal of European
Béland, D., & Cox, R. H. (Eds.) (2010). Ideas and Politics in Social Science Research.
New York: Oxford University Press.
Berman, S. (2001). Review Article: Ideas, Norms, and Culture in Political Analysis.
Comparative Politics, 33(2), 231–50.
Birkland, T. A. (1997). After Disaster; Agenda Setting, Public Policy and Focusing
Events. Washington, DC: Georgetown University Press.
Birkland, T. A. (2004). ‘The World Changed Today’: Agenda-Setting and Policy
Change in the Wake of the September 11 Terrorist Attacks. Review of Policy
Research, 21(2), 179–200.
Blyth, M. (1997). Moving the Political Middle: Redefining the Boundaries of State
Action. The Political Quarterly, 68(3), 217–314.
Blyth, M. (2001). The Transformation of the Swedish Model: Economic Ideas,
Distributional Conflict and Institutional Change. World Politics, 54(1), 1–26.
Blyth, M. (2011). Ideas, Uncertainty and Evolution. In D. Béland & R. Cox (Eds.),
Ideas and Politics in Social Science Research (pp. 83–102). Oxford: Oxford
University Press.
Blyth, M. (2013). Paradigms and Paradox: The Politics of Economic Ideas in Two
Moments of Crisis. Governance, 26(2), 197–215.
Campbell, J. L. (1998). Institutional Analysis and the Role of Ideas in Political
Economy. Theory and Society, 27(5), 377–409.
Campbell, J. L. (2002). Ideas, Politics, and Public Policy. Annual Review of Sociology,
28(1), 21–38.
Capano, G. (2009). Understanding Policy Change as an Epistemological and
Theoretical Problem. Journal of Comparative Policy Analysis: Research and Practice,
11(1), 7–31.
Carson, M. (2004). From Common Market to Social Europe?: Paradigm Shift and
Institutional Change in European Union Policy on Food, Asbestos and Chemicals,
and Gender Equality. Stockholm: Stockholm University.
Carstensen, M. (2011a). Paradigm Man vs. the Bricoleur: Bricolage as an
Alternative Vision of Agency in Ideational Change. European Political Science
Review, 3(1), 147–67.
Carstensen, M. (2011b). Ideas Are Not as Stable as Political Scientists Want Them to
Be: A Theory of Incremental Ideational Change. Political Studies, 59(3), 596–615.
Cohen, M., March, J., & Olsen, J. (1972). A Garbage Can Model of Organisations
choice. Administrative Science Quarterly, 17(1), 1–25.
Cox, R. H., & Béland, D. (2013). Valence, Policy Ideas, and the Rise of Sustainability.
Governance, 26(2), 307–28.
Daigneault, P.-M. (2014). Puzzling about Policy Paradigms: Precision and Progress.
Journal of European Public Policy, 21(3), 481–4.
Donnelly, P., & Hogan, J. (2012). Understanding Policy Change Using a Critical
Junctures Theory in Comparative Context: The Cases of Ireland and Sweden.
Policy Studies Journal, 40(2), 324–50.
Feldstein, M. (1994). American Economic Policy in the 1980s. Chicago, IL: University
of Chicago Press.
Garrett, G., & Lange, P. (1995). Internationalization, Institutions, and Political
Change. International Organization, 49(4), 627–55.
Haas, P. M. (1992). Introduction: Epistemic Communities and International
Policy Coordination. International Organization, 46(1), 1–35.
Haas, P. M. (2004). When Does Power Listen to Truth? A Constructivist Approach
to the Policy Process. Journal of European Public Policy, 11(4), 569–92.
Hall, P. A. (1989). The Political Power of Economic Ideas: Keynesianism across Nations.
Princeton, NJ: Princeton University Press.
Hall, P. A. (1990). Policy Paradigms, Experts and the State: The Case of Macro-
economic Policy Making in Britain. In S. Brooks & A. Gagnon (Eds.), Social
Scientists, Policy and the State (pp. 53–78). New York: Praeger.
Hall, P. A. (1993). Policy Paradigms, Social Learning, and the State: The Case of
Economic Policy Making in Britain. Comparative Politics, 25(3), 275–96.
Hall, P. A. (2013). Brother, Can You Paradigm. Governance, 26(2), 189–92.
Heclo, H. (1974). Modern Social Policies in Britain and Sweden: From Relief to Income
Maintenance. New Haven, CT: Yale University Press.
Hogan, J. (2006). Remoulding the Critical Junctures Approach. Canadian Journal
of Political Science, 39(3), 657–79.
Hogan, J., Donnelly, P. F., & O’Rourke, B. K. (2010). Reflections on Issues in
Contemporary Irish Business and Society. In J. Hogan, P. F. Donnelly, &
B. K. O’Rourke (Eds.), Irish Business and Society: Reflections on Governing, Participating
and Transforming in the 21st Century (pp. 1–27). Dublin: Gill and Macmillan.
Hood, C. (2010). The Blame Game: Spin, Bureaucracy, and Self-Preservation in
Government. Princeton, NJ: Princeton University Press.
Howlett, M. (2009). Process Sequencing Policy Dynamics: Beyond Homeostasis
and Path Dependency. Journal of Public Policy, 29(3), 241–62.
Howlett, M., & Cashore, B. (2009). The Dependent Variable Problem in the Study
of Policy Change: Understanding Policy Change as a Methodological Problem.
Journal of Comparative Policy Analysis: Research and Practice, 11(1), 33–46.
Howlett, M., & Ramesh, M. (2003). Studying Public Policy: Policy Cycles and Policy
Subsystems (2nd ed.). Oxford: Oxford University Press.
Jenkins-Smith, H. C. (1990). Democratic Politics and Policy Analysis. Pacific Grove,
CA: Brooks/Cole.
Jones, B., & Baumgartner, F. (2005). The Politics of Attention: How Government
Prioritizes Problems. Chicago, IL: University of Chicago Press.
Jones, M., & Jenkins-Smith, H. C. (2009). Trans-Subsystem Dynamics: Policy

Topography, Mass Opinion, and Policy Change. Policy Studies Journal, 37(1),
37–58.
Keeler, J. (1993). Opening the Window for Reform: Mandates, Crises, and
Extraordinary Policy-Making. Comparative Political Studies, 25(4), 433–86.
Kern, F., Kuzemko, C., & Mitchell, C. (2014). Measuring and Explaining Policy
Paradigm Change: The Case of UK Energy Policy. Policy & Politics, 42(2), 1–18.
Kingdon, J. W. (1984). Agendas, Alternatives, and Public Policies. Boston, MA: Little,
Brown.
Kleistra, Y., & Mayer, I. (2001). Stability and Flux in Foreign Affairs: Modelling
Policy and Organizational Change. Cooperation and Conflict, 36(4), 381–414.
Kuhn, T. S. (1962). The Structure of Scientific Revolutions. Chicago, IL: University of
Chicago Press.
Legro, J. W. (2000). The Transformation of Policy Ideas. American Journal of
Political Science, 44(3), 419–32.
Lewis, O., & Steinmo, S. (2010). Taking Evolution Seriously in Political Science.
Theory in Biosciences, 12(1), 235–45.
Mahoney, J., & Thelen, K. (Eds.) (2010). Explaining Institutional Change. Ambiguity,
Agency, and Power. Cambridge: Cambridge University Press.
McNamara, K. R. (1998). The Currency of Ideas: Monetary Politics in the European
Union. Ithaca, NY: Cornell University Press.
Meijerink, S. (2005). Understanding Policy Stability and Change: The Interplay of
Advocacy Coalitions and Epistemic Communities, Windows of Opportunity,
and Dutch Coastal Flooding Policy 1945–2003. Journal of European Public Policy,
12(6), 1060–77.
Mintrom, M. (2000). Policy Entrepreneurs and School Choice. Washington, DC:
Georgetown University Press.
Mintrom, M., & Norman, P. (2009). Policy Entrepreneurship and Policy Change.
Nohrstedt, D. (2011). Shifting Resources and Venues Producing Policy Change in
Contested Subsystems: A Case Study of Swedish Signals Intelligence Policy.
Oliver, M. J., & Pemberton, H. (2004). Learning and Change in 20th-Century
British Economic Policy. Governance, 17(3), 415–41.
Pérez-Caldentey, E., & Vernengo, M. (2007). Introduction: Ideas, Policies and
Economic Development in the Americas. In E. Pérez-Caldentey & M. Vernengo
(Eds.), Ideas, Policies and Economic Development in the Americas (pp. 1–8).
New York: Routledge.
Princen, S., & ’t Hart, P. (2014). Putting Policy Paradigms in Their Place. Journal of
European Public Policy, 21(3), 470–4.
Real-Dato, J. (2009). Mechanisms of Policy Change: A Proposal for a Synthetic
Explanatory Framework. Journal of Comparative Policy Analysis: Research and
Practice, 11(1), 117–43.
Rutland, P. (1994). The Politics of Economic Stagnation in the Soviet Union: The Role
of Local Party Organs in Economic Management. Cambridge: Cambridge University
Press.
Sabatier, P. A. (1988). An Advocacy Coalition Model of Policy Change and the
Role of Policy-Oriented Learning Therein. Policy Sciences, 21(3), 129–68.
Sabatier, P. A., & Jenkins-Smith, H. C. (Eds.) (1993). Policy Change and Learning: An
Advocacy Coalition Approach. Boulder, CO: Westview Press.
Schmidt, V. A. (2008). Discursive Institutionalism: The Explanatory Power of
Ideas and Discourse. Annual Review of Political Science, 11, 303–26.
Skogstad, G. (Ed.) (2011). Policy Paradigms, Transnationalism, and Domestic Politics.
Toronto: University of Toronto Press.
Skogstad, G. D., & Schmidt, V. A. (2011). Introduction: Policy Paradigms,
Transnationalism, and Domestic Politics. In G. D. Skogstad (Ed.), Policy
Paradigms, Transnationalism, and Domestic Politics (pp. 3–35). Toronto: Toronto
University Press.
Streeck, W., & Thelen, K. (2005). Introduction: Institutional Change in Advanced
Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond Continuity.
Institutional Change in Advanced Political Economies (pp. 1–39). Oxford: Oxford
University Press.
True, J. L., Jones, B., & Baumgartner, F. (2007). Punctuated-Equilibrium Theory:
Explaining Stability and Change in Public Policy. In P. A. Sabatier (Ed.), Theories
of the Policy Process (pp. 155–88). Boulder, CO: Westview Press.
Weible, C. M., Sabatier, P. A., Jenkins-Smith, H. C., Nohrstedt, D., Henry, A. D., &
de Leon, P. (2011). A Quarter Century of the Advocacy Coalition Framework:
An Introduction to the Special Issue. Policy Studies Journal, 39(3), 349–60.
Weir, M., & Skocpol, T. (1985). State Structures and the Possibilities for ‘Keynesian’
Responses to the Great Depression in Sweden, Britain, and the United States. In
P. B. Evans, D. Rueschemeyer, & T. Skocpol (Eds.), Bringing the State Back In
(pp. 107–64). New York: Cambridge University Press.
Wilder, M., & Howlett, M. (2014). The Politics of Policy Anomalies: Bricolage and
the Hermeneutics of Paradigms. Critical Policy Studies, 8(2), 183–202.
Zahariadis, N. (1999). Ambiguity, Time, and Multiple Streams. In P. A. Sabatier
(Ed.), Theories of the Policy Process (pp. 73–93). Boulder, CO: Westview Press.
2
What Is a Policy Paradigm?
Overcoming Epistemological
Hurdles in Cross-Disciplinary
Conceptual Adaptation
Matt Wilder
Introduction
The advent of the new institutionalism in the policy sciences has brought
with it a tendency to borrow concepts from disparate fields of study
(Immergut, 1998; Koelble, 1995). Though it is a discipline vulnerable
to critique for embracing metaphors rather than models (Dobuzinskis,
1992; Dowding, 1995), three decades of new institutionalism has pro-
duced theories whose conceptual origins can be traced to cybernetics
(Steinbruner, 1974), evolutionary biology (Krasner, 1984), seismic geol-
ogy (Jones et al., 2009), anthropology (Mahoney, 2000), economics
(North, 1990), and organizational theory (Agyris & Schön, 1978). From
the philosophy of science, the discipline has borrowed and adapted
Thomas Kuhn’s concept of paradigms to explain the dynamics of long-
term policy change (Kuhn, 1962, 1970a), culminating in the routine
mention of paradigms in policy journals since the early 1990s (Béland &
Cox, 2013; Skogstad, 2011).
Policy paradigms have been cited by thousands of scholars as a deter-
minant of policy outcomes since Peter Hall developed and popularized
the concept in two groundbreaking pieces (Hall, 1990, 1993; but see
also Allison, 1969; Jenson, 1989; Merton, 1957). Yet, in spite of the con-
cept’s popularity, consistency with the original Kuhnian model has been
accepted without serious examination by academics. Whereas Kuhn’s
(1962) The Structure of Scientific Revolutions precipitated tremendous
debate in the philosophy of science, Hall’s (1990) adaptation of para-
digms to the study of public policy has gone without much reflection on
the appropriateness of the analogy to the natural sciences and has sel-
dom been the subject of rigorous testing (Daigneault, 2014; Skogstad &
Schmidt, 2011). The argument put forth in this chapter is that policy
19
paradigms are typically viewed as background variables, considered

causally influential by most but rendered theoretically operable by few.
The reason paradigms have failed to become better integrated into other
theories of the policy process is that the concept as it was defined by its
architects, Thomas Kuhn and Peter Hall, is not sufficiently flexible to be
analyzed as an explanatory variable in the policy sciences.
To transcend its status as a poorly operationalized background varia-
ble, the concept of policy paradigms must undergo reconfiguration. This
is admittedly a dangerous enterprise for those interested in maintain-
ing the conceptual essence of paradigms (cf. Princen & ’t Hart, 2014).
Conceptual purity has, however, already been sacrificed as scholars have
attempted to locate paradigms in empirical accounts of policy change
and continuity. Consistent with an observation made long ago by
Lindblom (1959), Capano (2003), for example, argues that differences
exist between dominant and hegemonic paradigms, and suggests that
“administrative traditions” allow actors to operate within a subordinate
paradigm while projecting the illusion that they are pursuing new para-
digmatic goals. Howlett and Ramesh (1998) suggest that policy commu-
nities need not ever have a dominant idea set and that, even in instances
where few ideas exist, there is a strong tendency toward paradigmatic
contest. Béland (2007) speaks of contemporary policy paradigms as pol-
icy hybrids whereby new paradigms are layered upon existing structures
(cf. Hacker, 2004; Peters, 2011; Streeck & Thelen, 2005). Kay (2007) high-
lights the tension that exists between such layers in “synthetic” para-
digms, thereby bringing the core assumption that paradigms constitute
incommensurable world views and pure manifestations of causal inter-
pretations into question.
Given these developments, the softening of the conceptual rigidity
inherent to the original understanding of paradigms deserves comment.
Similar to developments in the history of science (see Crombie, 1994;
Dear, 2001; Shapin, 1996), policy scholars have by this point largely
abandoned the incommensurability thesis emphasized by Kuhn and
advocated by Hall (Surel, 2000). This conceptual reorientation has been
characteristic of attempts to render paradigms more consistent with
observable developments in public policy and public administration.
Unlike developments in the history of science, however, many policy
scholars have avoided recognizing this significant departure from the
original definition of policy paradigms (cf. Kuhn, 1982).
The objective of this chapter is to demonstrate that the notion of para-
digms continues to be a useful one for policy scholars, but that from an
epistemological point of view, policy paradigms do not (and should not)
Matt Wilder 21
retain complete synonymy with Kuhn’s (1962) scientific paradigms. The

following section outlines the original transposition of the Kuhnian
concept to the policy sciences, highlighting the conceptual rigidities
associated with Kuhn’s incommensurability thesis. This is followed by a
discussion on the recent trend toward a softer understanding of policy
paradigms before outlining a method by which this softened image of
paradigms may be rendered operable. Drawing on a method of depend-
ent variable disaggregation developed by Howlett and Cashore (2007),
and exploring its parallels to the concept of causal maps, the alternative
model links agents and ideas by theorizing about the role of epistemic
communities and advocacy networks in temporally distinct stages of the
policy process.
Following this sketch of the alternative method, the epistemological
consequences of commensurable paradigms are discussed in the interest
of expressing what policy paradigms have come to mean in their rela-
tion to Hall’s original theory. The conclusion of the chapter stresses that
better specification of how paradigms are operationalized will provide a
basis for fuller integration of the concept with other influential theories
of the policy process.
Hard paradigms, Kuhn’s gestalt metaphor and

the policy sciences
Kuhn’s ideas on the structure of scientific revolutions are the well-

known subject of five decades of philosophical and historical debate,
and so receive only brief summary here (see Nickles, 2002; Rorty & Wray,
2002). The epistemological crux of Kuhn’s paradigms—that is, normal
science and scientific revolutions—depends heavily on “world views”
and Kuhn’s appeal to gestalt images. Kuhn’s gestalt example is consist-
ent with observations made previously by Feyerabend (1962), Hanson
(1958), Polanyi (1964), and Quine (1960) as well as the vast scholarship
on “ways of seeing” (Berger, 1972; Brown, 1979; Goodman, 1978; Jacob &
Jeannerod, 2003; Strike & Posner, 1992). Kuhn (1962) characterizes
gestalt switches as being akin to ideological or religious conversion inso-
far as, according to Kuhn, one cannot simultaneously see two gestalts or
elements of one gestalt as the component parts of another.1
Kuhn’s metaphorical appeal to ideology and religion is however prob-
lematic in relation to his (somewhat contradictory) insistence that only
mature, observationally tested sciences may exist within a paradigm
(Kuhn, 1962, p. 12).2 That is, as Masterman (1970) points out, Kuhn’s
multiple definitions for paradigms give rise to conceptual inconsistency
since Kuhn posits an observationally derived concept of paradigms, on

one hand, and a “metaphysical” one, on the other. Kuhn’s recourse to
value-based analogies, however, has a firm basis with respect to what
Masterman calls Kuhn’s metaparadigms.3
In spite of Kuhn’s partial engagement with metaphysics, in drawing
a parallel between scientific and political revolutions, The Structure of
Scientific Revolutions took institutional overthrow to be a requisite fea-
ture of paradigm change.
Political revolutions aim to change political institutions in ways

that those institutions themselves prohibit. Their success therefore
necessitates the partial relinquishment of one set of institutions in
favour of another, and in the interim, society is not governed by
institutions at all. Initially it is crisis alone that attenuates the role of
political institutions as we have already seen it attenuate the role of
paradigms. In increasing numbers individuals become increasingly
estranged from political life and behave more and more eccentrically
within it. Then, as the crisis deepens, many of these individuals com-
mit themselves to some concrete proposal for the reconstruction of
society in a new institutional framework. At that point the society is
divided into competing camps or parties, one seeking to defend the
old institutional constellation, the others seeking to institute some
new one. And, once that polarisation has occurred, political recourse
fails. Because they differ about the institutional matrix within which
political change is to be achieved and evaluated, because they
acknowledge no supra-institutional framework for the adjudica-
tion of revolutionary difference, the parties to a revolutionary con-
flict must finally resort to the techniques of mass persuasion, often
including force. (Kuhn, 1962, pp. 93–4)4
Bringing metaparadigms down to earth by linking them to institutional

creation and destruction may be an explanation for why subsequent
authors in the policy sciences, namely Hall (1993) and Oliver and
Pemberton (2004), chose to emphasize this aspect of paradigm change.
This decision was made in spite of the fact that Kuhn clarified his defini-
tion of paradigms in the postscript of the second edition of The Structure
of Scientific Revolutions, where he acknowledged that paradigms had
previously been used in at least two different ways. In Postscript—1969,
Kuhn gives the first definition of paradigms as “the entire constella-
tion of beliefs, values, techniques, and so on shared by the members
of a given community,” while the second definition concerns, rather,
Matt Wilder 23
“one sort of element in that constellation, the concrete puzzle-solutions

which, employed as models or examples, can replace explicit rules as
a basis for the solution of the remaining puzzles of normal science”
(Kuhn, 1970b, p. 174). Kuhn thus defines paradigms in the first sense as
“sociological,” and the second as “paradigms as exemplary past achieve-
ments” (Kuhn, 1970b, p. 174).
Subsequent literature considers Kuhn’s 1969 postscript to be a sig-
nificant relaxation of the radical argument originally presented in The
Structure of Scientific Revolutions (see Kuhn, 1982). While Kuhn deemed it
important to better operationalize paradigms by differentiating between
the sociological aspects of paradigms and paradigms as disciplinary
matrixes upheld by positive exemplars, policy scholars’ definition of
paradigms has been less sophisticated (cf. Campbell, 2002). The series
of revisions that began with Postscript led the way to a softer and more
ontologically commensurable view of paradigms in the philosophy of
science (Horwich, 1993; Kuhn, 2000). Attention to the epistemological
debate that followed from The Structure of Scientific Revolutions is there-
fore a useful exercise in identifying and reconciling some of the prob-
lematic aspects of policy paradigms.
Hall’s (1990) understanding and adaptation of paradigms is suc-
cinctly captured in his introductory piece on policy paradigms, in
which he developed a three-order framework for understanding transi-
tions between incremental policymaking and paradigmatic revolution.
Incremental policymaking—a parallel to Kuhn’s normal science—
is defined by what Hall (1990) terms first and second order change. To
use Kuhnian terminology, first and second order changes are constitu-
tive of policymakers adjusting their techniques for solving paradigmatic
puzzles. These techniques, or policy means, are limited to what are
known in the policy literature as policy instruments and their specific
calibration, or what Hall terms policy settings (Hall, 1990, p. 55; Hood,
1983; Salamon, 2002). Hall’s (1990) third order of change, however,
goes beyond puzzle solving and means-related adjustments, and instead
involves changes to the policy puzzles themselves, or what Hall calls
the abstract goals guiding policy. According to Hall, while first and sec-
ond order alterations to instruments and settings are characteristic of
day-to-day policymaking, third order reorientation of all three elements
(most importantly reorientation of abstract goals) is a requisite feature
of policy revolution.5
While cognizant of the fact that the concept of scientific paradigms is
contentious in the philosophy of science, Hall (1990, pp. 60–1) argues
that the concept is not only a “great heuristic force” but in fact more
amenable to the examination of social science than it is to the study of

the natural sciences. This is presumably because individuals’ bounded
rationality exacerbates the role that belief systems play in the social sci-
ences, invalidating positivistic notions of policy paradigms (cf. Agyris &
Schön, 1978; Hall, 1990, pp. 55–6; March & Simon, 1958). Kuhn’s own
doubts about the applicability of the paradigms concept to social sci-
ences may figure heavily into why Hall adopted only a highly socio-
logical variant of the concept, mainly in the form of the controversial
incommensurability thesis (see Sankey, 1994; Waver, 1996).6 Kuhn’s
metaphorical usage of gestalt images as akin to world views is also reiter-
ated by Hall in his development of the policy paradigms concept.
[P]olicymakers customarily work within a framework of ideas and

standards that specifies not only the goals of policy and the kind of
instruments that can be used to attain them, but also the very nature
of the problems they are meant to be addressing. Like a Gestalt, this
framework is embedded in the very terminology through which poli-
cymakers communicate about their work, and it is influential pre-
cisely because so much of it is taken for granted and unamenable to
scrutiny as a whole. (Hall, 1993, p. 279)7
Hall (1990, pp. 66–7) points out, however, that in social science there
are few true observations in the scientific sense that may give concrete
meaning to terminology. It is evident that Hall’s observation applies
even to terminology that may be germane to a given theory or, still fur-
ther, competing sets of theories. The treatment of unemployment in
Keynesian and monetarist economic doctrines, for example, maintains
both observational and semantic consistency regardless of which para-
digmatic lens is doing the observing, yet unemployment means vastly
different things to Keynesians and monetarists in a theoretical sense
(cf. Doppelt, 1978; see Walters, 2000). That is, unemployment is con-
ceptually the same thing to Keynesians and monetarists, but the term
cannot be utilized in similar ways across the two paradigms due to the
assumptions of the theories. This is perhaps because scientific paradigms
are concerned more with what exists (i.e., the existence of mass or atoms)
whereas social scientific paradigms are concerned with cause–effect rela-
tionships (i.e., we know that inflation exists, but we are unsure of its
effect). Incommensurability in the social sciences is thus theory contin-
gent, while in the natural sciences it tends to be observation contingent.8
Social scientific debate also tends to center not on the meaning of
observations but rather on which observations are deemed important
Matt Wilder 25
or effectual. While this is problematic, it does not negate the concept

paradigms entirely.9 Difficulty in proving or disproving central axioms
in the social sciences suggests, however, that the rules governing the
diachronic aspects of the theory of social scientific paradigms ought to
be relaxed. As is shown in the next section, the ease with which the
meaning of terminology can be transplanted from one policy para-
digm to another is cause for a softer conceptualization of paradigmatic
commensurability.
Toward a softer image of policy paradigms
Attempting to replicate the findings of Hall’s archetypal case study,

Oliver and Pemberton (2004) revisited the case of paradigmatic change
in British monetary policy and made several significant findings.
Most consequential to the theory of policy paradigms was Oliver and
Pemberton’s observation that paradigm replacement need not be abso-
lute (cf. Hay, 2001). Contrary to Hall (1990, 1993), Oliver and Pemberton
(2004, pp. 427, 435) argued that consecutive efforts to affect paradigm
change actually led to “partial” adaptation of new paradigmatic goals
owing in large part to institutional and authoritative resistance to new
ideas. It goes without saying that allowing for the partial acceptance of
paradigmatic tenets vitiates Kuhn’s incommensurability thesis.
In spite of their revisions to Hall’s theory, Oliver and Pemberton do
not reflect on their abandonment of the axiom that paradigms are nec-
essarily incommensurable. Where other contemporary researchers have
bothered to operationalize paradigms at all, the incommensurability
thesis has been treated loosely (Schmidt, 2011; Surel, 2000). Skogstad
and Schmidt (2011) provide a definition of paradigms that is differenti-
ated from other ideational variables, but remains somewhat ambiguous
in the sense that paradigms are used synonymously with rival concepts,
such as image frames and réferéntiels (Jobert & Muller, 1987; Rein &
Schön, 1996). In his examination of paradigmatic change in the area
of global pension reform, Orenstein (2013, p. 276) hints at the possibil-
ity of the partial overthrow of a paradigm with his acknowledgment
that “second order changes themselves subtly reorient the fundamental
objectives of reform.” Orenstein thereby alludes to the use of instru-
ments whose purposes are incongruent with the prevailing paradigm.10
Similarly, Béland (2007) and Kay (2007) speak of policy hybrids or syn-
thetic paradigms. Though Béland (2007) describes a Kuhnian process, in
that early competing paradigms coalesced into a hegemonic paradigm
in the case of social security reform in the United States, Kay (2007)
describes the opposite process with respect to health insurance reform in

Australia, whereby some advocates’ favor of one paradigm over another
shifted, yielding a “tense” relationship between two competing yet
equally institutionalized paradigms before they were joined in synthe-
sis. Aside from violating the incommensurability thesis, observations of
reversibility toward paradigmatic contest, such as Kay’s, contravene the
Kuhnian assumption that once a paradigmatic revolution has occurred,
it will endure until the next revolution (cf. Nisbet, 1972).
The shift toward a softer conceptualization of policy paradigms sug-
gests that policy scholars may currently have more in common with
Kordig (1973) than they do with Kuhn, as Kordig was careful not to
deny that there is some validity to theory-laden observation, but argued
that competing scientific theories have so much in common that some
observations may be “neutral” to multiple theories.11 Partial denotation
as described by Field (1973), being consistent with Kordig’s view, is also
an idea that would improve researchers’ understanding of the semantic
commensurability of social scientific concepts.12 Rather than avoiding
these realities, greater familiarity and engagement with the debate on
scientific paradigms will assist researchers in parsing out the relative
influence of observation and “metaphysical” beliefs in the construc-
tion of policy paradigms. While there has been some effort to do this
in the policy sciences (Blyth, 2013), there remains polarization between
rational and sociological explanations for policy decisions (Hall &
Taylor, 1996).
That said, disciplinary emphasis on the sociological determinants of
paradigm change is apparent in the introduction to a recent special issue
of Governance devoted to the analysis of policy paradigms.
The study of policy paradigms offers a fundamental challenge to

explanations of politics that seek motivations in rational calculations
or the material interests of decision makers. This is not to say that
rationality or interests are unimportant, but that they are also con-
structed. Both are shared by dialogic and iterative processes where
communicating with other actors about goals, beliefs, and values
cause individuals to formulate their own sense of their interests, and
to devise rationalizations for the outcomes they seek. (Béland & Cox,
2013, p. 194)
While this view is compatible with Howlett’s (1994, p. 642) sentiment

that “in the policy world . . . the returns to individuals often run in the
direction opposite to those in scientific communities, with conformity
Matt Wilder 27
rather than iconoclastic behavior reaping the rewards,” Coleman,

Skogstad, and Atkinson (1996) observe that processes toward major
change need not be disproportionately influenced by political or socio-
logical factors, but may, depending on the institutional context, pro-
ceed in the sort of regimented and experiment-oriented fashion Kuhn
describes with respect to scientific progress. With the exception of the
latter authors’ view, explanations of paradigmatic change and stability
in the policy sciences has centered on the socio-political factors con-
tributing to paradigm change, on one hand (see Sabatier, 1988), and
actors’ metaphysical or “Third World” belief structures, on the other.13
Campbell (2002) takes a progressive step by distinguishing between cog-
nitive and normative paradigms, though very little has been explicitly
said about policy as conditioned by exemplary past achievements or
policy as rote puzzle solving under hegemonic disciplinary matrixes.14
Part of the reason policy researchers have focused on power and per-
suasion is that this aspect of paradigm change is consistent with other
influential theories of politics and the policy process (e.g., Heclo, 1974).
Paradigms are also not the only way ideational variables have been
treated in policy research. Goldstein and Keohane (1993) speak of the
mediating effects of actors’ abstract “world views” on their causal beliefs,
while Sabatier’s (1988) influential Advocacy Coalition Framework (ACF)
is predicated on the crystallization of political coalitions based on shared
“deep core” beliefs that structure and demarcate preferences within a
“policy core.” Although Sabatier (1988) moves one step forward by dis-
tinguishing between beliefs about goals and beliefs about appropriate
instruments, ideational variables outside of the paradigms literature suf-
fer from the same lack of operationalization in terms of what it is, spe-
cifically, that ideas affect.
It is possible that analytical inconsistency with respect to how para-
digms are operationalized stems from the fact that many researchers
use the concept for purely heuristic purposes. Hall (1993, p. 290) him-
self recognized the utility of the concept as an organizing device with
his classification of his own research as advancing “a specific argument
about the way in which ideas condition policymaking and how they
change, organized around the concept of ‘policy paradigms.’” If para-
digms are used simply for heuristic purposes (i.e., to organize patterns of
ideational change), researchers do not need to take the Kuhnian parallel
literally. In this case, paradigms need not be fully operationalized but
rather taken as a variable representing the social or cultural influence
of ideas. That said, any causal impact that learning and ideas have on
policymaking will continue to elude policy scholars’ understanding if
paradigms and gestalt switches are intended only to be used metaphori-

cally in the policy sciences (cf. Schmidt, 2011). Rendering ideational
influence operable requires a notion of paradigms that is workable given
the characteristics of policy ideas and the realities of the policy process
(cf. Daigneault, 2014).
Exploring paradigms in an operative sense
Taken in literal form, very few efforts have been made to operationalize
paradigms beyond Hall’s three-order definition. Among the few, Howlett
and Cashore (2007) build upon Hall’s three orders by developing a six-
part disaggregation of policy change. Addressing what has come to be
known as the dependent variable problem in the policy sciences (i.e.,
ambiguity concerning what is meant by “policy”), Howlett and Cashore
(2009, p. 3) contend:
[T]his [Hall’s] initial conceptual effort at classification requires recali-

bration in light of its own logic, as well as in light of the empirical evi-
dence gathered in many cases of policy change analyzed since Hall’s
work was first published. That is, according to Hall’s own emphasis
on distinguishing abstract or theoretical/conceptual goals from spe-
cific programme content or objectives, and operational settings or
calibrations, along with his distinction between the aims or “ends”
of policy and its actual policy requirements (“means”), it is possible
to discern six, rather than three, policy elements that can undergo
change. The implication of this taxonomy is that every “policy” is in
fact a more complex regime of ends and means related goals (more
abstract), objectives (less abstract), and settings (least abstract) than
was suggested by the use of Hall’s original decomposition and defini-
tion of the elements of policy into three “orders.”15
Following Howlett and Cashore, and in the interest of defining specifi-

cally which components of policy undergo change, Figure 2.1 provides
a six-part disaggregation of the dependent variable “policy.”
Three ends-related components of policy are located in Figure 2.1
above the axis, while their corresponding means-related components
are arranged below. These components range from most concrete
(or most observable) to most abstract (or most inductive) as indicated by
the values given at the extremes of the axis. The epistemological argu-
ment that lies behind the six-part disaggregation of policy paradigms
is communicated by the arrows linking the components. Solid arrows
Matt Wilder 29
Policy Ends
Instrument Programmatic Paradigmatic

Targets Objectives Goals
More More
Concrete Abstract
Instrument Instrument Implementation
Settings/ Choices/ Preferences
Calibration Selection
* Policy Means
Causal inference
*
Determinant of
Figure 2.1 Six-part disaggregation of policy

Source: Adapted and modified from Cashore and Howlett (2007) and Howlett (2009).
denote, non-problematically, that one component yields another (i.e.,

implementation preferences yield specific instrument choices). Dashed
arrows, on the other hand, indicate that the components are linked by
way of the causal inferences actors make with respect to cause–effect
assumptions (i.e., instrument calibration will result in instrument targets
being met, or instrument choices will achieve programmatic objectives).
The inferential sequence linking the components of policy can be
expressed as
Y → (A * B) * C * D * E
The “paradigm” is operationalized as the sequence of notations “*,” or the

belief that instrument settings A will assist policy instruments B achieve
instrument targets C which will lead to the fulfillment of programmatic
objectives D which will in turn satisfy paradigmatic goals E.16 In other
words, concrete observations are bridged by “metaphysical” beliefs
about cause–effect relationships. These beliefs are organized, even stand-
ardized, in the implementation preferences of a given policy regime Y,
which determine and delimit the menu of beginning-of-sequence instru-
ment options deemed appropriate by policymakers. Policy regimes that
adhere very strictly to a given paradigm will have very clearly defined
implementation preferences, while others will be more tolerant of
experimentation within a wider menu of policy means.
Having established an inferential basis for policy paradigms predicated
upon causal beliefs, it is possible to adapt the Howlett and Cashore (2007)
taxonomy to the concept of causal maps as per Axelrod (1973) and Eden
(1992). Understood in this way, abstract elements are linked to concrete
observation by the positive correlations actors make between lower and
higher order components of a given policy. Experience and learning are
then hypothesized to enhance or erode the strength of positive correla-
tions, though actors’ prior beliefs have a significant influence on the
effect of positive and negative observations (Gilardi, 2010; Wilder &
Howlett, this volume—for an empirical application of the causal map-
ping technique to policy paradigms see Princen & Van Esch [2013]).
More can be said, however, about which agents affect what elements of
policy, given that some agents and groups disproportionately focus their
cognitive capacities on some elements of policy problems and solutions
over others. The literature on transnational advocacy and learning is
informative in linking certain types of groups to these sorts of cognitive
biases. A distinction can be made between three types of groups in this
respect: those who engage in almost exclusively normative discourses,
those who approach policy in a more technocratic and means-oriented
fashion, and those who take a more holistic approach identifying policy
problems and solutions.
Keck and Sikkink (1998), for example, differentiate transnational
advocacy networks from epistemic communities, with the former being
much more invested in value-based advocacy than technical aptitude
(Stone, 2004). Meanwhile, according to Haas (1992), epistemic com-
munities are often able to monopolize the normative discourse because
they are able to draw clear linkages between their causal (cognitive)
beliefs and their principled (normative) beliefs. Dostal (2004) uncovers
a division of labor within the Organisation for Economic Cooperation
and Development (OECD) precisely along these lines, with policy advo-
cates in the Ministerial Council—those who engage state representatives
to popularize OECD policy—being distinct from those formulating the
specifics of OECD policy recommendations.
Thus, what we typically conceive of as advocacy groups will have a
much greater influence on agenda-setting than they will on the actual
formulation of policy. This is because they will tend to possess well-
articulated ideas about which goals they wish to advocate for and which
program areas need attention, but typically vague ideas about the appro-
priate instrument mixes required to achieve those goals, and seldom any
ideas at all about the more specific settings of those instruments or what
their targets should be. Epistemic communities and technocrats, on the
other hand, possess both principled beliefs and the expertise required to
engage in the cognitive exercises associated with instrument selection
Matt Wilder 31
and policy design. As Dostal (2004) points out, however, the ways in
which normative values and technical expertise impact the policy pro-
cess may be temporally discrete, with the former influencing agenda-
setting and the latter policy formulation.
Figure 2.2 lays out these elements and aspects of policy paradigms
as causal maps. As indicated in Figure 2.2, policy decisions regarding
program areas in need of attention will govern choices having to do
with instrument mixes, which will in turn govern the determination of
instrument settings and instrument targets. The chain of positive cor-
relations moves from instrument settings through to the achievement
of goals with the fulfillment of instrument targets and programmatic
objectives constituting links in the causal chain.
With respect to temporality, the abstract elements of policy are
hypothesized to congeal during agenda-setting. Given their common
ability to engage in normative discourses, advocacy groups, individual
policy entrepreneurs, and epistemic communities all have access to
this stage of the policy process. However, the thoroughness or com-
plexity of the causal maps employed by most advocacy groups is lim-
ited by their lack of technical expertise, oftentimes to the neglect of
clearly articulated ideas about the means-related elements of policy.17
Agenda Setting
Agents: Agenda-oriented AGENDA SETTING
groups.
Goals
Mechanisms: Normative
discourse/principled beliefs. Goals Goals
Change characteristics: universe of

Respond poorly/belatedly to societal Goals
anomalies.
program
area + Goals
Programmatic
instruments objectives Formulation
Agents: Formulation-oriented
Determines + groups (experts).
Yields Mechanisms: Cognitive discourse/

+ + Instrument causal beliefs.
(positive inference) targets
Settings
Change characteristics: Sensitive
to anomalies but respond in
FORMULATION incremental fashion.
Figure 2.2 Normative and technical elements of policy in a causal map

In contrast, epistemic communities possess the cognitive causal under-

standings required to influence processes in the bottom portion of
Figure 2.2 as well as the principled beliefs required to influence agenda-
related developments in the upper portion. What is more, these actors
possess a theoretical or “paradigmatic” inferential understanding of
how abstract and concrete elements are positively or negatively cor-
related. As both Hall (1993) and Haas (1992) remind us, because epis-
temic communities and experts within the state tend to have a highly
contingent (or paradigmatic) understanding of policy, they will be slow
to reconfigure their beliefs, especially in the absence of alternative the-
ories, notwithstanding exogenous pressure from the political executive
(cf. Lindblom, 1959).
That said, recent case studies attempting to disaggregate changes to
specific components of policy have found that change to one compo-
nent in the matrix often occurs independent of other alterations (Baker,
2013). Known as policy conversion and policy drift respectively (Hacker,
2005; Thelen, 2004), means-related changes may occur in the absence
of corresponding shifts to policy aims and, conversely, ends may shift
without any alteration to the means of achieving them (Kern & Howlett,
2009). Considered alongside the phenomenon of paradigmatic layer-
ing discussed earlier, it is clear that forging a policy paradigm is a fluid,
contested, and highly inferential undertaking (Carstensen, 2011a).
Dependence on what are often small-n inferences in many policy fields
explains why it is that policymakers may adjust their views in ways that
are leagues shy of being tantamount to religious or even ideological con-
version (Carstensen, 2011b; Lindblom, 1959). That said, there is noth-
ing immediately damaging about a strong metaphysical slant to social
scientific paradigms, though the degree of this slant opens the doors
for inter-paradigmatic borrowing and, therefore, the violation of Kuhn’s
incommensurability thesis.
Epistemological consequences of bringing

commensurability back in
Responding to the directive given by Masterman (1970)—and shortly

thereafter, Kuhn (1970b)—to shift analytical focus away from paradigms
as sociological constructs and toward paradigms as series of exemplary
achievements has been difficult for policy scholars only insofar as com-
mensurability has been forbidden. Several developments in the phi-
losophy of science that emphasize the linkages and disconnects among
concrete observations, value or theory contingent “observations,” and
Matt Wilder 33
metaphysical aspects may serve to better develop policy scholars’ under-

standing of policy paradigms.18 There are several epistemological conse-
quences associated with taking such a path that ought to be noted, though
it should be stressed that so long as policy paradigms are to stand for any-
thing more than metaphor, it is probably the most viable way to proceed.
The most important consequence, theoretically speaking, of aban-
doning the incommensurability thesis is that policy progression per-
haps ought to be conceived of less in revolutionary terms and more
so as an evolutionary process. Without discarding completely the
themes and parallels discussed by Kuhn, this could be done by apply-
ing some of the concepts developed by Toulmin (1972) to theories of
the policy process, such as field invariability and the relative ability
of ideas to withstand alteration in successive forums of competition.
This approach would be consistent with the methods for understand-
ing processes of policy formulation suggested by Thomas (2001) and
would assist in explaining why a major change often does not occur in
spite of popular expectations (cf. Hajer, 2005). Of course, the influential
Punctuated Equilibrium Theory already takes an evolutionary view of
the policy process, relying on “the politics of attention” to account for
the natural selection of policy ideas (Baumgartner & Jones, 2002; cf.
Gould & Eldredge, 1977).
Perhaps most deserving of attention in the policy literature are the
specific ways in which the components of policy outlined in Figures 2.1
and 2.2 are linked by human participants’ belief structures. To this end,
it is probably best to turn to philosophers like Lakatos (1968), who
argues that scientific theories possess a metaphysical “hard core” that
is surrounded by a permeable and much more observable “protective
belt of auxiliary hypotheses” (cf. Kuhn, 1962, p. 11). What is refutable,
according to Lakatos (1968, p. 183), and only with “a dogged patience,”
are those hypotheses that are part of this protective belt of falsifiable
hypotheses. This view helps to explain why it is that both science and
policy undergo a more fruitful change when actors either direct their
efforts toward “constructively” manipulating auxiliary hypotheses or
use existing policy means to pursue new policy ends.19 While the latter
statement may stretch what Lakatos had in mind, such “drifting” of the
more abstract components of policy is consistent with Laudan’s (1996)
view of scientific progress, whereby significant discursive changes result
not from conflicting empirical observations, but hinge instead on pro-
fessional discord regarding more abstract conceptual issues.
That said, it is important to keep in mind the limitations inherent
to borrowing from the philosophy of science. Policy doctrines are not
spatiotemporal statements as they are in the natural sciences, but are

rather “if-then” statements. Hence, policy, like many social sciences, dif-
fers from the natural sciences in that the purpose of policy is not simply
to observe the natural world, but effectually act upon it. Instrumentation
in this respect differs quite drastically between the two sciences, since
the instruments of the natural scientist are predominantly for measure-
ment purposes whereas the instruments of the policymaker are intended
to act upon the environment. It is only for the sake of measuring the
success of previous policy actions that more positivistic observation, as
emphasized by Kuhn in Postscript—1969, becomes the focus of policy
analysis. Even then, observation is seldom removed from political con-
siderations. As Hall (1990, 1993) pointed out in his discussion of policy
anomalies, the absence of measurement criteria for evaluating societal
problems gives rise to highly contested and highly political episodes of
social learning (see Wilder & Howlett, this volume).
Conclusion
Whether we consider metaphors borrowed from microbiology or seis-

mic geology, few hard assumptions inherent to concepts adapted from
natural sciences hold when speaking of policy. Though the Kuhnian
conception of paradigms has posed as an attractive candidate for cross-
disciplinary adaptation for over two decades, policy paradigms do not
pass the test of incommensurability that Kuhn had stressed as the essen-
tial characteristic of scientific paradigms. Much more akin to pseudo-
scientific paradigms, and therefore prone to all manner of rescue and
“soothsaying,” the notion of policy paradigms has posed both epistemo-
logical challenges and opportunities to policy scholars.20
Though the epistemological flexibility of soft paradigms has no doubt
enhanced the concept’s theoretical usefulness for social scientists, such
an escape from positivism has brought with it considerable conceptual
murkiness that is in need of redress if the notion of social scientific para-
digms is to adhere to any strict definitional meaning. This chapter has
provided a survey of how paradigms have been defined in the literature
and has pointed out a disparity in diligence afforded to the impact and
operationalization of policy paradigms. Policy researchers will be wise to
pay close attention to how they analyze ideas’ influence on policy out-
comes beyond describing operative paradigms as blanket variables that
simply “filter” the effects of other determinants (cf. Berman, 1998; Bleich,
2002; Jacobs, 2009). Though I have advocated for an alternative image
of policy paradigms, this is not to say that a six-part disaggregation is
Matt Wilder 35
the best or only approach to operationalizing paradigmatic variables (see

Daigneault, 2014). That said, routine elaboration on what researchers take
policy paradigms to mean is the most likely avenue toward the develop-
ment of disciplinary standards. Only after such consensus is reached will
the question posed in the title of this piece be definitively answered.
Notes
1 Kuhn (1962, p. 93) contends: “Why should a change of paradigm be called a
revolution? In the face of the vast and essential differences between political
and scientific development, what parallelism can justify the metaphor that
finds revolutions in both? . . . In both political and scientific development
the sense of malfunction that can lead to crisis is prerequisite to revolution.
Furthermore, though it admittedly strains the metaphor, that parallelism
holds not only for the major paradigm changes, like those attributable to
Copernicus or Lavoisier, but also for those far smaller ones associated with the
assimilation of a new sort of phenomenon, like oxygen or X-rays. Scientific
revolutions . . . need seem revolutionary only to those whose paradigms are
affected by them. To outsiders they may, like the Balkan revolutions of the
early twentieth century, seem normal parts of the developmental process.”
2 Hall (1993, p. 291) adheres to the view that paradigms only exist within
mature and regimented policy fields: “Only in some cases, then, will it be
appropriate to speak of a fully elaborated policy paradigm. In others, the web
of ideas affecting the direction of policy will be looser and subject to more
frequent variations.”
3 Masterman (1970, p. 65) points out that metaparadigms are “the only kind
of paradigm to which, to my knowledge, Kuhn’s philosophical critics have
referred.”
4 Italics in original. Kuhn (1962, p. 94) continues in the same passage, “Like
the choice between competing political institutions, that between competing
paradigms proves to be a choice between incompatible modes of community
life. Because it has that character, the choice is not and cannot be determined
merely by the evaluative procedures characteristic of normal science, for these
depend in part upon a particular paradigm, and that paradigm is at issue.
When paradigms enter, as they must, into a debate about paradigm choice,
their role is necessarily circular. Each group uses its own paradigm to argue in
that paradigm’s defense.”
5 Hall (1993, p. 293, n. 21) briefly mentions “fourth order learning” or “deutero-
learning” as learning to learn, but does not explore the possibility any further.
6 Kuhn (1962, pp. 163–4) doubts the conceptual utility of social scientific
paradigms precisely because problems are exogenously created in the social
sciences rather than endogenously created (i.e., problems are interpreted
rather than observed), social scientific problems being defined as problems
that society lacks the tools for solving. Due in large part to the lack of means
of solving social problems, social scientific disciplines are often character-
ized by too many competing causal interpretations for these causal stories to
be called paradigms. In this sense, most social sciences are pre-paradigmatic,
not paradigmatic.
7 Italics in original.
8 This is to say that in social scientific paradigms, observations themselves
are theory contingent, which is no different from instances in the natu-
ral sciences where the metrics governing Popperian “observations” (i.e.,
observations made using theory or paradigm enforcing instruments) are in
dispute (Popper, 1959, p. 85). It is primarily in this respect that the con-
ceptual consistency of paradigms is maintained, whether speaking of social
scientific paradigms or paradigms in the natural sciences.
9 Hall (1993, p. 59) notes that “theories specify the relationships between the
conventional goals of policy and the likely effectiveness of the various instru-
ments used to attain them. Even the statistical observations used to monitor
and align policy are themselves largely defined and generated in terms of this
paradigm.” Hall’s attitude toward empirical observation in the social sciences
is therefore aligned somewhere between those who emphasize the unfalsifi-
ability of “pseudoscientific” theories and Kuhn’s on the structure of scientific
revolutions. That is, Hall concedes that social scientific theories will make
greater and more problematic use of built-in mechanisms to defend against
empirical falsification than theories in the natural sciences. As Popper (1963)
observes, pseudoscientific theories will be particularly susceptible to amend-
ment in the face of developments that are contrary to their premises.
10 On inconsistency and incongruence between means and ends, see Kern and
Howlett (2009).
11 Kordig (1973, p. 560) argues that cross-paradigm transposition of terms and
concepts is actually how science progresses: “I have stressed that experiments
and observations are not ‘completely laden with,’ nor do they presuppose,
the particular theory being tested . . . Having noted all this, however, I also
noted that there is a viable sense for ‘theory-ladenness of observation.’ What
scientists observe (i.e. the corpus of observations) does change; it increases.”
Parentheses in original.
12 Field (1973, p. 479) argues that “if I am right in thinking that denotational
refinement is a fairly common feature of scientific revolutions, that suggests
that future scientists may very well refine many of our current scientific
terms, and hence that many of our current scientific terms are referentially
indeterminate. (In fact, induction from the indeterminacy of terms in earlier
theories may even suggest that science will never reach the stage where all of
its terms are perfectly determinate.)” Parentheses in original.
13 Popper (1978), for example, borrows and adapts Plato’s concept of a “Third
World” in his taxonomy for understanding theory development.
14 Capano (2003) has introduced the notion of hegemonic paradigms and has
thus scratched the surface in terms of providing a basis for a research pro-
gram dedicated to a Kuhnian analysis of puzzle solving under normal science
(or normal policymaking). Lindblom (1959) discusses incremental policy-
making under normal circumstances, but Lindblom’s argument is not explic-
itly in Kuhnian terms.
15 On the dependent variable problem, see Green-Pedersen (2004).
16 Gale and Walter (1973, pp. 419–20) use the notation “*” to “capture the
sense in which a theory ‘hangs together,’ while permitting each individual
element to be held tentatively at the same time.”
Matt Wilder 37
17 For the sake of analytical symmetry, if one conceives of bureaucrats in a

strictly technocratic sense, one could argue that bureaucrats only have well-
articulated ideas about the technical minutiae of their departments’ business,
though this image of the objective bureaucrat has not been sustained empiri-
cally (Irvine, 2011).
18 Popper (1959) describes observations made possible by way of instrumenta-
tion as “observations” (cf. Kordig, 1973; Lakatos, 1968, p. 156, n. 20).
19 Lakatos (1968, p. 183) comments: “It is usually only constructive criticism
which, with the help of rival research programmes, can achieve major
successes, but even so, dramatic, spectacular results become visible only with
hindsight and rational reconstruction.” Italics in original.
20 See Popper’s comments on pseudoscientific “soothsaying” in Popper (1963,
p. 49).
References
Agyris, C., & Schön, D. (1978). Organizational Learning. Reading, PA: Addison-
Wesley.
Allison, G. (1969). Conceptual Models and the Cuban Missile Crisis. The American
Political Science Review, 63(3), 689–718.
Axelrod, R. (1973). Schema Theory: An Information Processing Model of
Perception and Cognition. The American Political Science Review, 67(4),
1248–66.
Baker, A. (2013). The Gradual Transformation? The Incremental Dynamics of
Macroprudential Regulation. Regulation & Governance, 7(4), 417–34.
Baumgartner, F., & Jones, B. (Eds.) (2002). Policy Dynamics. Chicago, IL: University
of Chicago Press.
Béland, D. (2007). Ideas and Institutional Change in Social Security: Conversion,
Layering, and Policy Drift. Social Science Quarterly, 28(1), 20–38.
Béland, D., & Cox, R. (2013). The Politics of Policy Paradigms. Governance, 26(2),
193–5.
Berger, J. (1972). Ways of Seeing. Penguin: London.
Berman, S. (1998). The Social Democratic Moment: Ideas and Politics in the Making of
Interwar Europe. New York: Cambridge University Press.
Bleich, E. (2002). Integrating Ideas into Policy-Making Analysis: Fames and Race
Policies in Britain and France. Comparative Political Studies, 35(9), 1054–76.
Brown, H. (1979). Perception, Theory, and Commitment: The New Philosophy of
Science. Chicago, IL: University of Chicago Press.
Campbell, J. (2002). Ideas, Politics and Public Policy. Annual Review of Sociology,
28(1), 21–38.
Capano, G. (2003). Administrative Traditions and Policy Change: When Policy
Paradigms Matter. The Case of Italian Administrative Reform during the 1990s.
Public Administration, 81(4), 781–801.
Carstensen, M. (2011a). Ideas Are Not as Stable as Political Scientists Want Them to
Be: A Theory of Incremental Ideational Change. Political Studies, 59(3), 596–615.
Carstensen, M. (2011b). Paradigm Man vs. the Bricoleur: Bricolage as an

Review, 3(1), 147–67.
Cashore, B., & Howlett, M. (2007). Punctuating Which Equilibrium?
Understanding Thermostatic Policy Dynamics in Pacific Northwest Forestry.
American Journal of Political Science, 51(3), 532–51.
Coleman, W., Skogstad, G., & Atkinson, M. (1996). Paradigm Shifts and Policy
Networks: Cumulative Change in Agriculture. Journal of Public Policy, 16(3),
273–302.
Crombie, A. (1994). Styles of Scientific Thinking in the European Tradition: A History
of Argument and Explanation in the Mathematical and Biomedical Sciences and the
Arts. London: Duckworth.
Daigneault, P. (2014). Reassessing the Concept of Policy Paradigm: Aligning
Ontology and Methodology in Policy Studies. Journal of European Public Policy,
21(3), 453–69.
Dear, P. (2001). Revolutionizing the Sciences: European Knowledge and Its Ambitions,
1500–1700. Princeton, NJ: Princeton University Press.
Dobuzinskis, L. (1992). Modernist and Postmodernist Metaphors of the Policy
Process: Control and Stability vs. Chaos and Reflexive Understanding. Policy
Sciences, 25(4), 355–80.
Doppelt, G. (1978). Kuhn’s Epistemological Relativism: An Interpretation and
Defense. Inquiry, 21(1–4), 33–86.
Dostal, J. (2004). Campaigning on Expertise: How the OECD Framed EU Welfare
and Labour Market Policies – And Why Success Could Trigger Failure, Journal of
Dowding, K. (1995). Model or Metaphor: A Critical Review of the Policy Network
Approach. Political Studies, 43(1), 136–58.
Eden, C. (1992). On the Nature of Cognitive Maps. Journal of Management Studies,
29(3), 261–6.
Feyerabend, P. (1962). Explanation, Reduction, and Empiricism. In H. Feigl &
G. Maxwell (Eds.), Scientific Explanation, Space, and Time (pp. 28–97). Minneapolis:
University of Minnesota Press.
Field, H. (1973). Theory Change and the Indeterminacy of Reference. The Journal
of Philosophy, 70(14), 462–81.
Gale, G., & Walter, E. (1973). Kordig and the Theory-Ladenness of Observation.
Philosophy of Science, 40(3), 415–32.
Gilardi, F. (2010). Who Learns from What in Policy Diffusion Processes? American
Journal of Political Science, 54(3), 650–66.
Goldstein, J., & Keohane, R. (Eds.) (1993). Ideas and Foreign Policy. Ithaca, NY:
Cornell University Press.
Goodman, N. (1978). Ways of World Making. Cambridge, MA: Hackett.
Gould, S., & Eldredge, N. (1977). Punctuated Equilibria: The Tempo and Mode of
Evolution Reconsidered. Paleobiology, 3(2), 115–51.
Green-Pedersen, C. (2004). The Dependent Variable Problem within the Study of
Welfare State Retrenchment: Defining the Problem and Looking for Solutions.
Haas, P. (1992). Introduction: Epistemic Communities and International Policy
Coordination. International Organization, 46(1), 1–35.
Matt Wilder 39
Hacker, J. (2004). Reform without Change, Change without Reform: The Politics
of US Health Policy Reform in Comparative Perspective. In M. Levin &
M. Shapiro (Eds.), Transatlantic Policymaking in an Age of Austerity: Diversity and
Drift (pp. 13–63). Washington, DC: Georgetown University Press.
Hacker, J. (2005). Policy Drift: The Hidden Politics of US Welfare State
Retrenchment. In W. Streeck & K. Thelen (Eds.), Beyond Continuity: Institutional
Change in Advanced Political Economies (pp. 40–82). Oxford: Oxford University
Press.
Hajer, M. (2005). Setting the Stage: A Dramaturgy of Policy Deliberation.
Administration & Society, 36(6), 624–47.
Hall, P. (1990). Policy Paradigms, Experts and the State: The Case of Macro-
Economic Policy Making in Britain. In S. Brooks & A. Gagnon (Eds.), Social
Scientists, Policy and the State (pp. 53–78). New York, NY: Praeger.
Hall, P. (1993). Policy Paradigms, Social Learning, and the State: The Case of
Economic Policymaking in Britain. Comparative Politics, 25(3), 275–96.
Hall, P., & Taylor, R. (1996). Political Science and the Three New Institutionalisms.
Political Studies, 44(5), 936–57.
Hanson, N. (1958). Patterns of Discovery: An Inquiry into the Conceptual Foundations
of Science. Cambridge: Cambridge University Press.
Hay, C. (2001). The “Crisis” of Keynesianism and the Rise of Neoliberalism in
Britain: An Ideational Institutionalist Approach. In J. Campbell & O. Pedersen
(Eds.), The Rise of Neoliberalism and Institutional Analysis (pp. 193–218).
Heclo, H. (1974). Modern Social Politics in Britain and Sweden. Yale, MA: Yale
University Press.
Hood, C. (1983). The Tools of Government: Public Policy and Politics. London:
Macmillan Press.
Horwich, P. (Ed.) (1993). World Changes: Thomas Kuhn and the Nature of Science.
Cambridge, MA: MIT Press.
Howlett, M. (1994). Policy Paradigms and Policy Change: Lessons from the Old
and New Canadian Policies towards Aboriginal Peoples. Policy Studies Journal,
22(4), 631–49.
Howlett, M. (2009). Governance Modes, Policy Regimes and Operational Plans:
A Multi-Level Nested Model of Policy Instrument Choice and Policy Design.
Policy Sciences, 42(1), 73–89.
Howlett, M., & Cashore, B. (2007). Re-visiting the New Orthodoxy of Policy
Dynamics: The Dependent Variable and Re-aggregation Problems in the Study
of Policy Change. Canadian Political Science Review, 1(2), 50–62.
Howlett, M., & Ramesh, M. (1998). Policy Subsystem Configurations and Policy
Change: Operationalizing the Postpositivist Analysis of the Politics of the
Policy Process. Policy Studies Journal, 26(3), 466–81.
Immergut, E. (1998). The Theoretical Core of the New Institutionalism. Politics
and Society, 26(1), 5–34.
Irvine, S. (2011). Internationalization of Canadian Refugee Policy: Understanding
the Role of International Bureaucratic Networks in Domestic Paradigm Change.
In G. Skogstad (Ed.), Policy Paradigms, Transnationalism and Domestic Politics

(pp. 171–201). Toronto: University of Toronto Press.
Jacob, P., & Jeannerod, M. (2003). Ways of Seeing: The Scope and Limits of Visual
Cognition. Oxford: Oxford University Press.
Jacobs, A. (2009). How Do Ideas Matter? Mental Models and Attention in German
Pension Politics. Comparative Political Studies, 42(2), 252–79.
Jenson, J. (1989). Paradigms and Political Discourse: Protective Legislation in
France and the United States before 1914. Canadian Journal of Political Science,
22(2), 235–58.
Jobert, B., & Muller, P. (1987). L’Etat en action. Politiques publiques et corporatismes.
Paris: Presses Universitaires de France.
Jones, B., Wlezien, C., Green-Pedersen, C., Walgrave, S., Koski, C., & François,
A. (2009). A General Empirical Law of Public Budgets: A Comparative Analysis.
Kay, A. (2007). Tense Layering and Synthetic Policy Paradigms: The Politics of
Health Insurance in Australia. Australian Journal of Political Science, 42(4),
579–91.
Keck, M., & Sikkink, K. (1998). Activists beyond Borders. Ithaca, NY: Cornell
University Press.
Kern, F., & Howlett, M. (2009). Implementing Transition Management as Policy
Reforms: A Case Study of the Dutch Energy Sector. Policy Sciences, 42(4),
391–408.
Koelble, T. (1995). The New Institutionalism in Political Science and Sociology.
Comparative Politics, 27(2), 231–43.
Kordig, C. (1973). Observational Invariance. Philosophy of Science, 40(4), 558–69.
Krasner, S. (1984). Approaches to the State: Alternative Conceptions and Historical
Dynamics. Comparative Politics, 16(2), 223–46.
Kuhn, T. (1962). The Structure of Scientific Revolutions. Chicago, IL: University of
Chicago Press.
Kuhn, T. (1970a). Logic of Discovery or Psychology of Research. In I. Lakatos &
A. Musgrave (Eds.), Criticism and the Growth of Knowledge (pp. 1–23). Cambridge:
Cambridge University Press.
Kuhn, T. (1970b). Postscript—1969. In T. Kuhn (Ed.), The Structure of Scientific
Revolutions (2nd ed.) (pp. 173–208). Chicago, IL: University of Chicago Press.
Kuhn, T. (1982). Commensurability, Comparability, Communicability. In PSA:
Proceedings of the Biennial Meeting of the Philosophy of Science Association, 669–88.
Kuhn, T. (2000). The Road since Structure: Philosophical Essays, 1970–1993. Chicago,
IL: University of Chicago Press.
Lakatos, I. (1968). Criticism and the Methodology of Scientific Research
Programmes. In Proceedings of the Aristotelian Society, 69(1), 149–86.
Laudan, L. (1996). Beyond Positivism and Relativism: Theory, Method, and Evidence.
Boulder, CO: Westview.
Lindblom, C. (1959). The Science of “Muddling through.” Public Administration
Review, 19(2), 79–88.
Mahoney, J. (2000). Path Dependence in Historical Sociology. Theory and Society,
29(4), 507–48.
March, J., & Simon, H. (1958). Organizations. New York: Wiley.
Matt Wilder 41
Masterman, M. (1970). The Nature of a Paradigm. In I. Lakatos & A. Musgrave

(Eds.), Criticism and the Growth of Knowledge (pp. 59–90). Cambridge: Cambridge
University Press.
Merton, R. (1957). Social Theory and Social Structure. London: Free Press.
Nickles, T. (2002). Thomas Kuhn. Cambridge: Cambridge University Press.
Nisbet, R. (1972). Social Change and History: Aspects of the Western Theory of
Development (2nd ed.). London: Oxford University Press.
North, D. (1990). A Transaction Cost Theory of Politics. Journal of Theoretical
Politics, 2(4), 355–67.
Oliver, M., & Pemberton, H. (2004). Learning and Change in 20th-Century
Orenstein, M. (2013). Pension Privatization: Evolution of a Paradigm. Governance,
26(2), 259–81.
Peters, B. (2011). Steering, Rowing, Drifting, or Sinking? Changing Patterns of
Governance. Urban Research & Practice, 4(1), 5–12.
Polanyi, M. (1964). Personal Knowledge: Towards a Post-Critical Philosophy.
New York: Harper & Row.
Popper, K. (1959). The Logic of Scientific Discovery. London: Hutchinson.
Popper, K. (1963). Conjectures and Refutations: The Growth of Scientific Knowledge.
London: Routledge.
Popper, K. (1978, April). Three Worlds. The Tanner Lecture on Human Values.
University of Michigan, Ann Arbor, MI.
Princen, S., & Van Esch, F. (2013). Paradigm Formation and Paradigm Change in
the EU’s Stability and Growth Pact. Presented at the European Consortium for
Political Research General Conference, Bordeaux, France.
Quine, W. (1960). Word and Object. Boston, MA: MIT Press.
Rein, M., & Schön, D. (1996). Frame-Critical Policy Analysis and Frame-Reflective
Policy Practice. Knowledge, Technology & Policy, 9(1), 85–104.
Rorty, R., & Wray, K. (2002). Solidarity or Objectivity. In K. Wray (Ed.), Knowledge
and Inquiry: Readings in Epistemology (pp. 422–37). Toronto: Broadview Press.
Sabatier, P. (1988). An Advocacy Coalition Framework of Policy Change and the
Role of Policy-Oriented Learning Therein. Policy Sciences, 21(2/3), 129–68.
Salamon, L. (2002). The Tools of Government: A Guide to the New Governance.
Sankey, H. (1994). The Incommensurability Thesis. Avebury: Aldershot.
Schmidt, V. (2011). Ideas and Discourses in Transformational Political Economic
Change in Europe. In G. Skogstad (Ed.), Policy Paradigms, Transnationalism and
Domestic Politics (pp. 36–63). Toronto: University of Toronto Press.
Shapin, S. (1996). The Scientific Revolution. Chicago, IL: University of Chicago
Press.
Skogstad, G. (Ed.) (2011). Policy Paradigms, Transnationalism, and Domestic Politics.
Skogstad, G., & Schmidt, V. (2011). Introduction: Policy Paradigms, Transna-
tionalism and Domestic Politics. In G. Skogstad (Ed.), Policy Paradigms,
Transnationalism and Domestic Politics (pp. 3–35). Toronto: University of Toronto
Press.
Steinbruner, J. (1974). The Cybernetic Theory of Decision: New Dimensions of Political

Analysis. Princeton, NJ: Princeton University Press.
Streeck, W., & Thelen, K. (Eds.) (2005). Beyond Continuity: Institutional Change in
Advanced Political Economies. Oxford: Oxford University Press.
Strike, K., & Posner, G. (1992). A Revisionist Theory of Conceptual Change. In
R. Duschl & R. Hamilton (Eds.), Philosophy of Science, Cognitive Psychology, and
Educational Theory and Practice (pp. 147–76). Albany, NY: SUNY Press.
Stone, D. (2004). Transfer Agents and Global Networks in the “Transnationalization”
of Policy, Journal of European Public Policy, 11(3), 545–66.
Surel, Y. (2000). The Role of Cognitive and Normative Frames in Policy-Making.
Thelen, K. (2004). How Institutions Evolve. Cambridge: Cambridge University Press.
Thomas, H. (2001). Towards a New Higher Education Law in Lithuania: Reflections
on the Process of Policy Formulation. Higher Education Policy, 14(3), 213–23.
Toulmin, S. (1972). Human Understanding. Princeton, NJ: Princeton University Press.
Walters, W. (2000). Unemployment and Government: Genealogies of the Social.
Cambridge: Cambridge University Press.
Waver, O. (1996). The Rise and Fall of the Inter-paradigm Debate. In S. Smith,
K. Booth, & M. Zalewski (Eds.), International Theory: Positivism and Beyond
(pp. 149–86). Cambridge: Cambridge University Press.
3
Can You Recognize a Paradigm
When You See One? Defining
and Measuring Paradigm Shift1
Pierre-Marc Daigneault
Introduction
Whether as a subfield of political science or as a discipline in its own

right, policy studies have come a long way since Harold Lasswell’s semi-
nal work on the policy cycle (Howlett & Ramesh, 2003, p. 702; Sabatier,
2007; Savard & Banville, 2012). Nowadays, policy scholars use various
theoretical frameworks to understand policymaking, including advocacy
coalitions (Sabatier & Jenkins-Smith, 1993), multiple streams (Kingdon,
2003), and punctuated equilibrium (Baumgartner & Jones, 2009). To a
significant extent, these frameworks emphasize the influence of “ideas,”
for instance, worldviews, ideologies, cognitive filters, and causal beliefs,
on policy change (Real-Dato, 2009). This should not come as a surprise
since ideas have recently gained ascendency in social research alongside
the “usual suspects” of interests, institutions, and socioeconomic condi-
tions (Béland & Cox, 2011; Jacobs, 2015).
The concept of policy paradigm and associated typology of policy
change proposed by Peter Hall (1993), in particular, is an ideational
framework that has been extensively used by scholars (for a few exam-
ples of recent research, see Kay, 2007; Kern, Kuzemko, & Mitchell, 2014;
Morel, Palier, & Palme, 2012; Orenstein, 2013; Skogstad, 2011). Beyond
its metaphorical appeal, the value of the paradigm framework lies pri-
marily in its ability to illuminate the connection between ideas and
various degrees of policy change (see Baumgartner, 2013; Daigneault,
2014a). Applications of this framework have not gone without theoreti-
cal and operational problems, however.
The purpose of this chapter is threefold. First, the theoretical and
operational problems associated with the concept of policy paradigm
are reviewed in light of the original formulation by Hall (1993) and its
43
subsequent use by policy scholars. Second, an alternative conceptualiza-

tion and a set of methodological guidelines to conduct sound research
on policy paradigm are proposed to overcome these problems. Third,
the value of these guidelines is discussed, taking as a case in point a
recent empirical application in the field of social policy.
The roots of the paradigm concept
Paradigms in scientific inquiry

The concept of paradigm has a long history that can be traced back
to antiquity, from the Greek paradeigma which means “pattern, model;
precedent, example” (Harper, 2001, n.p.). However, it is the famous
philosopher of science Thomas Kuhn (2012 [1962]) who really main-
streamed the concept. While there is some controversy over the inter-
pretation of Kuhn’s work, his use of the concept reflects two different
meanings:
The key sense of “paradigm” is exemplar: an exemplary instance of

puzzle-solving in that discipline that provides a context and a model
for future puzzle-solving. Paradigm is also used to refer to a discipli-
nary matrix, a set of commitments shared by practitioners of a par-
ticular scientific field, including a special vocabulary and established
experimental techniques, as well as accepted theoretical claims. (Bird,
2012, p. 3)
“Normal” science occurs when scientists solve research problems within

a current paradigm. However, “anomalies”—puzzles that cannot be
accounted for within the current paradigm—accumulate and eventually
lead to a crisis. Eventually, a “scientific revolution” may occur as a new
paradigm of inquiry replaces the old one. Kuhn argued that paradigms
are often incommensurable, which implies that scientific research con-
ducted under different research traditions is not cumulative (Bird, 2012).
Incommensurability can also mean that people working within differ-
ent paradigms cannot understand each other’s respective research. This
stronger version of the “incommensurability thesis” is hard to defend
because even people who hold competing worldviews (including clash-
ing principles and values) can often recognize and discuss their differ-
ences. In any case, incommensurability is not a defining characteristic of
a Kuhnian paradigm shift (Bird, 2012, p. 12).
Pierre-Marc Daigneault 45
Paradigms in public policy

Kuhn’s (2012 [1962]) conceptual framework was borrowed by Hall
(1993) to explain patterns of institutional change in British macroeco-
nomic policy. Hall identified three modes of policy changes that are
lexically ordered, meaning that each mode of change builds on and
extends the characteristics of the preceding mode. First order change
is defined by adjustments in the settings or levels of the instruments
used to achieve the goals of policy. Second order change is characterized
by the use of new types of policy instruments, while the hierarchy of
goals behind policy remains stable. Hall draws a parallel between the
incremental mode of policymaking which characterizes first and sec-
ond order change, on the one hand, and normal science in Kuhn’s
account of scientific progress, on the other. Third order change, by con-
trast, is defined by a shift in the goals behind policy which translates
into substantial policy change: “Third order change [. . .] is likely to
reflect a very different process, marked by radical changes in the over-
all terms of policy discourse associated with a ‘paradigm shift’” (Hall,
1993, p. 279). Hall defined the concept of policy paradigm—an “inter-
pretive framework” fundamental to understand third order change—in
the following way:
[P]olicy makers customarily work within a framework of ideas and

instruments that can be used to attain them but also the very nature
framework is embedded through the very terminology through
which policymakers communicate about their work, and is influen-
tial because so much is taken for granted and unamenable to scrutiny
as a whole. (Hall, 1993, p. 279)
Policymaking in Hall’s account proceeds in stages that are closely aligned

with Kuhnian framework of scientific development. When anomalies
arise, that is, policy problems that cannot be solved by actual policy,
policymakers first attempt to solve these by modifying the setting of
instruments (first order change) or by using new instruments (second
order change). If the problems persist, however, the authority of policy-
makers becomes fragmented and there is political contestation by differ-
ent parties over alternative policy paradigms (for a more sophisticated
account, see Oliver & Pemberton, 2004).
Taking stock of the policy paradigm literature:

Advances and challenges
Hall’s (1993) work is now a “classic” that has given rise to a huge lit-
erature (e.g., Berman, 2013; Carson, Burns, & Calvo, 2009; Carstensen,
2011; Howlett & Cashore, 2009; Jenson & Saint-Martin, 2006; Kay,
2007; Skogstad, 2011). Beyond illuminating an important case of policy
change (i.e., British macroeconomic policymaking), Hall’s most impor-
tant contribution with this article has been to recognize the importance
of “ideas” and integrate them into his analysis (Daigneault, 2014a; Kay,
2011). Moreover, the typology he proposed provided a much-needed
bridge between incremental and radical policy change (Daigneault,
2014a; Howlett & Cashore, 2007).
The research conducted in Hall’s footsteps contributed to fine-tuning
the original framework. The most important advance has been to call
into question the dichotomy between “normal” (first and second order
change) and “radical” (third order change) policymaking. Indeed,
many have found that within-paradigm change is more significant
and/or that paradigm shifts can occur in a more gradual fashion than
allowed for by Hall’s framework (Coleman, Skogstad, & Atkinson, 1996;
Greener, 2001; Kay, 2011; Oliver & Pemberton, 2004). Moreover, dif-
ferent policy paradigms have been found to be fully compatible, in
contrast with the incommensurability and “relative commensurabil-
ity” positions defended by some scholars (see, respectively, Princen &
’t Hart, 2014; Wilder, 2014). For instance, Kay (2007) has analyzed the
formation in the 1990s of a synthetic policy paradigm with respect to
health insurance in Australia, namely “universalism plus choice.” It is
important to stress that Hall’s (1993) original account mentioned three
dimensions of policymaking—instrument settings, type of instruments,
and policy goals—but did not specify the constitutive dimensions or
fundamental attributes of the paradigm concept. Other contributions
have served to unpack the dimensions of policy change and/or policy
paradigms (Cashore & Howlett, 2007; Greener, 2001; Kern et al., 2014).
In that regard, four-dimensional conceptualizations of policy paradigm
appear to have been favored by scholars. For instance, Greener (2001,
p. 148) proposed (1) beliefs about cause and effect, (2) desired policy
outcomes, (3) main policy instruments and indicators, and (4) under-
girding ideas. For their part, Kern et al. (2014) proposed another version
composed of (1) ideas about the subject and how it should be governed
(interpretive framework), (2) goals, (3) instruments, and (4) governance
institutions.
Despite these advances, however, scholars interested in policy para-

digms still face a number of challenges. First, the concept of policy
paradigm is undertheorized (Blyth, 1997). Indeed, the nature of policy
paradigms remains elusive, which leads to obvious problems in identi-
fying what precisely changes when there is a paradigm shift (Berman,
2013; Campbell, 2004; Carson et al., 2009; Daigneault, 2014a; Kay, 2011;
Kern et al., 2014; Oliver & Pemberton, 2004; Surel, 2000). In particular,
there seems to be a frequent confusion among scholars between the idea-
tional and policy levels, which could stem, in part, from the way Hall
(1993) discussed third order change as intimately linked with a shift in
policy paradigm. Metaphorically, policy paradigms and public policies
could be compared to “Siamese twins” because it is sometimes difficult to
determine where one precisely begins and another finishes (Daigneault,
2014a). Indeed, while the ideational nature of policy paradigms is rather
non-contentious, we tend to forget that “policies are not only ‘material’
and ‘institutional’—made of resources, activities, laws, official declara-
tions, rules and regulations—but also ideational, in that they minimally
embody a conception of the problem to be solved and goals to be pur-
sued” (Daigneault, 2014a, p. 458). Second, analyses of the influence of
policy paradigms on public policies are not as credible as they could be:
“[A]s historical institutionalists have shown, world-views change, and they
have material consequences. Such explanations have problems. They are
not conceptually clear; the causal mechanism is opaque; and the meth-
odology lacks rigor” (Blyth, 1997, p. 245). Third, there is confusion about
the scale of policy paradigms. Indeed, paradigms can be apprehended at
the macro/societal or meso/policy level of analysis (Kay, 2011). Fourth,
agency is often a neglected aspect in research on ideas and policy para-
digms (Berman, 2013; Blyth, 1997; Campbell, 2004; Carstensen, 2011).
Related to these challenges, the first one in particular, are the prob-
lematic methodological practices of some scholars who tend to equate
paradigm shift with policy change. Implicitly, these scholars adopt a
“revealed ideas” approach where the ideas of policy actors are inferred
from the policy they adopt, rather than directly studied (Daigneault,
2014a).2 Although the distinction between ideas and public policy is
rarely spelled out in such explicit terms, the revealed ideas approach
usually follows the same logic. First, two policy paradigms, a and c, are
discussed in relationship with the type of policy expected under each
paradigm respectively, b and d. Second, the actual policies adopted by
one or more states e are then compared to b and d, that is, the poli-
cies expected under each paradigm. Third, if e has changed from a close
alignment with b to a close alignment with d, then a paradigm shift
from a to c is inferred. This approach has been applied more or less

explicitly by many scholars (e.g., Huo, 2009; White, 2012). For instance,
in an insightful article, White (2012) discussed two social policy para-
digms, namely liberalism and social investment, and then analyzed
whether policies and programs in various liberal welfare states conform
to the social investment paradigm. She concluded that there is no indi-
cation of the social investment paradigm taking hold; rather, the liberal
paradigm is becoming more influential. An advantage of the revealed
ideas approach is its simplicity; public policies are easier to measure than
abstract ideas. Therefore, it might appear sensible for scholars to study
the consequences of ideas rather than the ideas themselves.
Yet, the revealed ideas approach can lead to invalid inferences.
Assuming a direct and perfect correspondence between a policy para-
digm and a set of policies (the a–b and c–d relationships in the previous
discussion) is indeed problematic. On the one hand, a policy paradigm
does not necessarily exist in a given sector, although policy ideas are
always present to a certain extent (Daigneault, 2014b). On the other,
even if a policy paradigm does exist, its translation into public policy is
not automatic. That is so even though policy paradigms are influential—
Baumgartner (2014, p. 476) argued that they are like “ideas on steroids.”
While decision making is clearly influenced by policy ideas, the role
of institutions (i.e., the rules of the game) and realpolitik (i.e., electoral
gains, coalition building) should not be neglected. Therefore, the trans-
lation of ideas into public policies is an empirical question and should
remain so. There is a striking parallel to be made with the interpretivist
critique of rational choice theory: “[I]t is poor social science to con-
struct a model in which observed behaviour maximizes the interests of
agents, and then assume that fit between the interest and the behaviour
explains the behaviour. It may be just a coincidence” (Kay, 2006, p. 66).
Indeed, even though there are clear affinities between certain ideas and
policies, a given public policy can be compatible with different policy
paradigms. Inferring policy ideas from policies is therefore a risky busi-
ness. The following two examples from the social policy literature will
suffice to demonstrate this point:
It has been suggested that continuing agreement in Congress on the

desirability of extending old age insurance stems from liberal desires
to strengthen the welfare programs of the federal government and
from conservative desires to reduce union demands for private pen-
sion plans. If so, this is an excellent demonstration of the ease with
which individuals of different ideologies often can agree on concrete
policy. Labor mediators report a similar phenomenon: the contest-

ants cannot agree on criteria for settling their disputes but can agree
on specific proposals. (Lindblom, 1959, p. 83)
New measures are accepted by a wide range of different groups (politi-
cal parties, administrations, trade unions, employers and others) who
agree on the new measure, but for different reasons and with differ-
ent interests. They share neither a common vision of the reforms
nor the same interests in the measures. During the decision making
process, the measure which is selected from the alternatives is the
one which is able to aggregate different visions and interests. (Palier,
2005, p. 131)
A related limitation of the revealed ideas approach is that it cannot be

used to investigate the influence of ideas on public policy, which is an
important theme for ideational scholars (Baumgartner, 2013; Béland,
2009; Berman, 2013; Boothe, 2013; Campbell, 2004; Parsons, 2007).
Clearly, it makes no sense to investigate whether a paradigm shift has
caused policy change if paradigmatic change is inferred from policy
change in the first place: this is circular reasoning (Daigneault, 2014a).
Given these limitations, should scholars abandon the concept of
policy paradigm altogether? Definitely not; as argued above, the policy
paradigm literature has contributed to our understanding of important
real-world cases by focusing on the hitherto neglected ideational dimen-
sion of policymaking. Rather, the solution proposed here is to define the
concept of policy paradigm in a consistent way, to operationalize it more
precisely, and to put forward a set of methodological guidelines that will
facilitate the conduct of rigorous ideational research. In other words,
there is a need to “align ontology with methodology”3 with respect to
policy paradigms (Daigneault, 2014a).
Aligning ontology and methodology in research

on policy paradigms
Unpacking the concept

I define a policy paradigm as a set of coherent cognitive and normative
ideas intersubjectively held by people in a given policy community
about the nature of reality, social justice and the appropriate role of the
state, the problem that requires public intervention, policy ends and
objectives that should be pursued, and appropriate policy “means” to
achieve those ends (see also Daigneault, 2014a, pp. 461–2).
There are many important elements to this definition. First, at an onto-

logical level, the nature of policy paradigm is ideational. In other words,
paradigms are exclusively “made of” ideas (e.g., values, beliefs, princi-
ples, assumptions), in contrast with policies which possess a “material”
dimension. This may seem obvious but, in light of the frequent confu-
sion between paradigms and policies discussed above, this clarification
is nevertheless essential. Moreover, the ideas of a paradigm are both
cognitive (i.e., they help actors interpret their environment and assign
meaning to their actions) and normative/programmatic (i.e., they pro-
vide a “moral compass” to actors and guide their actions) (Béland, 2005;
Surel, 2000). I therefore argue against restricting the nature of policy
paradigms to cognitive ideas located in the background of the political
debate (see Campbell, 2004). Ideas also have an “affective” dimension
in that they can exert more or less appeal on policy actors (see Cox &
Béland, 2013; Parsons, 2007) and can serve as “discursive weapons” in
the political game (Béland, 2009).
Second, I argue that the ideas which compose a policy paradigm
pertain to four fundamental, constitutive dimensions of the concept:
(1) ideas about the nature of reality, social justice, and the appropriate
role of the State, (2) a conception of the problem that requires pub-
lic intervention, (3) ideas about policy ends and objectives that should
be pursued, and (4) ideas about appropriate policy “means” to achieve
those ends (Daigneault, 2014a). By “fundamental,” I mean that each
dimension is necessary to the concept (Daigneault & Jacob, 2012; Goertz,
2006; Sartori, 2009 [1984]). Admittedly, this fourfold conceptualiza-
tion is similar to the ones developed by Greener (2001) and Kern et al.
(2014). At the same time, however, the conceptualization proposed here
has the advantage of clearly outlining the policy problem dimension of
the concept which is only partly addressed in Greener (2001, see “beliefs
about cause and effect”) and absent from Kern et al. (2014). Contrary
to Kern et al. (2014), I argue that governance institutions should not
be considered a fundamental dimension of the concept. In fact, gov-
ernance institutions are covered in other dimensions of my framework,
in particular the fourth one (i.e., ideas about appropriate policy means
to achieve policy ends). Moreover, while comprehensive, the conceptu-
alization proposed here remains relatively simple compared to others
(e.g., Carson et al., 2009).
Third, the ideas that compose a policy paradigm must be internally
coherent, an attribute which is emphasized in the literature (e.g.,
Baumgartner, 2014; Carson et al., 2009; Hall, 1993; Kay, 2007) even
though there are disagreements as to its defining character (Skogstad &
Schmidt, 2011). However, it is important to note that Hall (1993) devel-

oped the policy paradigm concept in the context of macroeconomic
policymaking which is guided by a relatively coherent body of ideas.
What level of coherence is required for ideas to acquire paradigmatic sta-
tus? Minimally, the ideational content of all dimensions is compatible
and logically consistent (Daigneault, 2014a). In the absence of this con-
dition, paradigms do not exist and are only a loose collection of ideas.
To use an example from the field of social assistance (welfare), there
would be a clear inconsistency between “welfare dependency” (i.e., the
policy problem to be solved) and ideas favoring the use of generous and
unconditional cash transfers to welfare clients (i.e., ideas about appro-
priate policy means). Moreover, because individual ideas do not evolve
in a vacuum (Carstensen, 2015; Kay, 2009), the ideational content of the
four dimensions of policy paradigms is certainly related in other ways
beyond internal coherence. For instance, the more general and funda-
mental ideas about the nature of reality and the role of the state could
influence the ideas of the intermediate dimensions of policy problem
and ends which, in turn, influence the more “ground level” dimension
of ideas about policy instruments. Therefore, more research on the inter-
relatedness of the four dimensions is required.
Fourth, in order to possess paradigmatic status, ideas must be widely
shared or at least shared by a significant number of actors in a given pol-
icy community (Baumgartner, 2014; Daigneault, 2014a). Admittedly, it
might be difficult to establish a precise threshold at which ideas are held
by a sufficient number of actors to acquire paradigmatic status. Despite
this operational challenge, paradigms are intersubjective ideas—not
merely personal or private ideas. This brings back the level-of-analysis
issue: Are policy paradigms located at a meso/policy or macro/societal
level (Kay, 2011)? In other words, are policy paradigms restricted to the
mind of a few key policy actors or are they held more largely by persons
outside the “inner ring”—people in academia, media, business, and civil
society? I argue that both options are possible provided that a significant
number of people in the policy community share the same set of coher-
ent ideas directed at what the government should do, how, and why.
Empirical research on policy paradigms

The clear definition and operationalization of the concept of policy
paradigm presented in this chapter should contribute to improving the
rigor of research on ideas and policy change. A better conceptualiza-
tion is only a first step toward this goal, however, as the methodological
practices of scholars must also conform to the highest standards of rigor.
Therefore, students of policy paradigms should primarily rely on direct

evidence of the ideas of policy actors, in contrast with the indirect evi-
dence as generated by the revealed ideas approach (Daigneault, 2014a).
Admittedly, ideas cannot be studied “directly”: for technical and ethical
reasons, we simply cannot open the heads of policy actors and examine
their brains to see what they really think! The challenges associated with
measuring policy ideas should not be exaggerated, however (Daigneault,
2014a). As Kay (2009, p. 49) argues, “the measurement of beliefs is, of
course, notoriously difficult, but in policy studies we have developed a
range of qualitative research methods from elite interviewing through
to large scale attitudinal social surveys that can contribute to this post-
positivist ambition.” Scholars can rely on accounts of these ideas as
expressed in official documents (e.g., Hansard, policy statements, dis-
cussion papers, press releases) and other documents (e.g., memoirs,
scholarly literature), during research interviews with the policy actors
or their entourage, and in “real-life” settings (observation of interac-
tions of policy actors). Scholars should nevertheless be cautious with
respect to the ideas expressed by policy actors because the latter can
lie, have memory lapses, disclose information selectively, or reinterpret
their beliefs and actions under a positive light (Campbell, 2004; Jacobs,
2015). For that reason, scholars should use various techniques to ensure
the credibility of findings, including triangulation of sources and meth-
ods. Moreover, scholars must ensure that their data is sufficiently rich
(1) to infer the ideas of policy actors with respect to the four dimensions
of the paradigm concept and (2) to determine whether these ideas are
widely shared within a policy community.
Once a given policy paradigm has been noted to characterize a given
policy community, how can scholars recognize whether a paradigm
shift did occur? I hold that a paradigm shift occurs when there is sig-
nificant change on all four dimensions of the concept between two time
periods (Daigneault, 2014a; see also Kern et al., 2014). Of course, the
other conditions necessary to the existence of policy paradigms—for
instance, the internal coherence of ideas and their prevalence within a
policy community—must also be satisfied. Assessing the significance of
ideational change is a matter of interpretation but I consider that “sub-
stantial departure” from a given policy equilibrium is a good indicator
of this (Cashore & Howlett, 2007). However, ideational change can be
significant yet gradual and cumulative (Carson et al., 2009; Coleman
et al., 1996; Greener, 2001; Kay, 2011; Palier, 2005).
Beyond descriptive studies, many scholars are interested in analyz-
ing the influence exerted by ideas—including policy paradigms—on
policy dynamics. The guidelines presented so far have focused on how

to improve the measurement of policy paradigms—the “independent
variable” in many studies of policy change. Yet, specifying the “depend-
ent variable” (Howlett & Cashore, 2009)—policy change—can also be a
serious challenge known as the problem of the explanandum (Real-Dato,
2009). Because public policies have an ideational dimension, a “policy
theory” (Leeuw, 2003; Rossi, Lipsey, & Freeman, 2004), it may be delicate
to assess the influence of ideas on policy. For that reason, I argue that
ideational scholars should focus on the “material” dimension of public
policy (Daigneault, 2014a). This pragmatic guideline is consistent with
Real-Dato’s (2009, p. 122) suggestion to focus on “policy designs – defined
as ‘observable phenomena found in statutes, administrative guidelines,
court decrees, programs, and even the practices and procedures of street
level case workers as they interact with policy recipients’ (Schneider &
Ingram, 1997, p. 2).” Moreover, studying the impact of policy paradigms
on public policy entails identifying the causal mechanisms which medi-
ate this impact (Berman, 2013; Blyth, 1997; Campbell, 2004). Process trac-
ing is a method particularly apt to this task (Campbell, 2004; George &
Bennett, 2004; Jacobs, 2015). Ideas can influence policymaking in at
least three ways:
First, such processes help to construct the problems and issues that
enter the policy agenda. Second, ideational processes shape the
assumptions that impact the content of reform proposals. Third,
these processes can become discursive weapons that participate in
the construction of reform imperatives. (Béland, 2009, p. 702)
In all these cases, however, ideas influence policymaking through a

“logic of interpretation” which lead actors to make sense of “what is
possible and/or desirable” (Parsons, 2007, p. 13).
When the rubber meets the road: An application
The conceptual and methodological guidelines presented above are

expected to contribute to improve the rigor of ideational research. Now,
how does this conceptual framework fare in practice? An application
within a research program on ideas and policy change with respect to
Canadian social assistance (welfare) (Daigneault, 2014c, 2015) generated
lessons that could serve to improve research on policy ideas. It is beyond
the scope of this chapter to provide a full analysis of this case—this vol-
ume already contains five interesting cases which are analyzed in detail.
However, a cursory presentation of this case is necessary to understand

the significance of the lessons learned.
The empirical case

In 1998, the government of Saskatchewan embarked on a substantial
reform of its social policy regime, namely “Building Independence:
Investing in Families” (Saskatchewan. Government of Saskatchewan,
January 1996, March 1997, n.d.). This reform, which targeted low-
income families, sought to tear down the “welfare wall,” fight welfare
dependency, and reduce child poverty (Daigneault, 2015). In terms
of programs, Building Independence introduced relatively generous
income supplements and health benefits for low-income families who
are active on the labor market, as well as a training allowance for low-
income people enrolled in basic education and related courses. This
case of policy change was selected because it was a major, coherent, and
high-profile reform. Moreover, Building Independence was character-
ized by a tightened relationship between social assistance and the labor
market, which could, in turn, indicate a paradigm shift.
The application
As a first step to empirical research, I conducted a selective review of
the social assistance literature. Based on this review and the fourfold
operationalization of the concept of policy paradigm I had proposed
(Daigneault, 2014a), I then developed a typology of social assistance.
Three ideal types, namely the entitlement, workfare, and activation para-
digms, were discussed (Daigneault, 2014c). This deductive framework
was used to examine the actual policy ideas associated with the Building
Independence reform. Data came from qualitative interviews with a
sample of 14 policy actors who were closely involved with the initia-
tive at the time (politicians, public servants, and advocates). Interview
data were analyzed thematically using the NVivo 8 software (QSR
International Pty Ltd, 2008). Codes were first used to organized data
and nine themes were then constructed from these codes. The themes,
which are largely deductive, were classified under one of the four dimen-
sions of the policy paradigm concept outlined above (Daigneault, 2015).
I then assessed the level of alignment between these nine themes and
each ideal type of social assistance paradigm. I found that the policy
ideas informing Building Independence are coherent and align closely
with the activation paradigm, but also share some similarities with the
entitlement and workfare paradigms.
In a different but related (unpublished) study, I explored the extent

to which the ideas associated with Building Independence differed from
the ideas held during the 1980s. I hypothesized that a paradigm shift
had taken place with respect to social assistance between the two peri-
ods. Data came from the interviews presented above and secondary
sources. First, the ideas held by policy actors during the 1980s were iden-
tified and compared to those held under the 1998 reform. Second, inter-
view respondents were asked to assess, qualitatively and quantitatively,
the extent to which the ideas associated with Building Independence
represented a marked shift from previous ideas. Although the evidence
was generally supportive of the paradigm shift hypothesis, it was too
tentative and limited in scope to allow for generating sound inferences
on the nature of ideational change.
The lessons learned

I learned many lessons while conducting this research program on ideas
and welfare reform. The first is that a deductive approach, in which
policy paradigms are developed before data is collected, was fruitful.
When the theoretical literature is abundant in a given policy sector (as is
the case with social assistance), building ideal typical policy paradigms
indeed facilitates the analysis and interpretation of data. The only pit-
fall within this approach is that researchers may be tempted “to fit a
square peg into a round hole,” that is, to force the assignment of ideas
into predetermined categories. Various techniques can be used to coun-
teract this natural tendency, including developing intimate knowledge
of one’s data, conducting analysis iteratively and analyzing disconfirm-
ing evidence. Moreover, I found that analyzing ideational alignment on
an ordinal scale using various levels of shading was useful to nuance
the analysis (see Daigneault, 2015). As an alternative to the deductive
approach to the measurement of policy paradigm discussed here, schol-
ars could use an inductive approach according to which the ideas that
compose a paradigm emerge from the data.
A second lesson is that measuring policy paradigms is a challeng-
ing and labor-intensive endeavor. On the one hand, policy paradigms
are composed of numerous ideas (i.e., at least one idea per dimension)
which must be coherent. Rich data is required to bring to light complex
configurations of ideas. On the other hand, to have paradigmatic sta-
tus, ideas must be widely shared in the policy community. While semi-
structured interviews are the perfect research method to analyze ideas
in “depth,” they provide less “breadth.” Indeed, qualitative methods are
disadvantaged when the goal is to assess ideational prevalence within a

given policy community. Of course, a researcher or research team can
always conduct a high number of interviews, but this may be unfeasible
in a context where time and/or financial resources are limited. In particu-
lar, assessing claims of paradigm shift are particularly resource-intensive
because they involve measuring ideas at two different times, ideally fol-
lowing a longitudinal design. A less demanding method is to rely on a
retrospective design. In the unpublished study described above, respond-
ents were asked about the policy ideas held during two different periods,
namely at the time of reform in 1998 and during the 1980s. The magni-
tude of change between two periods then needs to be assessed according
to a logic that is quite similar to the “retrospective pretest” used in quasi-
experimental methods where human subjects studied at t2 are asked to
report their condition at t1. While this is a promising method, scholars
should ensure that they have sufficient data to derive valid inferences on
the nature of ideational change which, unfortunately, was not my case.
Third, determining whether policy ideas are internally coherent and
prevalent in a particular case is difficult. While these research decisions
must be based on data, they will inevitably involve judgment calls and
interpretations. Researchers must therefore be as transparent as pos-
sible as to the procedures they use to produce their inferences (King,
Keohane, & Verba, 1994).
The fourth and last lesson I learned is that while it may be a challenge
to get access to some policy actors, the real difficulty lies in getting access
to the values, beliefs, and assumptions they hold. Indeed, policy actors,
as mere human beings, may suffer from memory lapses, or can have cog-
nitive biases that lead them to a distorted perception of their own ideas.
Even worse, policy actors may be tempted to lie, to disclose selectively
what they know, or to distort reality to picture them under a positive
light (Campbell, 2004; Jacobs, 2015). I must admit that the policy actors
I interviewed seemed for the most part collaborative and transparent
about their ideas and actions. However, there were times when I felt that
some information was held back from me and/or I was being told the
“official line.” Many techniques can be used to mitigate these problems,
including triangulating data sources and methods and using carefully
designed probes during interviews (Rubin & Rubin, 2011). Respondent
selection is crucial as is, in particular, striking a balance between “insid-
ers,” who have intimate knowledge of policy ideas but presumably a
vested interest in defending them, and “outsiders,” who, while reason-
ably knowledgeable about the case at hand, are more prone to examine
ideas under a different, including critical, light.
Conclusion
The concept of policy paradigm developed by Hall (1993) has given rise
to an insightful literature on ideas and policy change. However, in its
original formulation, the concept was underspecified and its use within
empirical research has not always met the highest methodological
standards. In particular, in this chapter, I have taken issue with a per-
sistent and problematic confusion between ideas and policies. I sought
to overcome these limitations by suggesting guidelines relative to the
conceptualization, operationalization, and measurement of policy para-
digms. I also discussed the lessons I learned from applying these guide-
lines to a concrete case of policy change. Beyond these guidelines, the
key message is that scholars should be very rigorous with respect to how
they define concepts and how they apply them in empirical research.
Notes
1 I wish to thank the editors, Michael Howlett and John Hogan, for their kind
invitation to contribute to this volume and for their comments on this chap-
ter. I also thank Daniel Béland for his comments on various versions of this
work. This chapter is based on and extends my previous work on policy para-
digms, which was mostly conducted while I was a postdoctoral fellow at the
Ministère de l’Emploi et de la Solidarité sociale du Québec and at the Johnson-
Shoyama Graduate School of Public Policy (JSGS), University of Saskatchewan
Campus. Postdoctoral funding from the Fonds de recherche du Québec—
Société et culture (FRQ-SC) and the Canada Research Chair on Public Policy at
JSGS is gratefully acknowledged for this period.
2 The expression ‘revealed ideas’ is derived from the ‘revealed preferences’
approach in economics, where the preferences of economic agents are inferred
from their behavior.
3 This expression is borrowed from Hall (2003) who discussed the practices of
comparative scholars.
References
Baumgartner, F. R. (2013). Ideas and Policy Change. Governance, 26(2), 239–58.
Baumgartner, F. R. (2014). Ideas, Paradigms, and Confusions. Journal of European
Baumgartner, F. R., & Jones, B. D. (2009). Agenda and Instability in American Politics
(2nd ed.). Chicago: University of Chicago Press.
Béland, D. (2005). Ideas and Social Policy: An Institutionalist Perspective. Social
Policy & Administration, 39(1), 1–18.
Béland, D., & Cox, R. H. (Eds.). (2011). Ideas and Politics in Social Science Research.
New York and Oxford: Oxford University Press.
Berman, S. (2013). Ideational Theorizing in the Social Sciences since “Policy

Paradigms, Social Learning, and the State.” Governance, 26(2), 217–37.
Bird, A. (2012). The Structure of Scientific Revolutions and Its Significance: An
Essay Review of the Fiftieth Anniversary Edition. The British Journal for the
Philosophy of Science, 63(4), 859–83.
Blyth, M. M. (1997). “Any More Bright Ideas?” The Ideational Turn of Comparative
Political Economy. Comparative Politics, 29(2), 229–50.
Boothe, K. (2013). Ideas and the Limits on Program Expansion: The Failure of
Nationwide Pharmacare in Canada since 1944. Canadian Journal of Political
Science/Revue canadienne de science politique, 46(2), 419–53.
Campbell, J. L. (2004). Institutional Change and Globalization. Princeton, NJ:
Princeton University Press.
Carson, M., Burns, T. R., & Calvo, D. (Eds.). (2009). Paradigms in Public Policy:
Theory and Practice of Paradigm Shifts in the EU. Frankfurt: Peter Lang GmbH.
Carstensen, M. B. (2011). Paradigm Man vs. the Bricoleur: Bricolage as an
Review, 3(1), 147–67.
Carstensen, M. B. (2015). Conceptualising Ideational Novelty: A Relational
Approach. The British Journal of Politics & International Relations, 17(2), 284–97.
Cashore, B., & Howlett, M. (2007). Punctuating Which Equilibrium? Under-
standing Thermostatic Policy Dynamics in Pacific Northwest Forestry. American
Coleman, W. D., Skogstad, G. D., & Atkinson, M. M. (1996). Paradigm Shifts and
Policy Networks: Cumulative Change in Agriculture. Journal of Public Policy,
16(03), 273–301.
Cox, R. H., & Béland, D. (2013). Valence, Policy Ideas, and the Rise of Sustainability.
Governance, 26(2), 307–28.
Daigneault, P.-M. (2014a). Reassessing the Concept of Policy Paradigm: Aligning
21(3), 453–69.
Daigneault, P.-M. (2014b). Puzzling about Policy Paradigms: Precision and
Progress. Journal of European Public Policy, 21(3), 481–4.
Daigneault, P.-M. (2014c). Three Paradigms of Social Assistance. Sage Open, 4(4),
1–8. doi: 10.1177/2158244014559020
Daigneault, P.-M. (2015). Ideas and Welfare Reform in Saskatchewan: Entitlement,
Workfare or Activation? Canadian Journal of Political Science/Revue canadienne de
science politique. FirstView [article published online before print], 1–25. doi:
10.1017/S0008423915000098
Daigneault, P.-M., & Jacob, S. (2012). Les concepts souffrent-ils de négligence
bénigne en sciences sociales? Eléments d’analyse conceptuelle et examen
exploratoire de la littérature francophone à caractère méthodologique. Social
Science Information/Information sur les sciences sociales, 51(2), 188–204.
George, A. L., & Bennett, A. (2004). Case Studies and Theory Development in the
Social Sciences. Cambridge, MA: MIT Press.
Goertz, G. (2006). Social Science Concepts: A User’s Guide. Princeton, NJ: Princeton
University Press.
Greener, I. (2001). Social Learning and Macroeconomic Policy in Britain. Journal
of Public Policy, 21(02), 133–52. doi: 10.1017/S0143814X01001076
Hall, P. A. (2003). Aligning Ontology and Methodology in Comparative Research.
In J. Mahoney & D. Rueschemeyer (Eds.), Comparative Historical Analysis in the
Social Sciences (pp. 373–404). New York: Cambridge University Press.
Harper, D. (2001). Paradigm. Online Etymology Dictionary. Retrieved from http://
www.etymonline.com/index.php?term=paradigm
Dynamics: The Dependent Variable and Re-aggregation Problem in the Study
Subsystems (2nd ed.). Don Mills, ON: Oxford University Press.
Huo, J. (2009). Third Way Reforms: Social Democracy After the Golden Age. New York:
Jacobs, A. M. (2015). Process Tracing the Effects of Ideas. In A. Bennett & J. T.
Checkel (Eds.), Process Tracing: From Metaphor to Analytic Tool (pp. 41–73).
Cambridge: Cambridge University Press.
Jenson, J., & Saint-Martin, D. (2006). Building Blocks for a New Social Architecture:
The LEGO Paradigm of an Active Society. Policy & Politics, 34(3), 429–51.
Kay, A. (2006). The Dynamics of Public Policy: Theory and Evidence. Cheltenham,
UK: Edward Elgar.
Kay, A. (2007). Tense Layering and Synthetic Policy Paradigms: The Politics of Health
Insurance in Australia. Australian Journal of Political Science, 42(4), 579–91.
Kay, A. (2009). Understanding Policy Change as a Hermeneutic Problem. Journal
of Comparative Policy Analysis: Research and Practice, 11(1), 47–63.
Kay, A. (2011). UK Monetary Policy Change during the Financial Crisis: Paradigms,
Spillovers, and Goal Co-ordination. Journal of Public Policy, 31(2), 143–61.
Kern, F., Kuzemko, C., & Mitchell, C. (2014). Measuring and Explaining Policy
Paradigm Change: The Case of UK Energy Policy. Policy & Politics, 42(4), 513–30.
King, G., Keohane, R. O., & Verba, S. (1994). Designing Social Inquiry: Scientific
Inference in Qualitative Research. Princeton, NJ: Princeton University Press.
Kingdon, J. W. (2003). Agendas, Alternatives, and Public Policies (2nd ed.). New
York: Longman.
Kuhn, T. S. (2012 [1962]). The Structure of Scientific Revolutions (4th ed.). Chicago,
Leeuw, F. L. (2003). Reconstructing Program Theories: Methods Available and
Problems to be Solved. American Journal of Evaluation, 24(1), 5–20.
Lindblom, C. E. (1959). The Science of “Muddling Through.” Public Administration
Review, 19(2), 79–88.
Morel, N., Palier, B., & Palme, J. (Eds.). (2012). Towards a Social Investment Welfare
State? Ideas, Policies and Change. Bristol: Policy Press.
Orenstein, M. A. (2013). Pension Privatization: Evolution of a Paradigm.
Governance, 26(2), 259–81.
Palier, B. (2005). Ambiguous Agreement, Cumulative Change: French Social

Policy in the 1990s. In W. Streeck & K. Thelen (Eds.), Beyond Continuity:
University Press.
Parsons, C. (2007). How to Map Arguments in Political Science. Oxford and New
York: Oxford University Press.
QSR International Pty Ltd. (2008). NVivo Qualitative Data Analysis Software
(Version 8).
Real-Dato, J. (2009). Mechanisms of Policy Change: A Proposal for a Synthetic
Explanatory Framework. Journal of Comparative Policy Analysis: Research and
Practice, 11(1), 117–43.
Rossi, P. H., Lipsey, M. W., & Freeman, H. E. (2004). Evaluation: A Systematic
Approach (7th ed.). Thousand Oaks, CA: Sage.
Rubin, H. J., & Rubin, I. S. (2011). Qualitative Interviewing: The Art of Hearing Data
(3rd ed.). Thousand Oaks, CA: Sage.
Sabatier, P. A. (Ed.). (2007). Theories of the Policy Process (2nd ed.). Boulder, CO:
Westview Press.
Sabatier, P. A., & Jenkins-Smith, H. C. (1993). Policy Change and Learning: An
Sartori, G. (2009 [1984]). Guidelines for Concept Analysis. In D. Collier &
J. Gerring (Eds.), Concepts and Method in Social Science: The Tradition of Giovanni
Sartori (pp. 97–150). London: Routledge.
Saskatchewan. Government of Saskatchewan. (January 1996). Redesigning Social
Assistance: Preparing for the New Century—Discussion Paper. Regina, SK:
Department of Social Services.
Saskatchewan. Government of Saskatchewan. (March 1997). Children, Families
and Independence: Social Assistance Redesign—Progress Report. Regina, SK:
Government of Saskatchewan.
Saskatchewan. Government of Saskatchewan. (n.d.). Building Independence:
Investing in Families. Regina, SK: Government of Saskatchewan.
Savard, J.-F., & Banville, R. (2012). Policy Cycles. In L. Côté & J.-F. Savard (Eds.),
Encyclopedic Dictionary of Public Administration. Québec. Retrieved from www.
dictionnaire.enap.ca
Skogstad, G. (Ed.). (2011). Policy Paradigms, Transnationalism, and Domestic Politics.
Skogstad, G., & Schmidt, V. (2011). Introduction: Policy Paradigms, Transnationalism,
and Domestic Politics. In G. Skogstad (Ed.), Policy Paradigms, Transnationalism,
and Domestic Politics (pp. 3–35). Toronto: University of Toronto Press.
White, L. A. (2012). Must We All Be Paradigmatic? Social Investment Policies and
Liberal Welfare States. Canadian Journal of Political Science/Revue canadienne de
science politique, 45(3), 657–83.
Wilder, M. (2014). Ideas Beyond Paradigms: Relative Commensurability and the
Case of Canadian Trade-Industrial Policy, 1975–95. Journal of European Public
Policy, 22(7), 1004–21.
4
Is There a Fourth Institutionalism?
Ideas, Institutions and the
Explanation of Policy Change
Jeremy Rayner
Introduction
While the concepts of policy paradigms and paradigmatic change now

range freely through the policy literature, one of the most influential
contributions to the currency of the paradigm and related concepts is
that of Hall (1993). This is not to say that Hall invented the idea of
transferring Kuhn’s (1970) original ideas from the sociology of science
to politics and policy. Scattered references are found much earlier, for
example in Manning (1976, pp. 26–7), where the succession of domi-
nant ideologies is treated as an example of paradigm shifts. Nonetheless,
for modern policy studies, Hall (1993) is the canonical reference. This
chapter explores the connection between the contemporary debates
about policy paradigms and the tradition in which Hall (1993) was writ-
ten, namely neo-institutionalism. It is argued that much of the value
of the concept, together with some of the unresolved problems and
ambiguities surrounding its use, can be traced to its origins in the new
institutionalism and the debates around variant “institutionalisms”. By
returning to these debates, we get a better picture of both the strengths
and the weaknesses of the idea of policy paradigm.
In the first section, I briefly review the emergence of the new insti-
tutionalism in political science and policy studies and the differences
of approach among the three main variants: rational, historical and
sociological institutionalism. The next section analyses the role of the
paradigm idea in Hall (1993) in detail, paying particular attention to
the institutionalist context of his argument. The penultimate section
traces the career of paradigms and paradigm shifts in three versions of a
“fourth institutionalism”, where each version is proposed as a remedy to
an alleged failing on the part of the three older variants with respect to
61
the role of ideas in policy change. The conclusion argues that whether
there really is a need for a fourth new institutionalism is much less
important than how variant institutionalisms treat policy paradigms
and the closely related idea of paradigmatic change in public policies.
The new institutionalisms
The advent of a new institutionalism – to be distinguished from a much

older focus on the central institutions of government that character-
ized early twentieth-century political science – was given the status of
a distinctive school by March and Olsen (1984, 1989). When the latter
launched their now-classic attack1 on the narrow theoretical vision that
excluded organizational factors from political life, their target was the
behaviourist orthodoxy that had arisen at mid-century, primarily in the
United States. The orthodoxy, they argued, was contextual, in the sense of
embedding politics in society and reducing political life to the outcome
of social forces, (what has been called epiphenomenalism); reductionist,
in seeing politics as the aggregation of individual decisions; utilitarian,
in ascribing calculated self-interest to the agents making these decisions;
functionalist, in believing that history tends towards equilibrium and
balance; and instrumentalist, in holding that decisions about the alloca-
tion of resources, rather than decisions about the allocation of meaning,
are the stuff of politics (March & Olsen, 1984).
In programmatic fashion, March and Olsen called for a “new institu-
tionalism” that will:
deemphasize the dependence of the polity on society in favor of an

interdependence between relatively autonomous social and political
institutions; deemphasize the simple primacy of micro processes and
efficient histories in favor of relatively complex processes and histori-
cal inefficiency; deemphasize metaphors of choice and allocative out-
comes in favor of other logics of action and the centrality of meaning
and symbolic allocation. (1984, p. 738)
For understanding policy, the most important argument derived from

this programme turned out to be the seemingly innocuous observa-
tion that underlying non-institutional theories is the assumption that
“things are ordered by their consequential connections”, in the sense of
means linked to ends, causes to effects, solutions to problems and so on.
Institutionalism, on the other hand, alerts us to the possibility of other
kinds of ordering, particularly the importance of temporal ordering:
Jeremy Rayner 63
“things are connected by virtue of their simultaneous presence or arrival”

(March & Olsen, 1984, p. 743). Once the connection between problem
and solution has been put into question – the latter is not the simple
consequence of some characteristics of the former, for example, problem
structure or actor interests – the way is open to investigate the connec-
tion as an empirical rather than a logically necessary or “natural” one.
However, both the claim that the novelty of the new institutionalism
is a matter of emphasis or degree and the lengthy shopping list of poten-
tial challenges to the behaviourist orthodoxy militated against a sin-
gle identity for neo-institutionalism. The new institutionalism rapidly
fractured into three main schools, each of which takes up and empha-
sizes one particular challenge, largely at the expense of the others.
Thus, historical institutionalism (HI) has followed March and Olsen’s
(1984) suggestion about de-emphasizing micro processes and focused on
explaining complex processes and inefficient histories. From this little
hint would spring the historical institutionalists’ interest (critics might
say “obsession”) with the macro processes of path dependency, “lock-in”
and the inefficient histories that constrain possible futures (Pierson,
2000). The small steps that might allow the construction of a narra-
tive account of path dependency in terms of actors negotiating a world
structured by institutions are almost always subsumed under the larger
rubric of the constraints imposed by the original path and the various
positive feedback mechanisms that keep actors locked in.
Both rational choice institutionalism (RI) and sociological institu-
tionalism (SI) can be understood as efforts to restore some agency to
HI-inspired accounts that had certainly managed to impose order on
an inchoate world by appealing to the structuring role of institutions,
but were now in danger of reducing outcomes to the logic of path
dependency. Both RI and SI respond to the original request for sensitiv-
ity to the “interdependence” of social and political institutions and for
“de-emphasis” rather than a complete rejection of micro processes.
RI, however, notably disregards the point that a genuine institutional-
ism will also de-emphasize the logic of choice and allocation, while SI
enthusiastically embraces the suggestion of alternative logics of action
and the importance of the symbolic allocation of meaning.
Much has already been written on the relative merits of the three
institutionalist traditions that have emerged from this common start-
ing point (Campbell, 2004; Hall & Taylor, 1996; Rockman, Rhodes, &
Binder, 2008; Schmidt, 2010). However, the claim that all three need
to take the role of discourse and discursive practices more seriously is
intended as a response to a distinctive problem of all neo-institutionalist
explanations noted above. By stressing that institutions structure social

and political life, institutionalism has a natural tendency to focus on the
explanation of continuity rather than change (Schmidt, 2008a). Change
is the exception rather than the rule and institutionalist explanations
of change, most clearly HI explanations, appeal in such cases to exog-
enous factors that work by causing a temporary failure in institutions
and “business as usual”. The need to appeal to two kinds of explanation,
one for continuity and another for change, violates a rule of theoretical
parsimony. The concepts of a policy paradigm and a paradigm shift are
very often intended to tackle this problem. To decide whether they do
so successfully requires a journey back through the neo-institutionalists
debates on the relationship between ideas and institutions, a journey
that leads us ultimately to Peter Hall’s famous contribution (Hall, 1993).
Peter Hall’s founding contribution
Of all the contending schools of thought in contemporary policy stud-

ies, the “new” institutionalism is most closely associated with the origi-
nal concept of a policy paradigm. Many of the characteristic strengths
and weaknesses of the policy paradigm idea flow both from its connec-
tion with the work of Hall (1993) and from the broader context of a
revived interest in the role of institutions in political life in which his
work is located. This point of departure gives rise to the distinctive dou-
ble aspect of the concept of policy paradigm in new institutionalist writ-
ing. In its first aspect, “paradigm” is a term of art in the broad stream
of neo-institutionalist policy theory (a historical policy paradigm).
In its other, more reflexive, aspect, “paradigm” reminds us that neo-
institutionalism, and other policy theories, may constitute, or at least
substantially contribute to, a theoretical paradigm in policy studies with
consequences for how actors understand each other and the world of
policy and politics itself. The chapter focuses on the consequences of a
paradigm’s double aspect for neo-institutionalism itself and specifically
on the question of whether any specific version of new institutionalism
can use the double aspect to theoretical advantage.
Hall’s own work displays the double aspect in a negative way in the
form of a frustrating lack of explicit connection between historical policy
paradigms and theoretical paradigms. In retrospect, it is perfectly clear
that his landmark piece on the policy paradigm and its role in explain-
ing policy change (Hall, 1993) was written from the point of view of
a committed neo-institutionalist. Even the characteristic puzzle – how
do we explain change when so much of the everyday world of politics
Jeremy Rayner 65
and policy is dedicated to reproducing stability? – belongs squarely in

the new institutionalist problematique. In addition, Hall’s starting point
in late twentieth-century state theories, though chosen because of state
theorists’ relative sophistication in dealing with institutions and in
avoiding behaviourism, now obscures rather than clarifies this interest.
Hall’s (1993) state-centric approach to policy making and policy change
is qualified a little by his use of the expert policy community idea, but it
is a complicating factor in the later trajectory of policy paradigms. It has
made it more difficult for the two main sub-disciplines where the para-
digm idea has gained traction – comparative politics and international
relations – to speak to each other because the latter has typically taken
a less state-centric approach. Moreover, Hall’s (1993) starting point has
contributed to confusion over the level of generality of the ideas iden-
tified as paradigms with an uneasy distinction running through the
literature between paradigms in policy fields (of interest to students of
comparative politics) and paradigms as very broad movements of ideas
(e.g., “neoliberalism” in international politics).
In Hall’s original formulation, the policy paradigm constitutes a
framework of ideas and standards that specifies not only the goals of
policy and the kind of instruments that can be used to attain them,
but also the very nature of the problems that they are meant to be
addressing. (1993, p. 279)
Loosely following Kuhn’s usage in The Structure of Scientific Revolutions

(1970), the policy paradigm is presented as a concept that can solve
the puzzle of stability and change raised by the new institutionalism.
A paradigm is a “way of seeing” that not only causes some issues to
appear as problematic and in need of policy attention while others are
backgrounded but also suggests some kinds of policy instruments as
more likely to succeed in problem solving than others. Paradigms, in
this sense, are powerful sources of relative policy stability leading to an
incremental approach to tinkering with policy instruments and settings,
akin to the “normal science” found in the original Kuhnian approach
to paradigms. Paradigms are also the explanation of the relatively rare
phenomenon of dramatic change, when one paradigm gives way to
another in the policy equivalent of a scientific revolution.
When Hall came to write explicitly about the new institutionalism,
however, policy paradigms play no part. Instead, Hall and Taylor (1996)
offer an account of the fracturing of new institutionalism into the clas-
sic trinity of RI, SI and HI. The schools are presented as differing in their
emphasis with respect to two contending approaches to the critical rela-

tionship between institutions and actors. One, the calculus approach,
argues that institutions (in the sense of “the rules of the game”) per-
sist because they provide optimal outcomes for actors who orient their
behaviour towards them, even if only in the sense of strategically “gam-
ing” the rules. The other, the cultural approach, works with a much
broader conception of institutions as sociological frames, explaining the
persistence of institutions as part of the “taken for granted” furniture of
social life, even though skilful actors can negotiate their way around the
furniture to reach their goals. The apparently peculiar assumption that
two approaches create three kinds of institutionalism is explained by
Hall and Taylor (1996) in a typically careful and fair-minded effort not
to impose artificial order on individual contributions. While the calculus
approach dominates RI and the cultural approach plays a similar role
in SI, neither are monoliths. HI (with which Hall aligns himself) is an
uneasy mixture of both approaches, constituting, in the authors’ view, a
potential bridge, or third way, between contending schools. Nonetheless,
the role of policy paradigms in bridging the gap is not explored at all.
Hall (2013) recently returned to a discussion of policy paradigms in a
commentary on a special issue, where his mood is detached and ironical.
He notes that his own title (“Brother can you paradigm?”) has an air of
the importunate about it, but it also suggests that the author has wea-
ried of his own creation and the torrent of academic debate that it has
unleashed. The context this time is the recent global financial crisis and
the suggestion that the more interventionist style of policy making that
it evoked in many countries might usher in a revival of Keynesianism.
While Hall (1993) provides, of course, a detailed discussion of how a mon-
etarist economic policy paradigm came to supplant an older Keynesian
orthodoxy, he rejects the idea that the financial crisis marks a simple
return of the dispossessed. While claiming to discern some of the pre-
conditions for a major shift in policy (a “paradigm shift”) he notes that a
major change of this kind is a joint effort of theory and political action.
“[A] paradigm overarching enough to usher in a new era (as distinct from
the narrower paradigms that often dominate specific fields of policy)
must speak to broader political issues if it is to have compelling politi-
cal appeal” (Hall, 2013, p. 191). Keynesianism and monetarism spoke
to the issues of their times, not ours. A new paradigm will be the joint
work of political economists and “of politicians experimenting with new
approaches to the economic and political problems of our time . . . those
seeking such a paradigm would do well to begin in the academy but then
lift their eyes to the political arena” (Hall, 2013, p. 191).
Jeremy Rayner 67
Three responses
Ideational analysis
One response to Hall’s (1993) introduction of the policy paradigm con-
cept was part of a broad if rather diverse stream of commentary on the
role of ideas or “ideational factors” in neo-institutionalist explanations
of policy change. Much of this work has close affinities with both HI
and SI (Béland & Cox, 2010; Campbell, 2004), but there is also clear
tendency to become impatient with neo-institutionalist analysis and an
urge to move on. Ideational analysis and this strand of the debate over
the explanatory role of ideas in new institutionalism are usefully sum-
marized in Berman (2012), where the genesis is explicitly identified with
Hall’s (1993) discussion of paradigms. Significantly, after reviewing the
basic argument of Hall (1993), Berman (2012) makes no further mention
of paradigms, arguing instead that further progress is only possible by
clearly delineating what ideational scholarship means by “ideas”, distin-
guishing between concepts such as beliefs, norms, cultures and ideolo-
gies (not, it should be noted again, paradigms), attending specifically
to their level of generality and the specific roles that they could play as
independent variables in explanations of change.
Berman begins by noting the prominence of the discussion of state
theories in Hall (1993). She suggests that the early neo-institutionalists’
enthusiasm for state-centric explanations of policy change was quali-
fied by a concern that “the state” is a problematic agent of change, not
only in respect of identity (the so-called unitary actor problem in state
theories), but also with respect to motivation and mechanisms. In other
words, state-centric theories of policy change need to be able to answer at
least three questions: who are state agents, why do they act to promote or
hinder policy change and how do they do so? Both the concept of a pol-
icy paradigm and the broader idea of social learning found in the title of
Hall (1993) are directed in large part towards answering these questions.
Up to a point, the answer that Hall (1993) gives is very clear, building
on and extending the social learning literature in ways that had already
been taken up by other policy theorists, and not always by those working
in a neo-institutional tradition (Bennett & Howlett, 1992). Hall’s (1993)
analysis invokes the famous discussion of different “levels” or “orders” of
policy change and hence the distinction between policy changes where
broad policy goals remain constant while the instruments and their set-
tings are fine-tuned (second and first order change, respectively) and
the more striking case of third order change. In this latter case, policy
goals and possibly the whole edifice of problem definition and preferred
alternatives comes into question. It is during this discussion, of course,

that Hall (1993) appeals to the idea of a policy paradigm and the possi-
bility (or puzzle) of paradigmatic change. As Berman (2012) notes, there
is a preliminary answer here to the question of who acts: policy expert
communities are responsible for first or second order change, while the
involvement of broader social and political forces is needed for paradig-
matic change, the distinction Hall (2013) chose to restate in his reflec-
tions on the career of the policy paradigm idea. The questions around
motivation and mechanism are less well developed, but the challenge
of developing these distinctions is largely taken up by historical institu-
tionalists over the ensuing decade.
In Berman’s (2012) view, however, the outcome of the debates about
ideas and institutions was not to create a new, or improved, kind of
institutionalism, but eventually to leave institutionalism behind. Under
the impact of the ideational turn, the definition of an institution
became less distinct, including, for example, ideational features such
as norms, cultural practices and belief systems (Berman, 2012, p. 222).
Actors’ preferences came to be seen as socially constructed by dominant
ideas rather than by interests or roles, giving a much more empirical
account of motivation and rationality, and so on. These developments
led Berman (2012) to posit a “fairly logical progression” from the early
efforts of Hall through HI to “ideational scholarship”.
Above all, the verdict of ideational analysis on the new institutional-
ists’ understanding of the relationship between policy paradigms and
policy change has been to focus on the absence of serious candidates
for causal mechanisms. Shpaizman (2014), for example, identifies an
additional problem arising from the original discussion of the “orders”
of learning and change in Hall (1993). To Berman’s (2012) charge of lack
of precision, she adds the claim that the very weak theorization of the
connection between paradigm change and third order policy change in
Hall (1993) allows the distinctive picture of ideational change found in
Kuhn (1970) to do the explanatory work. Periods of “normal science”
correspond to first and second order policy change, in which learning
about success and failure results in tinkering with existing instruments
and settings within an established framework of problem definition.
When “anomalies” arise that cannot be accommodated within that
framework, the stage is set for the kind of “Gestalt switch” that Kuhn
(1970) identified with a scientific revolution and Hall (1993) illustrated
with the replacement of Keynesianism by monetarism in economic pol-
icy. As Shpaizman points out, this picture tends to privilege the idea of
punctuated equilibrium over other patterns of policy change (2014, p. 4).
Jeremy Rayner 69
By failing to address questions of mechanism, “paradigm shift” is either

an example of an exogenous shock or, illustrating a fundamental con-
fusion over causality, the consequence of such a shock. She is surely
right to conclude that a consistent institutionalism of any kind needs
a better understanding of how ideas play a role in the key processes of
endogenous change identified in neo-institutionalist scholarship. The
other two schools of thought represent new institutionalists struggling
to meet this challenge, with limited success.
Discursive institutionalism
In a brief but telling aside, Berman (2012, p. 225, n. 5) remarks that
she considered the idea of “discourse” as a potential ideational vari-
able, but ultimately rejected it on the grounds that its lack of a clear
and generally agreed upon definition renders it unsuitable for rigorous
social science analysis. She also distinguishes efforts such as her own
that retain the broadly positivist approach to explanation she detected
in Hall (1993) and attempt to develop causal models from the more
far-reaching attempt to substitute a constructivist mode of explanation
for the positivist foundations of institutional analysis (2012, p. 233). In
doing so, she sets her project apart from the two other developments
analysed in this chapter.
In contrast to Berman’s (2012) approach, it is in large part the broad
church and suggestive character of the term “discourse” that recom-
mends it to Vivien Schmidt. She adopted the term precisely in order
to recognize the wide range of interpretivist approaches in the social
sciences “that take ideas and discourse seriously” (Schmidt, 2011, p. 107)
without having to choose between them. What is needed, she argues,
is not a series of neat categorical distinctions at all, but “an umbrella
concept” under the shelter of which these disparate but related interpre-
tivist theories can “discuss, deliberate and contest” their mutual under-
standing of the central role of ideas to theories of change (Schmidt,
2011, p. 107). Paradigms play a correspondingly more important role in
discursive institutionalism (DI) and are explicitly linked to similar con-
ceptions of foundational ideas, or ideas that exert influence on action
in other traditions (Skogstad & Schmidt, 2011), notably the sociological
literature on frames and framing (Schoen & Rein, 1994) and the policy
literature on frames of reference or référentiels (Jobert, 1989). However,
the result of trying to keep so many players in the same tent is that
many of the problems with the concept that emerged in ideational anal-
ysis reappear and the solutions, while more clearly neo-institutionalist,
are still sketches rather than clearly worked-out designs.
Writing in the introduction to a special issue on policy change in the

European Union, Schmidt and Radaelli (2004) noted that
[methodologically], the study of European policy change has also

become increasingly split among those who emphasize interest-based
rationality and game theoretic behaviour; institutional path depend-
encies and historically-shaped patterns of development; social con-
structions of action, culture and identity; or, most recently, ideas and
discourse. (p. 183)
Several important themes in the DI tradition emerge from this pas-

sage. First, the characterization of the “split” is clearly set up in terms
of the neo-institutionalist schools: RI, HI and SI, respectively, followed
by something presented as relatively new in the shape of “ideas and
discourse”. Second, the general theoretical problem that is the occasion
for the emergence of DI is identified as the problem that policy change
presents for orthodox neo-institutionalist explanations. Third, the con-
text is the study of policy change in the EU, specifically, as the whole
piece makes clear, “second generation” studies of the process of national
adjustment to EU policies.
Taking these themes in order, Schmidt (2010) has, first, always main-
tained that DI is not intended to replace the other institutionalisms, but
to supplement them by providing a more complete explanation of polit-
ical and policy change. In this sense, at least, DI is similar to ideational
analysis in adding to the range of factors available in explaining change.
The point that sometimes discourse matters a great deal and sometimes
it does not matter very much, or even at all, is a reiterated and over-
looked claim in much of Schmidt’s writing. Second, the “institutional
setting” (Schmidt & Radaelli, 2004, p. 184) for any kind of explanation
of policy change is also emphasized. Discourse is important because it
is part of the “vast range of rules . . . that affect policy making in any
context” (2004, p. 184). And finally, the European theme is significant
because of the important and often unnoticed emphasis on explana-
tions that involve action at a distance. Discourse turns out to be espe-
cially useful in explaining how events at one level of governance affect
events at another, how rules can both constrain in one place and enable
at another or, in the case of policy learning, how events at one time can
have an effect at another time.
In terms of the March and Olsen (1984) programme for new institu-
tionalism, DI is thus in the forefront of explaining inefficient histories
in terms of alternative logics of action.
Jeremy Rayner 71
Missing from mainstream comparative political economy, in short,

is a framework for analysis capable of endogenizing agency in such
a way as to explain the dynamics of institutional change and (con-
tinuity) and which can provide information about how, why, and
when political economic actors may (re)shape their macro-historical
institutions, (re)conceptualize their strategic interests, and (re)frame
their cultural norms. This approach is discursive institutionalism.
(Schmidt, 2008b, pp. 2–3)
That said, the development of DI has not been without its character-
istic difficulties, many of which stem precisely from the effort to keep
all (explanatory) options open. An often unnoticed development of DI
in this respect is a central ambiguity in the concept of a paradigm in
terms of its level of generality. Referencing unpublished work by Mehta
(2013), Schmidt identified ideas at “three levels of generality” (2008b,
p. 8), namely, policy ideas, programmatic ideas and philosophical ideas.
These distinctions look superficially similar to Hall’s (1993) original dis-
cussion of paradigms and paradigm change, but they are distinguished
by their susceptibility to change or how they change rather than by
what changes. In Schmidt’s (2008a) version, policy ideas change most
rapidly, supplanted by new ideas that exploit “windows of opportunity”
(Kingdon, 1984) or loss of viability (Hall, 1993) or feasibility (Majone,
1989). Programmatic ideas reference not only Hall’s paradigms, but
also Jobert’s “frames of reference” and, while persisting longer than
policy ideas, are subject to revolutionary transformation at critical
moments of the kind described in Hall (1993). Philosophical ideas, or
Weltanschauungen, are least likely to change and their pattern of change
most resembles the evolutionary logic that underpins the broad neo-
institutionalist approach to change, stressing path dependency and
lock-in.
While Schmidt’s (2008a, 2008b) categories represent a genuine effort
to address the question of mechanism, the relationship to Hall’s (1993)
work is not easy to understand. Clearly, DI’s second level of “program-
matic” change appeals to the idea of a policy paradigm and to paradigm
shifts, although here Schmidt notes that “[p]aradigm shift may serve as
a metaphor for radical ideational change, but so far we really don’t know
how or why or even when the shift takes place” (2008b, p. 9). One inter-
pretation might be that the new second level of DI is equivalent to Hall’s
third level so that “policy ideas” include changes in both setting and
instruments (Hall’s [1993] first two “orders” of change). On this view,
DI’s third level of generality is simply beyond the scope of Hall’s original
analysis and particularly the latter’s interest in establishing a relation-

ship between policy change and the mechanism of social learning.
On an alternative reading, the levels do map onto the orders, so that
policy programmes are the equivalent of the contemporary idea of a pol-
icy mix or policy portfolio, including beliefs about the causal relation-
ships between policy instruments, target behaviours and the likelihood
of achieving policy goals. These would be supported by policy paradigms
in Hall’s (1993) original sense of the phrase and identified by him as the
“narrow sense” of a paradigm. Very general ideas, for example, about the
role of the state in the economy, or the scope of individual responsibility
in the face of beliefs about the social causes of behaviour, constitute par-
adigms in Hall’s broader sense and are referred to as philosophical ideas
in the lexicon of DI. Changes here would not involve the narrow sense
of paradigm, but be examples of the “paradigms overarching enough
to usher in a new era” identified in Hall (2013). At this level, change
does not come about as a result of learning (even social learning), but
because of the general uncertainties about previously taken for granted
ideas characteristic of a crisis or Kuhnian Gestalt shift. In later published
work, Mehta (2013) uses just such a distinction between paradigms at a
“middle level of generality . . . that of defining a problem within an issue
area” and paradigm change that entails “shifts in the climate of ideas at
the broader level of public philosophies” (2013, p. 292).
Constructivist institutionalism
In reaching this conclusion about the importance of shifts in public
philosophy as the mechanism of change on the scale of broad institu-
tional replacement, DI arrives at the starting point of the final variant
potentially vying for the title of a fourth new institutionalism. In spite
of the references to Gestalt shifts in Hall’s (1993) account of paradigm
change, there is little, in either his original account of social learning
or in his subsequent HI-inspired writing, to suggest that he would want
to adopt an account of discourse as anything more than partially con-
stitutive of the policy world. His remarks about seeking the elements of
a new paradigm in the arena of politics versus the primarily historical
explanatory value of academic neo-institutionalism is a clear indication
of where he would draw the line here. Schmidt’s (2008a) increasingly
acrobatic efforts to maintain a distinction between discourse and insti-
tutions without giving one priority over another is partly caused by her
own desire to remain connected to a less constructivist version of the
role of ideas in politics and policy. For a more thoroughgoing construc-
tivist account, we need to turn to the work of Colin Hay (2008) and, on
the latter’s attribution at least, Mark Blyth (2002).
Jeremy Rayner 73
In terms of the history of ideas presented in this chapter, Hay’s first

contribution appears on the scene in a response to Hall and Taylor
(1996). Hay and Wincott (1998, p. 955) argue that, properly understood,
HI retains a distinctive approach to causality and change that cannot
be reduced to either calculus (rational choice) or culture (sociology):
“[c]hange is seen as the consequence (whether intended or unintended)
of strategic action (however informed or misinformed) of an institu-
tional context that favours certain strategies actions and perceptions
over others”. This kind of strategic action results in change to both the
institutions themselves (“a partial transformation of the institutional
environment”) and “strategic learning” on the part of the actors, involv-
ing a revision of their assessments of the feasibility and desirability
of their goals and the means they have employed to realize them. In
support of their position, Hay and Wincott (1998) note that it has the
potential to overcome institutionalism’s “characteristic emphasis on
institutional inertia” without losing the sense the institutions continue
to structure the possibilities at any particular point in time. And they
stress the importance of ideas in their formulation, acting as “cognitive
filters” for the assessment of strategic action and its possibilities.
While Hay and Wincott (1998) is a generally favourable account of HI,
it also has a critical edge. The authors argue that both of the approaches
to explanation described by Hall and Taylor (1996) (they call them cal-
culus and cultural “ontologies”) are seriously deficient and to miscon-
ceive HI as somehow combining or bridging between these ontologies
risks creating the worst of both worlds, rather than remedying the weak-
nesses of one with the strengths of the other. In introducing ontological
language, Hay and Wincott (1998) ask their readers to take a further
(philosophical) step back to ask how the neo-institutionalisms address
the basic problem of any kind of institutionalism, the interaction of
structure and agency. As already noted, in their view, the most produc-
tive approach is found amongst those historical institutionalists, includ-
ing Hall himself in some moods, who posit a world of agents acting
strategically with respect to institutional constraints whose functional-
ity and efficiency are always in question. Neither rational calculating
machines nor cultural dopes (Garfinkel, 1967), actors reflexively moni-
tor their interactions with institutions, seeking to change their insti-
tutional context as they learn about their own strategic successes and
failures.
In addition to reintroducing a general concern with social learning
and change (a key idea in Hall’s (1993) original account of policy para-
digms), Hay and Wincott’s view of the distinctive contribution of HI also
grants “crucial space” to the role of ideas (1998, p. 956). The world is not
just composed of actors and institutions, but is also composed of actors,

institutions and ideas about institutions. Hall’s and Taylor’s (1996) mis-
conception about the ontological basis for a distinctive HI turns out to
derive from the disappearance of the paradigm: “perceptions about what
is feasible, legitimate, possible and desirable are shaped both by the insti-
tutional environment in which they find themselves and existing policy
paradigms and worldviews” (p. 956). Hay (2008) will later identify this
approach to institutions as “constructivist institutionalism” and seek to
differentiate it from what, in his view, HI has allowed itself to become
under the warring influence of the calculus and cultural ontologies.
Much (though by no means all) of Hay’s writing on constructivist
institutionalism has been a contribution to, and a reflection on, the bur-
geoning literature about “crises” (1996, 1999, 2004, 2010). There are, of
course, historical echoes here from Hall (1993). The state theorists who
formed Hall’s (1993) point of departure were much exercised by the idea
that state actors, normally fairly constrained in their freedom of action
by powerful societal actors, throw off their shackles during a crisis and
achieve at least a relative autonomy from social forces. Moreover, Hall
(1993) provides the original example of a link between crises and para-
digmatic change in his narrative of the economic crisis of the 1970s and
his description of “stagflation” as a Kuhnian anomaly that cast doubt
on the utility of the Keynesian paradigm in economic theory and public
policy. Hay (2008) notes the extension of these ideas in Blyth’s (2002)
study of the political role of business elites in the US and Sweden, where
an extra dimension of constructivism is added by the claim that crises
themselves are in large part constructed by crisis narratives deliberately
propagated in order to undermine stability and bring about change.
As Hay notes, the essence of constructivist institutionalism is found in
Blyth’s claim that, while actors may have material interests that could be
independently ascribed to them by an outside observer, it is their own
ideas about these interests that render the interests “actionable” (Blyth,
2002, p. 39, cited in Hay, 2008, p. 68). On this account, paradigms
offer interpretive frames for understanding interests and making them
“actionable” and hence paradigm change will have very profound con-
sequences for the political possibilities inherent in a particular clash of
interests. Crises and paradigm change go hand in hand precisely because
the old certainties about interests, interest-based alliances and interest
politics are undermined and a period of self-doubt and self-examination
ensues before Kuhnian normal politics is restored.
Constructivist institutional analysis was thus ready to hand when the
crisis of 2008/09 hit, but the challenge of interpreting current events
Jeremy Rayner 75
only served to expose the tensions and ambiguities in the constructivist

institutionalist account linking paradigms and paradigmatic change to
crises and the resolution of crises. As Hay (2008) pointed out in his origi-
nal discussion of Blyth (2002), there is a crucial ambiguity in the idea
that a crisis is a moment where the traditional interpretation of material
interest that connects interest with action is discarded. In some moods,
constructivists are tempted to say that the result is confusion about
interests which can be exploited by more clear-sighted actors who have
a crisis-explaining narrative that suits their purposes ready to hand. In
others, they seem to suggest that crises “strip away the veils” and allow
actors privileged access to material interests for a brief moment – these
are interpretations of the actions of trades unions in Britain during the
“Winter of Discontent” in the 1970s (Hay, 1996). As Hay (2008) points
out, the problem with the former account is that it fails to explain why
some actors are able to take advantage of the confusion engendered by
a crisis while others are merely confused; the problem with the latter
account is that it is not constructivist at all. As Hay (2011) has argued,
none of these difficulties should prevent us exploring the possibilities
of a constructivist institutionalism, but doing so productively requires
much more careful attention to the relationship between ideas that
actors act upon, the discourses from which ideas emerge and, most
important of all, the institutional contexts in which ideas acquire mean-
ing for actors. Failure to do so, he argues, has been a general feature of
interpretivist approaches to explaining policy change and not just the
constructivist institutionalist variant.
More important, however, is the fact that, until these ambiguities are
resolved, constructivist institutionalism finds itself with its own ver-
sion of the “two explanations” problem in institutionalist theory noted
above in the section “The new institutionalisms”. The well-known diffi-
culty that institutionalists have in explaining how an institutional con-
text that provides a very clear and consistent explanation of stability can
somehow find itself subject to more or less rapid and destructive change
is well known (Campbell, 2004). One response to this difficulty is to
argue that one kind of explanation (the institutionalist kind) works for
stability while some completely different account (exogenous events,
ideational factors) explains change. The suggestion that paradigms in
the sense of “public philosophies” provide a key part of the explanation
for political and policy stability by covering over an underlying clash of
material interests, while an exogenous “crisis” temporarily upsets things
until a new public philosophy takes its place, is a classic variant of this
“one explanation for stability, another for change” problem.
In the broader movement of the new institutionalism, this problem

has been tackled most successfully by HI, notably in the exploration
of patterns of endogenous change which can support, or undermine,
institutions and institutionalized policies over time. Processes such as
drift, layering and conversion have provided the basis for rich accounts
of gradual institutional change that include careful attention to ideas,
notably work by Hacker (2004) and Béland (2009) on changing social
policies in the US and Europe. As Hay (2008) notes, in the end, whether
constructivist institutionalism should be considered a distinctive vari-
ant – a fourth institutionalism – depends on how we understand HI.
If, as Hay claims, the dominant tendency in neo-institutionalism is to
fall back on the political calculus explanations characteristic of DI, then
constructivist institutionalism remains a distinctive effort to understand
the “dynamic interplay of structure and agent . . . and indeed material
and ideational factors” (2008, p. 62). If, on the other hand, as a more
optimistic reading of the continuing development of HI might suggest,
the goal of HI continues to be “to develop an endogenous institutional-
ist account of the mechanisms and determinants of complex institu-
tional change”, then reflection on the role of ideas will continue to be
central to the neo-institutionalist project, at least in its HI variant.
Conclusion
What can be learned from the career of the policy paradigm in the litera-
ture of the neo-institutionalists and their critics? In general, we should
note that the broad theoretical context of this whole debate is the
explanation of change and, by extension, the explanation of stability.
Explaining stability and change has a been a major concern of theorists
in political science, sociology, policy studies and other disciplines for
two decades or more, and the enduring interest in “paradigms” derives
in large part from the role the concept has played in the theoretical
debate around change. Neo-institutionalism, with its tendency to posit
stability imposed by institutions as a kind of natural order of things and
its subsequent difficulties with the explanation of change, has provided
fertile ground for experimentation with concepts like the paradigm that
promise a way out of this difficulty.
Three conclusions emerge. First, there is the continuing utility of a
general distinction between change as the result of what might be called
“rational processes”, such as learning and argumentative discourse, and
change that results from the collapse of widely held public philosophies,
or cultural frames, within which lessons and arguments are made sense
Jeremy Rayner 77
of and assessed. The utility of the distinction has been obscured, rather
than clarified, by the noisy debate over the role of ideas in the explana-
tion of change. To this extent, at least, Berman (2012) is surely correct to
ask for greater clarity in what is meant by “ideas”. The basic distinction
is lost if we conflate ideas as argument (or “ideas as weapons”) with ideas
as frames. Equally important, there is the question of mechanism implied
in the distinction between “ideas as argument” and “ideas as frames”.
Hall (1993) proposed an influential account of at least one such mecha-
nism in the shape of learning. Subsequent accounts often attempted
to develop the original distinction based on generality and scope, for
example, by distinguishing between policy learning and social learning,
but equally significant efforts are being made to identify other mecha-
nisms and processes. The early effort of Schmidt and Radaelli (2004) to
move the question of mechanisms forward, by introducing a distinction
between communicative and coordinative discourse, received a luke-
warm response from theorists of change, but is still invoked in empirical
studies (Casado-Asensio & Steurer, 2013). Dunlop and Radaelli (2012)
have returned to the learning mechanism in an attempt to impose some
order on an increasingly chaotic field.
Second, and rather more interesting from a theoretical point of view,
is the suggestion of the constructivists, especially Hay (2008), that idea-
tional change is linked to the processes of endogenous change investi-
gated by historical institutionalists as early as Thelen and Steinmo (1992).
While the framing of this suggestion in terms of Hay’s own interest in
“ontologies” renders it less accessible than it might otherwise be, it does
serve to reintroduce the characteristic concerns of HI with timing and
sequence. As erstwhile HI theorists such as Pierson have moved closer to
RI in stressing strict path dependence and the rational calculation of the
costs of retracing one’s steps along a particularly well-entrenched policy
path, paradigms have emerged as a major competitor in explaining how
a more broadly conceived set of “policy legacies” may both constrain and
enable change. Here progress seems to depend upon very careful specifi-
cation of the level of generality of the policy elements and the ideas asso-
ciated with them. The debates within DI about the difference between
a policy paradigm and a public philosophy may seem rather arcane, but
there is no benefit to be gained by confusing them. In fact, an even
more fine-grained analysis may very well pay dividends here, for exam-
ple, in Cashore and Howlett’s (2007) idea of “implementation logics”
as distinct from idea about appropriate policy goals, or the ideational
components of “instrument constituencies” as opposed to attempting
to construct these constituencies purely on the basis of ascribed interest.
Finally, from the earliest contribution through the recent debates

about the role of paradigms in the latest financial crisis, there is the
fascinating question about reflexivity and what our own knowledge
about paradigms can contribute to political action. Ironically, here a
new institutionalism more sensitive to the role of ideas in explanations
of change arrives at a very similar conclusion to one less sensitive, albeit
by a different route. While mainstream new institutionalism calls on
exogenous events, which by definition cannot be known in advance, to
bring about significant institutional and policy change, the ideational
versions, by giving such a strong role to the contest over meaning, can-
not be predictive either. As Hay (2008) puts it, if we truly acknowledge
the independent causal and constitutive roles of ideas and discourse,
then we can only know the meaning of events once they have been suc-
cessfully narrated. This may be a disappointing conclusion to those who
seek more guidance from theory in the world of action, but it is not a
call to quietism. As Hall (2013) points out, we should expect to find the
next paradigm in politics, not in theory.
Note
1 “The New Institutionalism” was ranked as the sixth most-cited article in the
history of the American Political Science Review in that journal’s centennial
year, 2006: http://www.apsanet.org/imgtest/apsrnov06top20.pdf, accessed
March 23, 2012. Hall and Taylor’s (1996) article is currently ranked the most-
cited article ever published by Political Studies.
References
Béland, D. (2009). Ideas, Institutions and Policy Change. Journal of European
Béland, D., & Cox, R. H. (Eds.). (2010). Ideas and Politics in Social Science Research.
Bennett, C. J., & Howlett, M. (1992). The Lessons of Learning: Reconciling
Theories of Policy Learning and Policy Change. Policy Sciences, 25(3),
275–94.
Berman, S. (2012). Ideational Theorizing in the Social Sciences since “Policy
Paradigms, Social Learning, and the State”. Governance, 26(2), 217–37.
Blyth, M. (2002). The Great Transformations. Cambridge: Cambridge University Press.
Casado-Asensio, J., & Steurer, R. (2013). Integrated Strategies on Sustainable
Development, Climate Change Mitigation and Adaptation in Western Europe:
Communication Rather Than Coordination. Journal of Public Policy, 34(3),
437–73.
Jeremy Rayner 79
Cashore, B., & Howlett, M. (2007). Punctuating Which Equilibrium? Under-

standing Thermostatic Policy Dynamics in Pacific Northwest Forestry. American
Dunlop, C. A., & Radaelli, C. M. (2012). Systematizing Policy Learning: From
Monolith to Dimensions. Political Studies, 61(3), 599–619.
Garfinkel, H. (1967). Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice-
Hall.
Hacker, J. S. (2004). Privatizing Risk without Privatizing the Welfare State: The
Hidden Politics of Social Policy Retrenchment in the United States. American
Hall, P. A. (2013). Brother, Can You Paradigm? Governance, 26(2), 189–92.
Hall, P. A., & Taylor, R. C. R. (1996). Political Science and the Three New
Institutionalisms. Political Studies, 44(5), 936–57.
Hay, C. (1996). Narrating Crisis: The Discursive Construction of the “Winter of
Discontent.” Sociology, 30(2), 253–77.
Hay, C. (1999). Crisis and the Structural Transformation of the State: Interrogating
the Process of Change. British Journal of Politics and International Relations, 1(3),
317–44.
Hay, C. (2004). Ideas, Interests and Institutions in the Comparative Political
Economy of Great Transformations. Review of International Political Economy,
11(1), 204–26.
Hay, C. (2008). Constructivist Institutionalism. In B. Rockman, R. Rhodes, &
S. Binder (Eds.), The Oxford Handbook of Political Institutions (pp. 56–74). Oxford:
Oxford University Press.
Hay, C. (2010). Chronicles of a Death Foretold: The Winter of Discontent and
Construction of the Crisis of British Keynesianism. Parliamentary Affairs, 63(3),
446–70.
Hay, C. (2011). Interpreting Interpretivism Interpreting Interpretations: The New
Hermeneutics of Public Administration. Public Administration, 89(1), 167–82.
Hay, C., & Wincott, D. (1998). Structure, Agency and Historical Institutionalism.
Jobert, B. (1989). The Normative Frameworks of Public Policy. Political Studies,
37(3), 376–86.
Kingdon, J. (1984). Agendas, Alternatives and Public Policies. New York: Longman.
Kuhn, T. A. (1970). The Structure of Scientific Revolutions (2nd ed.). Chicago, IL:
University of Chicago Press.
Majone, G. (1989). Evidence, Argument, and Persuasion in the Policy Process. New
Haven, CT: Yale University Press.
Manning, D. J. (1976). Liberalism. London: St. Martin’s Press.
March, J. G., & Olsen, J. P. (1984). The New Institutionalism: Organizational
Factors in Political Life. American Political Science Review, 78(3), 734–49.
March, J. G., & Olsen, J. P. (1989). Rediscovering Institutions: The Organizational
Basis of Political Life. New York: Free Press.
Mehta, J. (2013). How Paradigms Create Politics: The Transformation of American
Educational Policy, 1980–2001. American Educational Research Journal, 52(2),
285–324.
Pierson, P. (2000). Increasing Returns, Path Dependence, and the Study of Politics.
American Political Science Review, 94(2), 251–67.
Rockman, B. A., Rhodes, R. A. W., & Binder, S. (Eds.). (2008). The Oxford Handbook
of Political Institutions. Oxford: Oxford University Press.
Schmidt, V. A. (2008a). Discursive Institutionalism: The Explanatory Power of
Ideas and Discourse. Annual Review of Political Science, 11(1), 303–26.
Schmidt, V. A. (2008b). From Historical Institutionalism to Discursive Institution-
alism: Explaining Change in Comparative Political Economy. Paper presented
at the American Political Science Association Annual Meetings, Boston, August
2008.
Schmidt, V. A. (2010). Taking Ideas and Discourse Seriously: Explaining Change
through Discursive Institutionalism as the Fourth “New Institutionalism”.
European Political Science Review, 2(1), 1–25.
Schmidt, V. A. (2011). Speaking of Change: Why Discourse Is the Key to the
Dynamics of Policy Transformation. Critical Policy Studies, 5(2), 106–26.
Schmidt, V. A., & Radaelli, C. M. (2004). Policy Change and Discourse in Europe:
Conceptual and Methodological Issues. West European Politics, 27(2), 183–210.
Schoen, D. A., & Rein, M. (1994). Frame Reflection: Toward the Resolution of
Intractable Policy Controversies. New York: Basic Books.
Shpaizman, I. (2014). Ideas and Institutional Conversion through Layering: The
Case of Israeli Immigration Policy. Public Administration, 92(4), 1038–53.
Skogstad, G., & Schmidt, V. A. (2011). Introduction: Policy Paradigms,
Transnationalism and Domestic Politics. In G. Skogstad (Ed.), Policy Paradigms,
Transnationalism and Domestic Politics (pp. 3–35). Toronto: University of Toronto
Press.
Thelen, K., & Steinmo, S. (1992). Historical Institutionalism in Comparative
Politics. In S. Steinmo, K. Thelen, & F. Longstreth (Eds.), Structuring Politics:
Historical Institutionalism in Comparative Analysis (pp. 1–32). Cambridge:
Part II
5
Comparing and Contrasting Peter
Hall’s Paradigms and Ideas with the
Advocacy Coalition Framework
Paul Cairney and Christopher M. Weible
Introduction
The study of policy processes is growing and diversifying both in num-

bers of scholars and in theoretical and methodological approaches.
Scholars are increasingly developing established theories and creating
new theories, studying public policies in a variety of contexts that span
the globe, and applying a diversity of methodological and analytical
techniques. If communication among policy scholars is essential for les-
son learning and advancing the field then clear vocabulary lies at the
fulcrum of progress. A key way to improve that communication process
is to describe and compare, in great depth, the language of key concepts
and theories.
To that end, the purpose of this chapter is to compare and contrast the
conceptual definitions and theories associated with policy paradigms
and ideas (Hall, 1993) with the Advocacy Coalition Framework (ACF)
(Jenkins-Smith, Nohrstedt, Weible, & Sabatier, 2014; Sabatier, 1988).
The goal is not an integration but rather a communication between the
two approaches to clarify some of the differences and similarities and to
suggest areas and strategies for improvements and future research.
Challenges immediately arise, however, when comparing and con-
trasting Hall’s (1993) policy paradigms and ideas to any other theo-
retical approach or concepts found in public policy. The source of the
challenges lies in the obscure ontological origin of both paradigms and
ideas, which manifests in overly stretched and poorly defined defini-
tions that have befuddled communication and research among policy
scholars for decades. As such, two sets of questions must continue to
be asked. The first set is descriptive: what are ideas and paradigms and
how do the definitions of ideas and paradigms relate? The second set is
83
explanatory: what is paradigmatic policy change, what causes it, and

how long does it take?
Both sets of questions about paradigms and ideas remain largely unan-
swered. The principal reason is that ideas and paradigms are ontologically
obscure and, therefore, any attempt to draw explanatory leverage and
generalizable knowledge will be problematic. Yet, scholars continue to
study them. The popularity of ideas and paradigms stems probably from
their intuitive appeal, for the concepts are flexible enough to capture the
imagination of what is and what ought to be. Thus, policy scholars often
adopt an ‘I know it when I see it’ approach. As a result, they see what
they want to see in ideas and paradigms, and what they see is not always
the same. Despite plenty of attention among talented scholars, the defi-
nitions and relationship among paradigms and ideas remain nearly as
confusing and ambiguous today as they were more than two decades ago.
Given past efforts in attempting to clarify the meaning of paradigms
and ideas and strengthening the theory involving paradigmatic policy
change, this chapter takes a different path that involves an exploration
of Halls’ paradigms and ideas and the ACF. This exploration shows that,
while some commonalities can be identified, non-trivial differences
exist between the two approaches. It would be counterproductive to
the study of policy processes to overlook the nuanced differences in the
hope of building studies based on their similarities.
Descriptions of ideas and paradigms
For more than two decades, Hall (1993) has inspired and motivated schol-
ars to clarify the elusive meaning of paradigms and ideas. Recognizing
these past efforts and not wanting to replicate them, this section starts
with a brief summary of Hall (1993) to enable a later comparison with
the ACF. Fuller descriptions and interpretations of Hall (1993) can be
found in this volume and in, for example, special issues devoted to his
work (including Governance, 26, 2, 2013).
Hall’s (1993) article now commands 4,000 Google Scholar citations
and has become a key reference point in the field. If scholars want
to define a policy paradigm, or find a point of departure to present
their own argument, they usually start with ‘Policy Paradigms, Social
Learning, and the State’. Many discussions of ideas and paradigms start
with this quote:

Paul Cairney and Christopher M. Weible 85
instruments that can be used to attain them, but also the very nature
scrutiny as a whole. I am going to call this interpretive framework a
policy paradigm. (Hall, 1993, p. 279)
Part of Hall’s aim is to account for the role of ideas in policy processes (1993,
p. 276). This presents anyone with an immediate problem because the ‘idea’
concept is a very broad and not well-defined term (for examples of a wide
variety of attempts, see Cairney, 2012, pp. 222–3). Indeed, Cairney and
Heikkila (2014, p. 365) seek a way to categorize both ideas and beliefs as
[w]ays of thinking or the knowledge that plays a role in the policy

process. This [ideas] category may include beliefs, knowledge, world-
views, and shared definitions of policy problems, images, and solu-
tions within groups, organizations, networks, and political systems.
Some ideas or beliefs may be taken for granted or rarely questioned –
such as core beliefs, values, or paradigms. Others may be more malle-
able, such as proposed solutions to policy problems.
Similar to Cairney and Heikkila’s (2014) interpretation, a wide range

of authors have used ideas as a way to identify the role of beliefs and
give them analytically separate causal weight from ‘interests’ or ‘power’,
which are also difficult to define satisfactorily. In some cases, the bal-
ance of explanation – or, at least, the focus – appears to be in favour
of the independent role of ideas, such as when Richardson (2000,
p. 1019) describes them as viruses which provide a ‘shock to both exist-
ing institutional arrangements and the actors that benefit from them’,
Baumgartner and Jones (1993, p. 237) identify ‘powerful forces of
change that sweep through the entire system’, and Axelrod argues that
(1986, p. 1095) ‘an established norm can have tremendous power’ as a
‘mechanism for regulating conflict in groups’. However, further inspec-
tion suggests that the role of individuals and their choices is crucial – as
people show degrees of immunity to ideational viruses and resistance
to framing policy problems, and can choose to challenge or reinforce
norms. In other accounts, policymaking involves a battle between pro-
ponents of different policy ‘narratives’ (McBeth, Jones, & Shanahan,
2014), or the ‘social construction of target populations’ (Schneider,
Ingram, & deLeon, 2014). While many different uses of ideas exist in
the literature, the meanings and roles of ideas can be organized, ini-
tially, in three ways.
First, ideas relate to persuasion and argument, as resources in the policy
process, alongside the use of material and other resources (Hall, 1993,
pp. 291–2; Jenkins-Smith & Sabatier, 1993, pp. 44–5; Kettell & Cairney,
2010, p. 301; Kingdon, 1984, pp. 131–3; Majone, 1989, p. 2). A large
part of the literature considers the balance we need to strike between
explaining outcomes in terms of material power and persuasion, before
finding a way to say that both are important. This focus has strong links
to a large literature on agenda setting, problem definition, and ‘framing’
(Cairney, 2012, pp. 182–7).
Second, the meaning of ideas goes beyond persuasion. Ideas refer to
a shared language, sometimes implicitly or explicitly by policy partici-
pants. This is in reference to Hall’s (1993, p. 279) description of ideas
as part of an abstract framework of terminology. It is also a definition
that links ideas directly as constitutive components of policy paradigms.
That is, ideas constitute policy paradigms but not all ideas are policy
paradigms (Baumgartner, 2014).
Third, the meaning of ideas is associated with the phrase ‘I have an
idea.’ In other words, an idea refers to a proposed solution to a policy
problem. This definition reflects the interpretation of ideas in Kingdon’s
policy stream. For example, when Kingdon (1984) starts his work with
the following quote from Victor Hugo, ‘Greater than the tread of might
armies is an idea whose time has come’, the reference is not in relation
to any kind of value or belief but rather to a solution to a problem.
This interpretation of ideas in relation to paradigms suggests that solu-
tions are not simply resources that people choose to use, and are able to
promote through persuasion and argument. Rather, there is a ‘structural’
element to policy discussions; a paradigm provides the context in which
people use arguments and persuasion and within which certain solu-
tions are feasible or even conceivable. In one sense, this means manipu-
lating language explicitly to present a persuasive argument. In another,
it means using a language that everyone uses routinely, to the extent
that they (almost) take it for granted. Consequently, one might exert
power to frame a policy problem and propose an idea in the form of a
solution or to use ideas to influence the discursive context in which that
framing takes place. One might frequently challenge a policy, while a
challenge to the whole system of policymaking is rare.
The interpretation and reinterpretation of Hall

There is also a diverse interpretation of paradigms which builds on, or
reinterprets, the work of Hall (1993). At times, these interpretations
are similar to the three meanings we identify, but many have differ-
ent emphases and potentially different meanings. One reading of Hall’s
(1993) definition would suggest, for example, that we should treat a
paradigm as an encompassing gestalt, which exists independently of
the people operating within it. This interpretation treats paradigms as
something often unrecognized and implicit in shaping the thoughts,
language, discourse and narratives, and political behaviour of everyone
thinking and behaving therein.
Similar to the gestalt interpretation is one that focuses on structure
and is akin to a set of institutions which help explain ways of thinking
and regular patterns of behaviour in organizations and political systems.
Consequently, our focus could be on who constructed the institutions,
and in whose interests. This emphasis ties neatly to several discussions
of system-wide power, including Gramscian notions of the construction
of hegemony (1971; see Hindess, 1996, p. 5; Lukes, 2005, p. 27), as well
as the competition for ‘cognitive control’ – note the phrase ‘le referen-
tial’ (Jobert & Muller, 1987) which refers to a fundamental set of ideas –
which often look benign or innocuous – imposed by powerful elites
(Genieys & Smyrl, 2008, p. 23). From this definition of paradigms, poli-
cies or institutions that comprise them are translations or revealed ideas
(Daigneault, 2014).
In addition, the discussion of a ‘structural’ element to ideas links
strongly to the study of institutions (i.e., rules, norms, and ‘standard
operating procedures’) and ‘new institutionalism’. This is partly a posi-
tive sign, since it brings together two potentially separate literatures, to
aid the production of a common reference point for scholars, but also
a problem, since it adds to terminological confusion in two main ways.
First, although John (2003, p. 488; followed by Cairney & Heikkila,
2014; and Cairney, Studlar, & Mamudu, 2012) identifies commonality
among major policy theories, according to their focus on ‘five core causal
processes . . . institutions, networks, socioeconomic process, choices,
and ideas’, it is now difficult to distinguish between at least two of those
concepts. Second, the identification of a large number of approaches
to ‘new institutionalism’ shows us that, under the surface of a shared
concept, there is a huge amount of terminological confusion and debate
(for reviews, see Hall & Taylor, 1996; Peters, 2011). Most notably, there
is much debate about the nature of institutions as the rules and norms
that influence behaviour: for some, they almost represent structures that
bind behaviour (March & Olsen, 2006, p. 3); for others, they exist largely
in the minds of individuals (Ostrom, 2007, p. 23) or socialized practices
of groups (Rhodes, 2006, p. 91) and, as such, are relatively open to the
potential of challenge or change. Indeed, Mark Bevir rejected the use
of the word ‘institution’ altogether (at a Political Studies Association

Seminar, Manchester, 2009) because it has a ‘bewitching effect’ (‘Don’t
call practices institutions!’). As Cairney (2012, p. 75) argues, ‘This differ-
ence in language may seem trivial, but it highlights important debates
which often divide the discipline.’
‘Paradigm’ could also be used to describe a type of policy subsystem. In
particular, the hegemonic aspect of policy paradigms would suggest that
the policy subsystem were controlled by a policy monopoly attempting
to keep issues off the ‘macropolitical’ agenda by utilizing strategies of
dominant framing of policy issues (Baumgartner & Jones, 1993, 2009;
Jones & Baumgartner, 2005). Indeed, when discussing the many simi-
larities between his and Hall’s work, Baumgartner (2013, p. 251) tries
to clarify this connection: ‘When ideas are widely shared by an entire
policy community, they can be called a paradigm. Some policy com-
munities may well be dominated by a single paradigm, others may see
competition, and others may see the replacement of one dominant par-
adigm by another.’ So, for Baumgartner (2014, p. 476), paradigms relate
to the ‘grand’ or ‘fundamental’ aspects of systems. This distinction is
crucial, since a system-wide paradigm suggests that all participants oper-
ate within a common discursive framework, while a subsystem focus
suggests that the system contains a large number of competing frames in
different subsystems (or, perhaps, frames that are supported by different
groups in competition with each other), which either operate indepen-
dently of each other or occasionally combine to produce new subsystem
arrangements.
However, in this respect, Hall’s use of policy paradigm is not always
clear. Hall (1993, pp. 276, 288) focuses generally on ‘the political system
as a whole’ – ‘defined as the complex of political parties and interest
intermediaries that stand at the intersection between the state and soci-
ety in democratic polities’ – and furthers his argument with a focus on
the cross-cutting area of economic policy. However, he also suggests that
the maintenance of an economic paradigm allows routine aspects of pol-
icymaking (producing ‘first’ and ‘second order’ change) to be ‘relatively
insulated from the kind of pluralist pressures we often associate with the
broader political system’ (1993, p. 281). Hall’s (1993, p. 279) initial focus
is on the contribution of policy paradigms to ‘normal’ policymaking,
which involves policy continuity and non-radical change and the main-
tenance of a dominant way of thinking about the world. He draws on
Kuhn’s (1962) model of scientific progress, which suggests that natural
science is conducted and evaluated by communities who share a single,
common, view of the world, and rarely go back to ‘first principles’ to
question the assumptions on which their theories and models are built
(unlike social science, in which there is more competition to explain the
world).
Given the different definitions of ideas and paradigms, there are some
general interpretations that offer a basis for further discussions. First,
ideas are best interpreted as an umbrella concept or category of concepts
that involves a variety of mental constructs that operate differently in
the politics of policy processes. Second, policy paradigms are more of
a system-based concept that, if recognized, are generally agreed upon,
may represent an underlying implicit gestalt or explicit structure of the
system, and are drawn originally from ideas. Third, ideas and policy par-
adigms are related in that any paradigm consists of realized or translated
ideas but not all ideas are associated with policy paradigms.
Descriptions and explanations of paradigmatic change
Hall’s description of paradigmatic policy change has three orders. At the

most profound level is Hall’s (1993, p. 280) identification of ‘third order’
(paradigmatic) change, which has three main elements: ‘the power of
one set of actors to impose its paradigm over others’; a shift in authority
in the political system; and, policy failure associated with the failure of
experts to explain and solve the problem. In other words, there is wide-
spread attention to cumulative policy failure, and either a new govern-
ment takes over and introduces radically different policies, or existing
policymakers reject the advice of one set of experts in favour of another
(Cairney, 2012, p. 229). In the case of the United Kingdom’s economic
policy, paradigm change followed the decision by elected governments –
and the Thatcher governments in particular – to draw on monetarist
ideas put forward by international economists (and championed by key
political and media figures) and challenge the dominance of Keynesian
expertise within the Treasury (Hall, 1993, p. 287). This analysis served to
reinforce the argument that ‘ideas matter’ because the monetarist para-
digm became a resource for Thatcher and key ministers to use, to challenge
the economic orthodoxy within the government’s civil service –
an option not available to Edward Heath, who set off on a similar path
in the early 1970s before making a famous ‘U turn’ (Hall, 1993, p. 290).
Hall contrasts this event with the more routine ‘first’ or ‘second order’
policymaking that takes place within an existing paradigm when its
participants do not question the rationale for policy goals. It compares
with Kuhn’s (1962) model of scientific progress which identifies a simi-
lar process in which scientists take their paradigm for granted for long
spells, only to face major crises in key eras. Kuhn suggests that scientific
advances have not been incremental or based on the linear accumula-
tion of knowledge. Rather, when communities of experts fail to pro-
vide further scientific advance, they are overtaken and replaced by other
communities with different ideas and ways of thinking. This is often
caused by a crisis prompted by new information and the inability of
scientists to explain why the world does not work in the way they think
it does.
The policymaking equivalent is a political crisis prompted by policy
failure, calling into question the thinking behind policy and undermin-
ing the status of its advocates (Cairney, 2012, p. 230). It produces a bat-
tle of ideas, which only ends ‘when the supporters of a new paradigm
secure positions of authority over policymaking and are able to rearrange
the organization and standard operating procedures of the policy pro-
cess so as to institutionalize the new paradigm’ (Hall, 1993, p. 281). As
Wilder discusses, in this volume, there is little room for compromise in
this process if, as suggested by Kuhn, competing paradigms are ‘incom-
mensurable’; if the move from one world view to another involves a
‘gestalt switch’ ‘akin to ideological or religious conversion’, Hall’s (1993,
p. 280) emphasis is slightly different: ‘paradigms are by definition never
fully commensurable’. Rather, one scientific or policymaking commu-
nity replaces another. In science, this may take decades, as one genera-
tion is replaced by another. In policymaking, generational change can
be much faster (Cairney, 2013, pp. 290–1; Hall, 1993, p. 280).
The occurrence of paradigmatic policy change suggests some com-
bination of three elements: a profound shift in the context in which
policy debates take place; a profound shift in the way that policymakers
think and act; and, as a result, major policy change. There is general
agreement that it is rare. Yet, we need to be careful about at least two
aspects.
First, the literature tends to distinguish between a decision to make
policy differently and the longer term outcomes of those decisions.
Policy change can be major by one measure and minor by another. So,
can we identify paradigmatic policy change from the point of initial
decision to do things differently, or only after we measure its long-term
effect?
Second, this is a specific kind of major policy change, measured or char-
acterized in terms of the extent to which new policy, and policymaking
arrangements, diverged from, or contradicted, the old. Paradigmatic pol-
icy change suggests that a complete rethink by policymakers produced
a complete shift in policy direction. It may also imply a ‘big bang’ or
event which signals that clear shift. This is not the same as, for example,
a gradual and cumulative shift in a new direction, or major change in
terms of a massive new commitment to the same direction (the latter
was often signposted by Lindblom [1964, p. 157] as a possibility).
More recent studies consider whether we can talk meaningfully of
paradigmatic change in the absence of a big bang or event (Béland &
Cox, 2010; Blyth, 2002, p. 7; Goetz & Howlett, 2012; Hay & Wincott,
1998; Schmidt, 2010). There are also numerous terms – including ‘grad-
ual change with transformative results’ (Streeck & Thelen, 2005, p. 9),
‘punctuated evolution’ (Hay, 2002, p. 163), ‘gradual but profound’ third
order change unaccompanied by crisis (Palier, 2005, p. 129), and ‘phased
transition towards paradigm change’ (Studlar & Cairney, 2014) – which
identify a profound shift in institutions (or perhaps organizations),
beliefs, and policy over such a long period of time that it is difficult to
talk of change linked to one event or a short era. Such discussions chal-
lenge the idea that paradigm change necessarily involves a profound
and quick burst of change in institutions or policy. Yet, the problem is
that no study is clear on how long is long or how quick is quick.
These issues are not unique to Hall’s work. They can also be identified
in studies, such as punctuated equilibrium theory and multiple streams
analysis, grouped under the term ‘evolutionary theory’ (Cairney, 2013).
For example, similar questions have been raised about the meaning,
categorization, and measurement of policy punctuations (see John &
Bevan, 2012, part of a special issue on punctuated equilibrium in Policy
Studies Journal, 40, 1). There is also much disagreement about how to
conceptualize evolutionary change (is it gradual or punctuated?) but, in
that context, remarkably high agreement about how long ‘evolutionary’
policy change takes – largely because key authors identify the potential
for huge variation. For example, Kingdon’s (1984, pp. 122–36) range is
‘a while’ to ‘a few years’ to ‘twenty-five years’ (see Cairney, 2013, p. 12;
Studlar & Cairney, 2014, p. 4). In more general terms, the identification
and measurement of ‘major policy change’ is still problematic (Cairney,
2012, pp. 29–30). So too is the general idea of policy ‘direction’, which
is yet another metaphor, difficult to identify and operationalize (unless
an actor performs a ‘U turn’ to return to a previous policy).
One further complication: Variation by policy issue

One thing that might set Hall’s work apart from many others is his
primary object of study – economic policy in the UK – and the extent
to which comparable approaches study the same thing. For example,
Baumgartner and Jones’ (1993) early work focused on issues such as
nuclear power, pesticides, and tobacco, while the ACF has its roots in
studies of environmental issues – and both developed from studies of
the United States. It may be that the identification of paradigms is more
straightforward in economic policy than in other policy fields, partly
because economics as a discipline is often modelled closely on the natu-
ral sciences, its professionals are often trained in particular subfields,
and it makes sense to suggest that Keynesian and monetarist economists
had incommensurable understandings about how the world worked
(Hall, 1993, p. 284).
As Hall (1993, p. 290) suggests, this kind of outcome may be a com-
mon feature of areas ‘where policymaking involves some highly techni-
cal issues and a body of specialized knowledge pertaining to them’, but
there seem to be areas – such as environmental and tobacco policy –
where a large scientific consensus develops over time, and more debate
revolves around how to frame and address the problem than define it
technically, or in terms of cause and effect. Consequently, for example,
Studlar and Cairney (2014) identify phases of change because the scien-
tific evidence on tobacco was accepted more or less in different govern-
ment departments, but there was a more gradual acceptance of the need
to change policy markedly to reflect the weight of the evidence. In other
words, the definition of the problem and production of solutions were
often very separate processes.
Overview of the Advocacy Coalition Framework
The ACF is an approach for guiding research in principal areas of coali-

tions, learning, and policy change. The framework was originally devel-
oped in the early 1980s and has continuously evolved over time through
empirical applications and updates and revisions to the framework and
its theoretical emphases. A detailed summary of the ACF is not feasible
in this chapter; instead only the concepts and theories within the ACF
are presented to inform the comparison with Hall’s paradigm and ideas.
For further descriptions of the ACF, see Jenkins-Smith et al. (2014).
As a research guide, the ACF starts from several foundational assump-
tions, three of which are relevant to the current discussion and involve
individuals, subsystems, and policies. First, the ACF assumes individuals
are bounded rational and principally guided by a belief system. This
belief system is a three-tiered mental model that includes normative/
ontological values called deep core beliefs, normative and empirical beliefs
central to a policy subsystem called policy core beliefs, and instrumental
beliefs for achieving policy core belief goals called secondary beliefs.
Sabatier and Jenkins-Smith (1999, p. 133) provide a list of potential

belief system components per belief system level including, but not lim-
ited to: (1) sociocultural identity, whose welfare counts, view of human
nature, and priorities associated with basic values for deep core beliefs;
(2) problem perceptions, general policy solutions, causal understand-
ings, priorities for distribution of authority, and priorities associated
with policy instruments that span a policy subsystem as part of policy
core beliefs; and (3) instrumental decisions or beliefs related to a subset
of a subsystem as part of secondary beliefs. To address their cognitive
constraints, individuals rely upon their belief system to filter, interpret,
understand, and distort reality as well as to shape their networks and,
hence, the structure of coalitions. For scholars seeking to apply the ACF,
the three-tiered belief system and the belief system components offer
a broad but useful guide in conducting their research and in testing
hypotheses.
Second, the principal unit of analysis for studying policy processes is
the policy subsystem. Policy subsystems are defined by a topical area, a
geographic territory, and the policy actors involved. Policy subsystems
can occur at any level of government from local to national and can
occur cross-nationally. Policy subsystems are also nested and overlap-
ping and semi-autonomous. They may involve officials from any level
of government and non-government policy actors including those from
the private sector, non-profits, academia, consulting firms, the news
media, engaged citizens, and possibly others.
Third, public policies are conceptualized as translations of belief
systems of the coalition or coalitions whose members negotiated and
formulated it. This is one of the reasons why coalition politics are so
contentious; it involves debates about translating normative and empir-
ical beliefs into practice.
Relating ACF and paradigms and ideas
Pragmatically, Hall (1993) and Sabatier and Jenkins-Smith (1993) were

describing a very similar phenomenon in part to understand behaviour
and motivation. Consequently, Hall’s definition of ideas and paradigms
looks a lot like the definitions of belief systems and policy subsystems in
the ACF. However, it is critically important to explore the differences at
a deeper level of sophistication.
The differences begin with the role of the individual. In Hall’s (1993)
work, the role of the individual in relation to ideas and paradigms is
ambiguous and the gestalt emphasis suggests that agency exists at the
collective level. As such, questions have been asked about whether ideas
exist outside of individuals (Baumgartner, 2014). In contrast, the ACF is
grounded in a modified depiction of methodological individualism; that
is, individuals have agency-given contextual constraints and opportuni-
ties. Thus, belief systems and learning exist only at the individual level.
When a phrase like a ‘coalition belief system’ is mentioned, the term
‘coalition’ is used metaphorically or for convenience because coalitions
do not have beliefs – only individuals do. Therefore, a single, coherent
belief system that operates at the coalition or subsystem level does not
exist in the ACF. This does not mean that a researcher cannot measure
or describe the belief system attributes that might pervade a coalition or
dominate a policy subsystem; indeed, this is exactly what happens when
mapping the political landscape of a policy subsystem and the agree-
ments and disagreements of belief systems aggregated from individuals
involved. However, in studying belief systems of collectives, it is recog-
nized, sometimes implicitly, that the origins of belief systems occur only
within individuals involved in the policy subsystem.
As a point of comparison, the distinction is important for at least
two reasons. First, in the ACF, for example, the relationship between
belief systems and individuals is not a conceptual or theoretical debate,
and time and energy are not spent in clarifying the relationship as is
done in trying to understand ideas and individuals (Baumgartner, 2014;
Daigneault, 2014). Second, in the ACF, agency is clear. Individuals make
change or stasis happens. Their action might be in reaction to events
or information, they often operate in tandem with their organizations,
and they are definitely affected by their contextual settings, but only
individuals have agency. In Hall’s work, or at least in the way it is often
interpreted, the point of agency is less unclear and could exist within
individuals or from the whole. The explanation for stasis or change
could relate to individuals or to ideas. Thus, the causal driver in the
ACF is individuals and their belief systems; the same causal driver in
the broader literature on ideas and paradigms is much more difficult to
identify. In Daigneault’s (2014, p. 454) the ‘bearers of ideas’ are often
neglected.
Another point of similarity and subtle differences relates to ideas and
belief systems. For Hall (1993), a common interpretation of ideas is one
of an umbrella concept that can refer to any mental construct, includ-
ing beliefs, values, and knowledge. In this broad sense, ideas resemble
the belief components of the ACF. However, a belief system is not an
umbrella concept category. It is an integrated three-tiered model with
components that are interrelated. Aspects of the ACF’s belief system
can be measured, hypotheses about the belief systems components can
be empirically tested, and belief systems can be compared. In the work

inspiring Hall, and advanced by Hall, there is no equivalent sense of
how to operationalize the concept of ideas – and this is one of the rea-
sons why ideas as a category do not offer much theoretical guidance or
insights for explaining change and stasis in policy processes.
Similarly, it appears that there are parallels between the ACF and Hall’s
work with regard to their depiction of public policies. Public policies, as
written by individuals, are depicted as translations of belief systems in
the ACF. This matches some interpretations of Hall’s work of policies as
‘revealed ideas’ (Daigneault, 2014, p. 459). Given this interpretation of
policy, however, the ACF offers an extra benefit based on its belief sys-
tem model in conceptualizing minor and major change in public policy.
Both the ACF and Hall (1993) have descriptions of policy change. For
the ACF, policy change comes in two forms. The first is minor policy
change, which is equated with changes in the secondary aspects of the
policy subsystem, such as a change in the instruments for achieving the
same goal. Given that policies are translations of beliefs, minor policy
change is associated with changes in secondary beliefs of a belief system.
Minor policy change in the ACF is equivalent to Hall’s first and second
order paradigmatic policy change. The second is major change, which is
equated with changes in the policy core aspects of the policy subsystem,
which may (but not necessarily) involve changes in goals. Major policy
change is associated with changes in the policy core beliefs of a belief
system. Major policy change in the ACF is equivalent to Hall’s third
order paradigmatic policy change, at least in definition. However, the
ACF differs in that there can be major policy change in a policy sub-
system but, given that a policy subsystem consists of multiple policies,
there might not be a complete paradigmatic makeover. In this regard,
major policy change is not the same in practice as the paradigmatic
policy change implies. For example, there can be major policy change
that radically alters the distribution of resources in a policy subsystem,
but the goals of the policy subsystem remain the same.
To compare policy change in the ACF and paradigmatic policy change
in Hall’s (1993) work meaningfully requires us to define the policy
paradigm concept and translate policy paradigms into the lexicon and
theory of the ACF. A direct translation of policy paradigms does not
exist in the ACF. One might expect that some policy subsystems are
more coherent in their policies than others, or that some policy sub-
systems might be controlled by a single dominant coalition supported
by members with very similar belief systems. The latter is the analogy
that Baumgartner used in relating policy paradigms to the policy image
purported by a policy monopoly. At the same time, this interpretation
differs, as belief systems are what binds coalitions together and are not
the subsurface, underlying, and fundamental force that shapes world
views, language, and culture, which is one of the main interpretations
of policy paradigms. In this regard, equating policy paradigms to the
shared components of the belief systems of individuals in the domi-
nant coalition undervalues the indirect and implicit gestalt power of
policy paradigms, which is possibly one of the strongest interpretations
of Hall’s argument. This is one situation where the search for similari-
ties and likeness leads to a loss of the critical nuances and details that
make a particular theory interesting and important in contributing to
our understanding of policy processes.
Putting the differences aside, assume for the sake of argument that par-
adigmatic policy change is a sweeping subsystem or system level change
of beliefs, policy, or both. Policy paradigm change could be equated with
the ACF as a changeover of one dominant coalition to another domi-
nant coalition. With the exception of revolutions, where one dominant
coalition is completely removed and replaced with a rival dominant coa-
lition, such sweeping change is rare. More likely, a dominant coalition
and its associated policy paradigm would crumble, a dominant coalition
would be replaced with two or more adversarial advocacy coalitions,
and perhaps another or the same dominant coalition would again re-
emerge. Such a process would take decades to centuries and, on smaller
scale, has been described previously by Weible (2008). If such a compari-
son were made, the lesson would be that paradigmatic policy change is
extremely rare and most likely takes decades or longer. In this sense, the
description and theoretical arguments about policy change in the ACF
are distinct from Hall’s paradigmatic policy change because of the tem-
poral differences and potential differences in magnitude.
Conclusion
The purpose of this chapter has been to provide a meaningful explora-

tion of the similarities and differences between Hall’s (1993) paradigms
and ideas and the ACF. The field can benefit from thoughtful interpre-
tations and comparisons of different approaches, which we hope was
illustrated by this chapter. The underlying rationale of this chapter is
that policy process research deals with complexity that defies the emer-
gence of universal theories of the policy process and that requires careful
conceptualization and creation of causal theories.
In this exploration, we found some similarities. For example, Hall’s
ideas and the belief components that comprise the ACF’s belief system
are similar. However, we found far more differences: the structure of

ACF’s belief system is absent from Hall’s category of ideas; ACF’s major
policy change is most distinct from paradigmatic policy change; agency
lies within individuals in the ACF but is more ambiguous for Hall (or, at
least, the literature in which Hall’s work can be situated); and the pol-
icy paradigm concept does not exist, in a meaningful way, in the ACF.
Blurring these differences is to overlook the current descriptions and
explanations for behavioural change and stasis in both approaches. The
descriptions and explanations that exist in both theories are what must
be developed if both approaches have any hope in making significant
advances to the study of policy processes.
Thus, explorations as done in this chapter should be done not just to
combine or integrate different approaches for the sake of finding com-
mon ground, but also for appreciating the differences, recognizing the
differences in vocabularies and research approaches, and developing
strategies for advancing the field. Perhaps in time some theoretical inte-
gration might occur but any such integration – if it were possible and
fruitful – absolutely must begin with deliberate and accurate attempts
at understanding the differences, which are far too often overlooked.
While the purpose of this chapter has been to explore the relationship
between the ACF, ideas, and policy paradigms, undoubtedly other inter-
pretations exist and can be put forward.
References
Axelrod, R. (1986). An Evolutionary Approach to Norms. American Political Science
Review, 80(4), 1095–111.
Baumgartner, F. (2014). Ideas, Paradigms and Confusions. Journal of European
Baumgartner, F., & Jones, B. D. (1993; 2009). Agendas and Instability in American
Politics. Chicago, IL: Chicago University Press.
Baumgartner, F., & Jones, B. D. (2009). Agendas and Instability in American Politics
(2nd ed.). Chicago, IL: Chicago University Press.
Béland, D., & Cox, R. (2010). Introduction. In D. Beland & R. Cox (Eds.), Ideas and
Politics in Social Science Research. Oxford: Oxford University Press.
Blyth, M. (2002). Great Transformations: Economic Ideas and Institutional Change in
the Twentieth Century. Cambridge: Cambridge University Press.
Cairney, P. (2012). Understanding Public Policy. Basingstoke: Palgrave Macmillan.
Cairney, P. (2013). What Is Evolutionary Theory and How Does it Inform Policy
Studies? Policy and Politics, 41(2), 279–98.
Cairney, P., & Heikkila, T. (2014). A Comparison of Theories of the Policy Process.
In P. Sabatier & C. Weible (Eds.), Theories of the Policy Process (pp. 363–90).
Boulder, CO: Westview Press.
Cairney, P., Studlar, D., & Mamudu, H. (2012). Global Tobacco Control: Power,
Policy, Governance and Transfer. Basingstoke: Palgrave Macmillan.
Daigneault, P. (2014). Reassessing the Concept of Policy Paradigm: Aligning
21(3), 453–69.
Genieys, W., & Smyrl, M. (2008). The Problem of Policy Change. In W. Genieys &
M. Smyrl (Eds.), Elites, Ideas and the Evolution of Public Policy (pp. 1–17).
Basingstoke: Palgrave Macmillan.
Goetz, K. H., & Howlett, M. (2012). Time, Temporality, and Timescapes in Politics
and Policy. In ECPR Workshop, 10–15 April, Antwerp.
Gramsci, A. (1971). Selections from Prison Notebooks. London: Lawrence and
Wishart.
Hall, P., & Taylor, R. (1996). Political Science and the Three New Institutionalisms.
Hay, C. (2002). Political Analysis. Basingstoke: Palgrave Macmillan.
Hindess, B. (1996). Discourses of Power: From Hobbes to Foucault. Oxford: Blackwell.
Jenkins-Smith, H., & Sabatier, P. (1993). The Dynamics of Policy-Oriented
Learning. In P. Sabatier & H. Jenkins-Smith (Eds.), Policy Change and Learning:
An Advocacy Coalition Approach (pp. 41–56). Boulder, CO: Westview Press.
Jenkins-Smith, H., Nohrstedt, D., Weible, C., & Sabatier, P. (2014). The Advocacy
Coalition Framework: Foundations, Evolution, and Ongoing Research. In P.
Sabatier & C. Weible (Eds.), Theories of the Policy Process (pp. 183–224). Boulder,
CO: Westview Press.
Jobert, B., & Muller, P. (1987). L’Etat en Action: Politiques Publiques et Corporatismes.
Paris: Presses Universitaires de France.
John, P. (2003). Is There Life After Policy Streams, Advocacy Coalitions, and
Punctuations: Using Evolutionary Theory to Explain Policy Change? The Policy
Studies Journal, 31(4), 481–98.
John, P., & Bevan, S. (2012). What Are Policy Punctuations? Large Changes in the
Legislative Agenda of the UK Government, 1911–2008. Policy Studies Journal,
40(1), 89–107.
Jones, B., & Baumgartner, F. (2005). The Politics of Attention. Chicago, IL: University
of Chicago Press.
Kettell, S., & Cairney, P. (2010). Taking the Power of Ideas Seriously: The Case of
the 2008 Human Fertilisation and Embryology Bill. Policy Studies, 31(3),
301–17.
Kingdon, J. (1984). Agendas, Alternatives and Public Policies. New York: Harper
Collins.
Kingdon, J. (1995). Agendas, Alternatives and Public Policies (2nd ed.). New York:
Harper Collins.
Chicago Press.
Lindblom, C. (1964). Contexts for Change and Strategy. Public Administration
Review, 24(3), 157–8.
Lukes, S. (2005). Power: A Radical View (2nd ed.). Basingstoke: Palgrave Macmillan.
Majone, G. (1989). Evidence, Argument and Persuasion in the Policy Process. New
March, J., & Olsen, J. (2006). Elaborating the ‘New Institutionalism’. In
R. Rhodes, S. Binder, & B. Rockman (Eds.), The Oxford Handbook of Political
Institutions (pp. 3–22). Oxford: Oxford University Press.
McBeth, M., Jones, M., & Shanahan, E. (2014). The Narrative Policy Framework.
In P. Sabatier & C. Weible (Eds.), Theories of the Policy Process
(pp. 225–66). Boulder, CO: Westview Press.
Ostrom, E. (2007). Institutional Rational Choice. In P. Sabatier (Ed.), Theories of
the Policy Process 2. Cambridge, MA: Westview Press.
Palier, B. (2005). Ambiguous Agreement, Cumulative Change. In W. Streeck &
K. Thelen (Eds.), Beyond Continuity: Institutional Change in Advanced Political
Economies (pp. 127–44). Oxford: Oxford University Press.
Peters, B. G. (2011). Institutional Theory in Political Science: The ‘New Institutionalism’
(3rd ed.). London: Continuum.
Rhodes, R. A. W. (2006). Old Institutionalisms. In R. Rhodes, S. Binder, & B.
Rockman (Eds.), The Oxford Handbook of Political Institutions (pp. 90–110).
Oxford: Oxford University Press.
Richardson, J. J. (2000). Government, Interest Groups and Policy Change. Political
Studies, 48(5), 1006–25.
Sabatier, P. A. (1988). An Advocacy Coalition Model of Policy Change and the
Sabatier, P. A., & Jenkins-Smith, H. (Eds.) (1993). Policy Change and Learning:
An Advocacy Coalition Approach. Boulder, CO: Westview Press.
Sabatier, P. A., & Jenkins-Smith, H. (1999). The Advocacy Coalition Framework:
An Assessment. In P. Sabatier & H. Jenkins-Smith (Eds.), Theories of the Policy
Process (pp. 117–68). Boulder, CO: Westview Press.
Schmidt, V. (2010). Taking Ideas and Discourse Seriously: Explaining Change
through Discursive Institutionalism as the Fourth ‘New Institutionalism’.
Schneider, A., Ingram, H., & deLeon, P. (2014). Democratic Policy Design: Social
Construction of Target Populations. In P. Sabatier & C. Weible (Eds.), Theories of
the Policy Process (pp. 105–50). Boulder, CO: Westview Press.
Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond Continuity:
University Press.
Studlar, D., & Cairney, P. (2014). Conceptualizing Punctuated and Non-
punctuated Policy Change: Tobacco Control in Comparative Perspective.
International Review of Administrative Sciences, 80(3), 513–31.
Weible, C. M. (2008). Expert-Based Information and Policy Subsystems:
A Review and Synthesis. Policy Studies Journal, 36(4), 615–35.
6
Paradigm Construction and the
Politics of Policy Anomalies
Matt Wilder and Michael Howlett
Introduction: Ideas and policy-making
The concept of ‘paradigmatic’ policy-making was introduced in the early

1990s by figures such as Jenson (1989), Hall (1993) and Campbell (1998)
and continues to be the most well-known method for reconciling ideas
with interests and institutions in the description and understanding of
policy processes (Baumgartner, 2013; Berman, 2013). Hall’s (1990, 1993)
concept of policy paradigms, in particular, remains the most influential
and widely used means of theoretically integrating ideational influence
with other well-known determinants of policy behaviour (see Baker,
2013; Hodson & Mabbett, 2009; Larsen & Andersen, 2009; Orenstein,
2013; Skogstad, 1998). In this approach, policy-makers operate within
the parameters of more or less fixed sets of ideas—‘paradigms’—and
change policies only in so far as discrepancies between policy ideas and
realities—‘anomalies’—appear forcing them to reconsider their ideas.
Policy anomalies are informational signals that must be interpreted
by human agents. They do so with the help of cognitive filters or ‘image
frames’ that help them organize information into reasonably coherent
patterns (Jones & Baumgartner, 2005; Schön & Rein, 1994). Thus, while
anomalies contribute to the perception and definition of policy prob-
lems, they are not synonymous with problems but rather present images
necessary to understand them as such. Anomalies in this sense are a
metric used to assess the degree to which reality conforms to expecta-
tions, the interpretation of which assists in labelling or diagnosing prob-
lems. Policy anomalies are therefore in many ways analogous to medical
symptoms, the monitoring of which lends practitioners insight into
which treatments may or may not be working, which treatments may be
producing unwanted side effects, and which treatments may be creating
101
more problems than they are solving. As is the case in the practice of
medicine, such diagnoses are not straightforward, as symptoms may be
linked to many plausible causes or theories of causation.
Anomalies are especially significant in the policy world given their
links to theories of policy paradigms. These theories, most notably that
of Hall (1993), link the nature of policy change to changes in policy par-
adigms or the overall set of cognitive and normative beliefs that policy-
makers and the public hold about public problems and their solutions.
‘Minor’ intra-paradigmatic change within an existing paradigm is thus
seen as being separate from, but related to, ‘major’ or transformative
inter-paradigmatic change.
Despite the utility and intuitive appeal of this approach to under-
standing paradigm formation and policy change, Hall’s framework has
not been without its conceptual detractors (Daigneault, 2014). Campbell
(2002) has, for example, argued that the model pays insufficient atten-
tion to the political dimensions of paradigm formation and dissolution,
while others have challenged more specific aspects of the framework
such as the three-part disaggregation of public policy that Hall devel-
oped to distinguish major from minor change (Howlett & Cashore,
2007). Moreover, in what have been the only replications of Hall’s
(1990, 1993) study of 20th-century British monetary policy, Hay (2001)
and Oliver and Pemberton (2004) arrived at considerably different con-
clusions than Hall concerning the nature and process of policy change
and the roles played in it by paradigms and anomalies. Together these
conceptual and empirical critiques have led some to question whether
policy development is ever grounded within a dominant paradigmatic
frame at all (Schmidt, 2011).
Building on this research, the purpose of this chapter is to propose
a new method for explaining how ideational paradigms develop and
impact the policy-making process. Following Campbell (1997) and
Carstensen (2011a, 2011b), this method does not consider policy-makers
to be only strategic thinkers and technical problem-solvers, but also
‘policy bricoleurs’ whose behaviour often serves to stretch the param-
eters of the dominant ideational frame. This image of policy-makers
assumes that knowledge construction proceeds as policy-makers inter-
act in the attempt to reconcile policy means and ends in the pursuit of
their goals. Importantly, this is often a highly political process wherein
actors seek strategic advantage by bending interpretations of reality to
suit their needs (Johnson, 2012; Lejano & Leong, 2012).
This brings our focus to the hermeneutics of policy anomalies—that
is, upon the interpretive struggle that surrounds policy-makers as they
Matt Wilder and Michael Howlett 103
both identify and react to discrepancies between policy expectations

and outcomes. Rather than being entirely constrained by institutional
rules, political interest, or rigid paradigmatic frames, this model sug-
gests that policy behaviour often takes the form of the development
of ‘contingent strategies’ that operate around the edges of a dominant
paradigm (Lindblom, 1959; Ostrom, 1990). Anomalous observations
may allow dominant ideational frames to be stretched, promoting inno-
vative bricolage which brings solutions previously deemed not worthy
of consideration into the ambit of acceptable ideas (Carstensen, 2011a).
Thus, paradigm change need not be sudden, episodic and coincide with
a change in authority with respect to who is designated to monitor pol-
icy adherence to the operative ideational frame, as Hall (1990, 1993)
had argued, but may also proceed in a much more incremental and
disorderly fashion without any change in the locus of decision-making
authority necessarily accompanying transformative change (Coleman,
Skogstad, & Atkinson, 1996).
Hall’s model and its vicissitudes: Conceptual problems

in orthodox paradigm change models
Hall’s model of ideational politics and policy paradigms was derived

from his adaptation of Kuhn’s (1962) socio-historical theory of the
development of scientific paradigms to policy circumstances. Kuhn’s
Structure of Scientific Revolutions provided a framework for addressing
many of questions about policy dynamics by constructing a way to
describe how empirical observations of reality in elite groups are con-
tingent upon the outcome of internal struggles over the construction of
concepts and the accuracy of those observations (Kuhn, 1962, pp. 92–4;
Latour & Woolgar, 1986; Stone, 1989). Hall seized upon the many paral-
lels between scientific and policy discourses, situating his discussion of
policy paradigms in the terms of Heclo’s earlier questions about the rela-
tive importance of ‘puzzling’ versus ‘powering’ in policy development
(Hall, 1993; Heclo, 1974, pp. 305–6; cf. Evans, Rueschemeyer, & Skocpol,
1985; Krasner, 1984).
Kuhn’s initial 1962 conceptualization of paradigms was the one main-
tained by Hall when the latter translated the notion of a scientific para-
digm into terms compatible with the policy sciences. As he put it:

standards that specifies not only the goals of policy and the instru-
ments that can be used to attain them, but also the very nature of
the problems they are meant to be addressing. Like a Gestalt, this

scrutiny as a whole. (Hall, 1993, p. 279)
In this view, policy problems, and therefore problem solutions, are

observed through perceptual frames that are paradigm-contingent and
paradigm-exclusive. Hence, as long as the definition of a given problem
and its solution remains constant, policies will be stable, with major
policy change requiring a change in the paradigmatic frame which
allows new problem definitions to be framed and new solutions to be
articulated. This is because, following Kuhn (1962), competing policy
paradigms are thought to be incommensurable in the sense that two con-
tradictory positions cannot be held simultaneously by policy-makers. As
a consequence, paradigms cannot be meaningfully compared, nor can
they be combined without a concomitant change in how the elements
within one or both are interpreted.
Hall (1993) considered changes to existing policy instruments and
instrument settings to occur without violating paradigmatic incommen-
surability. In other words, such changes were considered by Hall (1993)
to take place ‘intra-paradigmatically’, that is, without disrupting the
ideational parameters of the paradigmatic frame. As such, these ‘first’
and ‘second order’ changes were considered by Hall to be constitutive
of incremental policy-making, a style of policy-making that, according
to Hall, is characterized by intra-paradigmatic dynamics (Hall, 1993,
p. 277; cf. Howlett & Migone, 2011; Lindblom, 1959). While incremen-
tal first and second order changes (change within a paradigm) routinely
occurred endogenously, wholesale paradigm change was thought almost
invariably to be prompted by exogenous events that destabilized the
existing consensus (Sabatier, 1988).
A major type exogenous event identified by Hall (1993) was what Kuhn
(1962, pp. 62–7) termed ‘anomalies’. These are discrepancies between
expected and actual outcomes of scientific or policy experimentation
which cast doubt on the veracity of problem diagnoses and the efficacy
of proposed solutions. Anomalies constitute observations of reality that
fall outside of a given paradigmatic frame, and thus require paradigm
modification in order to be accommodated.
A second type of exogenous event is one in which the community of
decision-makers itself changes (Merton, 1979). This second type of exog-
enous event was linked by Kuhn to the first in the sense that anomalies
were seen to discredit and delegitimize scientific elites. Similar to the

dynamic theory of ‘punctuated equilibrium’ (Jones & Baumgartner,
2005), delegitimization of the dominant image frame was considered
by Hall to be integral to the ability of proponents of new ideas to attain
positions of influence.
As shall be discussed below, while according to Kuhn (1962) the sim-
ple existence of anomalies rendered existing paradigms delegitimate,
in the policy world this is not necessarily the case. Rather, paradigm
stretching may be undertaken to retain or restore trust in existing strate-
gies for action (Mondou, Skogstad, & Houle, 2014).
Problems in the empirical basis of current

paradigm-based models
In their replication of Hall’s study, Oliver and Pemberton (2004) were

less concerned with developing or refining Kuhnian analogies and more
interested in assessing the accuracy of Hall’s paradigm model given the
empirical record of British economic policy development over the past
80 years. Despite this empirical emphasis, their findings were very sig-
nificant for the theoretical understanding of policy paradigms, leading
to several overt as well as some tacit suggestions for how Hall’s theory
might be modified to fit more closely with the observed reality of change.
While Oliver and Pemberton (2004) did not refute the claim that para-
digms exist, they significantly departed from Kuhnian assumptions by
arguing that the evidence from the British case did not display incom-
mensurability, but rather illustrated how paradigms overlap even in
such theory-driven or ‘idea-heavy’ areas such as macroeconomic policy-
making (Phillips, 1975). Observing that more than one paradigm often
coexisted, Oliver and Pemberton found that ‘partial’ acceptance and
rejection of new policy goals was a common outcome and emphasized the
centrality of political and administrative ‘battles’ that take place around
the institutionalization of new paradigmatic frameworks (Oliver &
Pemberton, 2004, p. 419). This is anathema to the idea put forward by
Hall that paradigmatic policy shifts are tantamount to policy ‘gestalt-
switches’ which occur relatively quickly and unproblematically (Hall,
1990, p. 59; 1993, p. 279).
Related to this abandonment of the incommensurability thesis,
paradigm change was also not considered by Oliver and Pemberton
(2004) to be absolute, as it was for Kuhn (1962) and Hall (1993). While
Hall viewed third order or paradigmatic change to encompass changes
to all lower-order elements of policy, Oliver and Pemberton’s (2004)
notion of iterative intra-paradigmatic cycling suggested that changes

to paradigmatic goals could occur independently of changes to instru-
ments and their settings (cf. Baker, 2013; Hay, 2001; Howlett, 2009).
Beyond this, doubt was cast on the assumption that first and second
order changes—changes to settings and instruments—necessarily occur
within the confines of a given paradigm. Indeed, more conventional
understandings of incremental policy-making posit the opposite, with
marginal policy changes often extending beyond the realm of the
familiar (Lindblom, 1959).
These revised findings from the UK economic policy case significantly
altered how the process of paradigm change is to be understood. Among
other possibilities, they also raise the possibility of a slight rather than
wholesale paradigm shift in instances in which there is no clear win-
ner in the politico-administrative struggles Oliver and Pemberton (2004)
found to be critical to the consolidation of a new policy paradigm (see
also Baker, 2013; Coleman et al., 1996). This is apparent in their find-
ing that ‘the evidence of our analysis is that such change can result
in alterations to the prevailing framework of policy that, while insuf-
ficient to justify the term “paradigm shift”, are certainly much more
significant than the “second order change” of Hall’s typology’ (Oliver &
Pemberton, 2004, p. 436; cf. Skogstad, 2005).
Oliver and Pemberton’s (2004) modifications to Hall’s model are
depicted in box 6 and path B in Figure 6.1. These amendments repre-
sent a considerable departure from Hall (1993) since the processes lead-
ing to paradigm change following ‘third order’ anomalies are not linear
or deterministic as they were for Hall, but are rather dependent upon
experts’ adherence to paradigmatically pure renditions of an alternative
paradigm (see also Surel, 2000, pp. 508–9). They allow for significant
policy change to be accomplished internally, so that even when new
paradigmatic ideas are developed elsewhere, these ideas need not be
imposed from without by a shift in the locus decision-making author-
ity or a change of decision-making venue (Baumgartner & Jones, 1991)
but can result from internal rethinking and re-designation of policy
aims and mechanisms. Whereas, following Kuhn’s (1962) lead, alterna-
tives are paradigmatically bound and constrained for Hall (1993), this
is not the case for Oliver and Pemberton (2004), who instead argue
that ideational bargaining is an important characteristic of contests to
institutionalize alternative policy trajectories. The potential for defeat
in discursive battles to institutionalize ideas also raises the possibility
that some or most elements of the existing paradigm may be retained
and not replaced, commonly yielding a what Kay (2007) has termed
Figure 6.1 Oliver and Pemberton’s evolutionary iterative framework

Source: Adapted from Oliver and Pemberton (2004).
paradigmatic synthesis, or a mix of policy ideas, many of which may be

logically contradictory.
While Oliver and Pemberton (2004) were convincing in their critique
of existing theory, they were almost entirely silent on how their amend-
ments might be reintegrated into a theory of policy change. For this,
we require means for understanding the conditions in which anomalies
may be dealt with in ways that promote the maintenance of paradig-
matic image frames, and the conditions in which anomalies warrant
experimentation with ideas that exist beyond the range of paradigmati-
cally acceptable solutions.
In policy settings, maintaining the integrity of the dominant image

frame is dependent on vigilant monitoring of policy behaviour. When
proposals involve solutions that extend beyond those tolerated by the
dominant image frame, we may speak of frame defection. Frame defec-
tion may be regulated in various ways depending on the structure of
the policy subsystem. In hierarchical subsystems where action may only
proceed with prior approval, ideas that extend beyond the pale of the
image frame may simply be rejected. In subsystems in which policy
actions are only regulated ex post (including most market behaviour and
many day-to-day functions of civil servants), those charged with the
responsibility to monitor may choose among options to reward, punish,
tolerate or otherwise forgive frame defection. When frame defection is
allowed, the policy image is extended to incorporate solutions previ-
ously deemed inappropriate. This will often have consequences for the
paradigmatic integrity of the dominant image frame.
As shown in Figure 6.2, the likelihood that policy solutions outside
the range of acceptability will be tolerated is much greater if the sever-
ity of the defection is low. In other words, low order defections that
only marginally contradict established policy will often be tolerated in
instances in which frame adherence has failed or proven intractable. So
long as the circumstances that led to their toleration are not determined
to be exceptional, when these contingent strategies produce successful
outcomes or otherwise resolve anomalies, the image frame will undergo
a slight shift rather than wholesale reorientation. If the ideas undergird-
ing an image frame are paradigmatic, the incorporation of the contin-
gent strategy serves to stretch the paradigm, drawing what were once
considered to be incommensurable alternatives closer towards the main-
stream. The reoccurrence of such episodes may yield what amounts to
paradigm replacement, but one arrived at by way of incremental adjust-
ment of the image frame, rather than the sort of upheaval characteristic
of a paradigmatic revolution (Carstensen, 2011b; Coleman et al., 1996).
For this reason, those with authority to monitor the type of experi-
mentation allowed within a given policy domain serve important
gatekeeping functions. While some of these actors have been dubbed
‘guardians’ of paradigmatic tenets in the literature (Orenstein, 2013),
‘gatekeepers’ better captures how these actors attempt to control the rel-
ative paradigmatic purity of policy ideas during processes of (re)formu-
lation (cf. McNamara, 1998). Consistent with Kuhn, such gatekeepers
should be considered to be politically motivated, but since gatekeep-
ers’ preferences may be determined by any number of considerations,
Figure 6.2 Contingent strategies and frame extension
including norm-adherence and the pursuit of policy preferences, their

interests should not be assumed a priori.
Rather, what is deterministic about the exercise of gatekeeping relates to
the exclusivity of the operative image frame, on the one hand, and who
is responsible for gatekeeping functions, on the other. In subsystems in
which membership is stable and hierarchically organized, frame defection
will seldom be attempted and rarely tolerated, especially when the actors
therein share common principled beliefs and cognitive understandings
of the policy area (Haas, 1992). In self-monitoring systems with rotat-
ing membership, actors who consider their interests to be suboptimally
served by the dominant image frame will be presented with opportuni-
ties to test the parameters of the dominant frame by engaging in frame
defection or by encouraging a culture of tolerance for minor defections.
The latter may be a key element in strategies that rely on salami tactics
to affect a very gradual change in policy ideas, even paradigmatic ones.
Outstanding research questions in the new

paradigmatic theory
The theory and findings discussed so far have important implications

for the understanding of the nature and drivers of paradigmatic change
processes in the policy sciences. The most significant consequence of
these revisions to Hall’s work is that policy change, especially paradig-
matic policy change, may be much less sudden and all-encompassing
than originally surmised, reflecting instead a more gradual, hermeneu-
tic and discourse-intensive activity (Gadamer, 1989; Lejano & Leong,
2012; Ricoeur, 1981). Understanding why this is the case and what role
anomalies play in this process are matters requiring further investiga-
tion and analysis.
Existing theory suggests two separate contexts in which the delib-
eration of policy problems and the articulation of their solutions take
place. One scenario, envisaged by scholars writing in the ‘advocacy
coalitions’ vein, posits that subsystemic actors engage in policy delib-
eration with policy ideas already well defined. As a consequence, the
advocacy coalition framework (ACF) suggests that the policy core beliefs
held by coalition actors will be rather resilient to change and impervi-
ous to compromise absent the exogenous influence of mandated policy
brokers (Sabatier, 1988). In a second scenario, while policy actors may
be resilient to alternative ideas at the outset, they may be ‘softened’ over
time, allowing for the recombination of what were previously consid-
ered to be incompatible alternatives (Kingdon, 1984).
In the first instance, policy brokers are able to force compromise
between crystallized coalitions in the construction of new image frames.
In the second, image construction is a conscious and voluntary exercise
necessary for successful policy entrepreneurialism. These two scenarios
are contrasted in Figure 6.3.
In instances in which unified and united coalitions of actors pos-
sess well-articulated and clearly demarcated ideas about what policies
are desirable, the struggle to authoritatively define policy anomalies
will take place between coalitions. This scenario has been the image of
policy-making most frequently cited by those interested in policy para-
digms since it implies reluctance on the part of coalitions to compromise
on their beliefs (and is as such consistent with the incommensurability
thesis). However, whether compromise is achieved earlier on in the
process, as per Kingdon’s (1984) theory of entrepreneurial coupling,
or more forcefully mandated by exogenous brokers with some meas-
ure of delegated authority, as per Sabatier’s (1988) understanding of
Figure 6.3 Two processes of problem definition and solution construction
brokerage in the ACF, the likelihood that paradigmatic ideas will with-
stand the formulation process is dependent upon a preference on the
part of authoritative actors to see policy conform to a given policy
paradigm over all others. Contexts in which there are multiple vetoes
should therefore be considered especially likely to produce compro-
mises, withering the paradigmatic purity of policy ideas in the process
(Thomas, 2001).
Conclusion: Paradigmatic change and the hermeneutics

of policy anomalies
Hall (1990, 1993) can be credited with moving the understanding of

policy dynamics and the role played by policy ideas a great distance for-
ward. His development of the policy paradigms concept brought with it
several major contributions to the understanding of policy processes and
the role of ideas, actors and institutions in processes of policy change.
Rather than pitting ideational variables against material or institutional
factors as previous work had tended to do (Schulman, 1988), Hall’s
framework helped to explain problems that had plagued policy scholars
for decades, such as how policy could be both change-oriented and con-
tinuous as well as both incremental and revolutionary (Durant & Diehl,
1989). This work was furthermore methodologically groundbreaking in
its ability to render the constituent elements of policy more transparent
by linking them to specific types of policy change.
In spite of these achievements, the areas in which Hall’s framework

gained its greatest heuristic purchase are also those in which subsequent
studies showed that it lacked explanatory specificity (Blyth, 2013).
Interestingly, though subsequent studies replicating Hall’s analysis of
long-term change in British macroeconomic policy have brought these
concerns to the fore, there have been few calls for a reformulation of the
policy paradigms concept (cf. Daigneault, 2014; Schmidt, 2011).
In addressing this issue, the overall argument of this chapter has been
that significant policy change is often much less logical than has tra-
ditionally been assumed. Bringing the study of the manner in which
policy-makers interpret policy problems to the theoretical forefront,
we have argued that actor perceptions of policy anomalies serve as
important variables affecting both paradigmatic and inter-paradigmatic
changes. Despite the literature on policy paradigms paying little ana-
lytical attention thus far to the latter outcome, the hermeneutic contest
over the definition of policy anomalies is important to understanding
the confluence and interaction of ideational and political factors in the
policy process. While it is well established that decision-makers use
political influence to structure policy outcomes, bringing about signifi-
cant change or ensuring policy continuity, we have suggested that these
actors control the policy process by behaving as ‘bricoleurs’ and idea-
tional gatekeepers. While gatekeeping and bricolage is often a means of
interpreting signals to conform to personal preferences or complement
cognitive schemas, instances in which such behaviour tolerates ideas
outside of the dominant image frame will often result in a lack of para-
digmatic coherence.
Accordingly, what has been perhaps the most useful, though an equally
troublesome, observation made by those working with the concept of
policy paradigms has been that the destruction of one paradigm often
does not lead to the kind of certain change implied by the very concept
of policy paradigms (Béland, 2007; Feindt & Flynn, 2009; Kay, 2007; van
der Heijden, 2011). As a consequence, the discipline has become divided
between those wishing to articulate a softer image of policy paradigms
and those wishing to resituate the concept according to original criteria
laid out by Hall (Daigneault, 2014; Princen & ’t Hart, 2014). In either
case, it would seem that in many of the case studies purporting para-
digmatic policy change, the extent to which the underlying ideas and
image frames qualify as ‘paradigmatic’ has been exaggerated. Developing
an understanding of the possibilities and limitations governing how
disparate ideas may congeal under a single policy frame will therefore
be one of the challenges to moving the theory of policy ideas forward.
Tools for accounting for the interpretation, contestation and strategic

use of policy anomalies should prove useful.
References
Baker, A. (2013). The Gradual Transformation? The Incremental Dynamics of
Macroprudential Regulation. Regulation & Governance, 7(4), 417–34.
Baumgartner, F., & Jones, B. (1991). Agenda Dynamics and Policy Subsystems.
The Journal of Politics, 53(4), 1044–74.
Berman, S. (2013). Ideational Theorizing in the Social Sciences since ‘Policy
Paradigms, Social Learning, and the State’. Governance, 26(2), 217–37.
Campbell, J. (1997). Mechanisms of Evolutionary Change in Economic
Governance: Interaction, Interpretation and Bricolage. In L. Magnusson &
J. Ottosson (Eds.), Evolutionary Economics and Path Dependence (pp. 10–32).
London: Edward Elgar.
Campbell, J. (1998). Institutional Analysis and the Role of Ideas in Political
Campbell, J. (2002). Ideas, Politics and Public Policy. Annual Review of Sociology,
28(1), 21–38.
Carstensen, M. (2011a). Paradigm Man vs. the Bricoleur: Bricolage as an
Review, 3(1), 147–67.
Carstensen, M. (2011b). Ideas Are Not as Stable as Political Scientists Want Them
to be: A Theory of Incremental Ideational Change. Political Studies, 59(3),
596–615.
Understanding Thermostatic Policy Dynamics in Pacific Northwest Forestry.
Coleman, W., Skogstad, G., & Atkinson, M. (1996). Paradigm Shifts and Policy
Networks: Cumulative Change in Agriculture. Journal of Public Policy, 16(3),
273–302.
Daigneault, P.-M. (2014). Reassessing the Concept of Policy Paradigm: Aligning
21(3), 453–69.
Durant, R., & Diehl, P. (1989). Agendas, Alternatives, and Public Policy: Lessons
from the U.S. Foreign Policy Arena. Journal of Public Policy, 9(2), 179–205.
Evans, P., Rueschemeyer, D., & Skocpol, T. (1985). On the Road to a More
Adequate Understanding of the State. In P. Evans, D. Rueschemeyer, &
T. Skocpol (Eds.), Bringing the State Back In (pp. 347–66). Cambridge: Cambridge
University Press.
Feindt, P., & Flynn, A. (2009). Policy Stretching and Institutional Layering: British
Food Policy between Security, Safety, Quality, Health and Climate Change.
British Politics, 4(3), 386–414.
Gadamer, H. (1989). Truth and Method. New York: Crossroad.

Haas, P. (1992). Introduction: Epistemic Communities and International Policy
Coordination. International Organization, 46(1), 1–35.
Hall, P. (1990). Policy Paradigms, Experts and the State: The Case of Macro-
economic Policy Making in Britain. In S. Brooks & A. Gagnon (Eds.), Social
Scientists, Policy and the State (pp. 53–78). New York: Praeger.
Hay, C. (2001). The ‘Crisis’ of Keynesianism and the Rise of Neoliberalism in
Britain: An Ideational Institutionalist Approach. In J. Campbell & O. Pedersen
(Eds.), The Rise of Neoliberalism and Institutional Analysis (pp. 193–218).
Heclo, H. (1974). Modern Social Politics in Britain and Sweden. Yale, MA: Yale
University Press.
Hodson, D., & Mabbett, D. (2009). UK Economic Policy and the Global Financial
Crisis. Journal of Common Market Studies, 47(5), 1041–61.
Howlett, M. (2009). The Dependent Variable Problem in the Study of Policy
Change: Understanding Policy Change as a Methodological Problem. Journal of
Comparative Policy Analysis: Research and Practice, 11(1), 33–46.
Dynamics: The Dependent Variable and Re-aggregation Problems in the Study
Howlett, M., & Migone, A. (2011). Charles Lindblom Is Alive and Well and Living
in Punctuated Equilibrium Land. Policy and Society, 30(1), 53–62.
Jenson, J. (1989). Paradigms and Political Discourse: Protective Legislation in
France and the United States before 1914. Canadian Journal of Political Science,
22(2), 235–58.
Johnson, C. (2012). Bricoleur and Bricolage: From Metaphor to Universal
Concept. Paragraph, 35(3), 355–72.
Jones, B., & Baumgartner, F. (2005). The Politics of Attention: How Government
Prioritizes Problems. Chicago, IL: University of Chicago Press.
Health Insurance in Australia. Australian Journal of Political Science, 42(4),
579–91.
Kingdon, J. (1984). Agendas, Alternatives, and Public Policies. Boston, MA: Little,
Brown.
Krasner, S. (1984). Approaches to the State: Alternative Conceptions and Historical
Dynamics. Comparative Politics, 16(2), 223–46.
Chicago Press.
Larsen, C. A., & Andersen, J. G. (2009). How New Economic Ideas Changed the
Danish Welfare State: The Case of Neoliberal Ideas and Highly Organized Social
Democratic Interests. Governance, 22(2), 239–61.
Latour, B., & Woolgar, S. (1986). Laboratory Life. Princeton, NJ: Princeton
University Press.
Lejano, R., & Leong, C. (2012). A Hermeneutic Approach to Explaining and
Understanding Public Controversies. Journal of Public Administration Research
and Theory, 22(4), 793–814.
Lindblom, C. (1959). The Science of ‘Muddling Through’. Public Administration

Review, 19(2), 79–88.
McNamara, K. (1998). The Currency of Ideas: Monetary Politics in the European
Union. Ithaca, NY: Cornell University Press.
Merton, R. (1979). The Sociology of Science: Theoretical and Empirical Investigations.
Chicago, IL: University of Chicago Press, 1979.
Mondou, M., Skogstad, G., & Houle, D. (2014). Policy Image Resilience,
Multidimensionality, and Policy Image Management: A Study of US Biofuel
Policy. Journal of Public Policy, 34(1), 155–80.
26(2), 259–81.
Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective
Action. Cambridge: Cambridge University Press.
Phillips, D. (1975). Paradigms and Incommensurability. Theory and Society, 2(1),
37–61.
Princen, S., &’t Hart, P. (2014). Putting Policy Paradigms in Their Place. Journal of
Ricoeur, P. (1981). Hermeneutics and the Human Sciences: Essays on Language, Action,
and Interpretation. New York: Cambridge University Press.
Role of Policy-Oriented Learning Therein. Policy Sciences, 21(2/3), 129–68.
Schmidt, V. (2011). Ideas and Discourses in Transformational Political Economic
Change in Europe. In G. Skogstad (Ed.), Policy Paradigms, Transnationalism and
Schön, D. A., & Rein, M. (1994). Frame Reflection: Toward the Resolution of
Intractable Policy Controversies. New York: Basic Books.
Schulman, P. (1988). The Politics of Ideational Policy. Journal of Politics, 50(2),
263–91.
Skogstad, G. (1998). Ideas, Paradigms and Institutions: Agricultural Exceptionalism
in the European Union and the United States. Governance, 11(4), 463–90.
Skogstad, G. (2005). The Dynamics of Institutional Transformation: The Case of
the Canadian Wheat Board. Canadian Journal of Political Science, 38(3), 529–48.
Skogstad, G., & Schmidt, V. (2011). Introduction: Policy Paradigms, Transnationalism
and Domestic Politics. In G. Skogstad (Ed.), Policy Paradigms, Transnationalism
and Domestic Politics (pp. 3–35). Toronto: University of Toronto Press.
Stone, D. (1989). Causal Stories and the Formation of Policy Agendas. Political
Science Quarterly, 104(2), 281–300.
Thomas, H. (2001). Towards a New Higher Education Law in Lithuania: Reflections
on the Process of Policy Formulation. Higher Education Policy, 14(3), 213–23.
van der Heijden, J. (2011). Institutional Layering: A Review of the Use of the
Concept. Politics, 31(1), 9–18.
7
From Policy Paradigm to Policy
Statement: A New Way to Grasp
the Role of Knowledge in the
Policymaking Process
Philippe Zittoun
Introduction
Thomas Kuhn’s (1962) concept of a paradigm, which he used to describe

the dynamics of knowledge production in the natural sciences, is par-
ticularly important for the social sciences. As is the case for many
famous concepts, the notion of a paradigm takes on a different meaning
depending on the interpreter, period, and the field to which the concept
is applied. After it was used by sociologists to examine the history of nat-
ural science, other social scientists adopted it to describe their own sci-
entific field. In sociology, for example, Robert Merton (1968) suggested
using this concept to highlight a complete set of sociological postulates,
concepts, and proposals to produce a coherent analysis.
While the concept of a paradigm was initially used to describe sci-
entific theories, it came to be used by some social scientists to describe
fields outside of the sciences. In the case of public policy, Peter Hall
(1993) used the concept of a policy paradigm to highlight the existence
of a set of beliefs and representations that structure the choice of policy
instrument and can help to understand the dynamics of policy change.
When applied to understand the turn in economic policy in France and
UK, this model was relevant and has also been used to explain other
instances of policy change.
After Hall (1993) adapted the paradigm concept to the policy field,
the concept of a policy paradigm became central to policy studies to
better understand the role of knowledge and ideas in the policy change
process. Unfortunately, the ways this concept was used in a lot of cases
of policy process were problematic. One of the main reasons for these
difficulties was that policy knowledge was considered to be equivalent
to scientific knowledge in terms of coherence, logic, and hierarchy, and,
117
furthermore, the scientific revolution and policy change are comparable.

A lot of epistemological differences were forgotten during the process of
adapting the paradigm concept from the natural sciences to policy stud-
ies, which in turn destabilizes suggested theories of policy change.
After exploring Kuhn’s (1962) concept of paradigm and Hall’s (1993)
adaption of it to the study of policy, and highlighting the problems with
this adaption, this article proposes drawing on the logic of Foucault’s
concepts of episteme and dispositif, which are more relevant for describ-
ing the role of knowledge and ideas in public policy (Foucault, 1966,
2008). The article introduces the concept of a ‘policy statement’ as a
new way to grasp the role of knowledge in the complex career (Gusfield,
1984) of a policy proposal (Zittoun, 2014).
The paradigm as a heuristic concept to grasp

natural science
The concept of a paradigm, as initially developed by Thomas Kuhn

(2012), was developed to explore the dynamic process of scientific dis-
covery. In an effort to understand the complication arising from the
adaption of the concept of a paradigm to policy studies, this section
focuses on the limitations of the concept as per Kuhn’s definition in the
broader field of natural science, to which it is epistemologically linked.
Kuhn (1962) considered the ‘paradigm’ to be a heuristic concept con-
structed to describe a specific social situation that he observed. This situ-
ation is linked to the field of science observed not simply as a social
space of knowledge making, but more specifically as some defined situ-
ation inside this field. In Kuhn’s view, the concept of ‘paradigm’ must
only be used during a period that he called ‘normal’ when science can
be considered as ‘normal.’ The concept of a ‘paradigm’ and the concept
of ‘normal science’ have ‘narrow link[s]’ and are ‘related.’
To understand when a period can be considered ‘normal’ in the field
of science, Kuhn (1962) insisted on five key aspects, using some exam-
ples he took from the history of natural science. First, in a ‘normal’
period, there is a relatively large, unique, and closed social group that
comprises the entire field and shares some common rules, including
some conditions on how to enter the group and how knowledge on spe-
cific topics is distributed. Kuhn did not believe that the ‘normal’ period
and ‘paradigm’ could be spoken of until the field became structured
and closed, nor did he consider whether there were fixed and stable
‘supporters’ of the field. To illustrate this idea, Kuhn took the exam-
ple of the ‘pre-Newton’ period on the study of optics. In this period,
Philippe Zittoun 119
he explained, because there was no common knowledge shared by a

large group – on the contrary there was limited common knowledge
between multiple groups which supported multiple theories – this
period could not be considered ‘normal.’ For him, the concept of a para-
digm could only be used after Newton and with the collective acknowl-
edgement of the existence of common principles and common tests to
confirm physical optics. In this perspective, the concept of paradigm is
‘inseparable’ from a sociological representation of the academic field in
which all the participants belong to a dominant group that shares com-
mon knowledge, principles, and recognized norms.
Second, paradigms are constructed to highlight the existence of a
coherent and ordering group of knowledge, principles, laws, and ideas.
The term ‘coherent’ here is used to emphasize the existence of a scien-
tific field in which knowledge is constructed like a house of cards with
some principles and methods on the ground floor and a lot of knowledge
logically deduced from this ground floor to build additional floors. For
Kuhn (1962), defining what constituted the paradigm was often com-
plex because, depending on the subject, it was not always comprised
of the same components. What was important for him was the idea
that the paradigm could help the researchers via its accepted principles
and methods to solve enigmas inherent to the field. In this way, a para-
digm was envisioned like a magnifying glass that helped researchers to
observe and grasp reality, but at the same time only allows a specific part
of this reality to be seen, with the rest being out of site.
Much of the time, examples helped Kuhn (1962) to define the concept
of a paradigm. In the case of optics, after Newton, Kuhn argued that
there were different successive paradigms that dominated the field. The
Newtonian paradigm began with the foundational principle that light
is a particle. But this knowledge was not just a principle. With this idea,
scientists mobilized all available methods, theories, and laws concerning
particles. In the eighteenth century, scientists shared a common belief in
these principles. In the nineteenth century, a new paradigm was devel-
oped by Young and Fresnel who argued that light was not a particle,
but rather a wave. In the natural sciences, the laws, methods, and theo-
ries used to describe waves were not the same as those for particles. It
was like another house of cards. These two examples of ‘paradigms’ and
‘scientific knowledge’ illustrate that knowledge is ordered in such a way
that logical thought informs the construction of theories and hypoth-
eses based on commonly shared principles.
The third aspect is the link between a paradigm and the concept of
an enigma. An enigma is a physical phenomenon identified and defined
by researchers as a phenomenon that is not ‘natural’ or ‘normal,’ but is

rather abnormal and surprising and can be explained by its causes. In
this case, the paradigm with all its methods, laws, and knowledge works
as an instrument to define and to solve a problem. It is very important
here to understand that, because scientists belong to the same commu-
nities and share knowledge, they also share the same methods for distin-
guishing how to decide whether an enigma exists, or whether an enigma
is solved. As a closed social group, scientists develop their own language
and their own methods of distinguishing what is true and what is not,
what constitutes an enigma, and what kinds of methods and results can
be recognized as problem solving.
Defining enigmas constitutes one of the main activities of the research-
ers that acknowledge them and present them as a risk for the paradigm.
In a certain way, an enigma is something used to test the stability of
a paradigm and its capacity to answer what scientists expect from it –
the capacity to solve every enigma. Enigmas work paradoxically as both
a challenge to and a consolidation of what the paradigm represents.
Returning to the optics example, during the nineteenth century, one
enigma related to light, that scientists identified, was the existence of
waves. In each case, the paradigm helped scientists to define and solve
each enigma. In this way, each enigma was both a challenge and also
a test for the paradigm. The paradigm became the concrete solution of
enigma which, as models or examples, can take the place of explicit rules
as the foundation for the solution in the normal science (Kuhn, 1962).1
The fourth aspect is that paradigms not only represent knowledge,
but also a dominant social group. In other words, the concept of a para-
digm considers both actors and ideas as part of the same phenomena.
As Kuhn (1962) stated, a paradigm represents all the beliefs, recognized
values, and techniques shared by members of a group at a certain period
of time. This definition of a paradigm suggests that paradigms do not
exist if knowledge is not shared by all members of the group. This idea
was complex to understand and Kuhn, in a postscript written a few years
later, explained that it is probably easier to argue that scientific com-
munities and the knowledge they share are separate phenomena, even
if they are ultimately thought to be inseparable.
Fifth and finally, the paradigm concept primarily supports a theory
of scientific revolution. Kuhn (1962) argued that a paradigm reflects
common knowledge during ‘normal’ periods, but the work done during
‘normal’ periods also transforms paradigms with the identification of
‘abnormalities’ that must be explained, ultimately leading to paradigm
changes. The first four characteristics take part in illustrating this theory.
Paradigms appear very stable because the members of the group support-
ing the paradigm share common knowledge. By examining the logical pro-
cess of constructing theories and hypotheses based on commonly shared
principles, Kuhn (1962) considered the entire knowledge production pro-
cess. In this way, Kuhn constructed a very stable characterization of the
knowledge-making process that can qualify it as ‘normal.’ This normative
characterization process helped him to argue paradigm change to be an
enigma inspired by the science he observed and one that he must solve.
To solve this enigma, Kuhn suggested taking into account the idea that
if, most of the time, an enigma is solved by the paradigm, there is some
period that he referred to as a ‘crisis’ period at which point the paradigm is
not able to solve some enigma. In Kuhn’s view, all communities have the
same rules to define when a paradigm fails. For example, one of the most
important aspects is the capacity of an experiment to predict results and to
verify them through observation or experience. The failure of a paradigm is
the failure of prediction – failure shared by an entire community.
To solve his own enigma, Kuhn (1962) suggested that the incapac-
ity to solve enigmas in the scientific community contributed to allow-
ing new paradigms to appear. Faced with this failure, he argued that
researchers look for new methods, new ideas, and new experiments to
break the spiral of failures. In this way, something like a law is obtained
with the idea that when a paradigm – as shared principles – is stable,
there are some conditions of change that come from the paradigm itself.
From scientific paradigm to policy paradigm
In 1993, Peter Hall suggested adapting the concept of paradigm to

the policy field. He followed a large movement in the social sciences
that adapted and used this concept – which appears quite heuristic –
to describe some phenomena. Probably more than any other field in
the social sciences, the adaption that he piloted was more alike to the
original concept defined by Kuhn (1962) in the sense that Hall adapted
not only the concept to represent a coherent collection of knowledge
shared by most actors, but also the theory of change suggested by Kuhn.
Adapting the concept in this way allowed him to propose a theory in
which ideas played a role in policy change.
To do this, Hall (1993) cited some similarities between the fields of
science and policy. For Hall, the concept of a paradigm was a way to
complete the ‘social learning’ theory developed by Hugh Heclo, which
explained an incremental process of policy change (Heclo, 1974; Heclo &
Wildavsky, 1974).
First, Hall (1993) argued that the policy process can be generally
linked to ‘normal’ and ‘stable’ periods in which there is marginal policy
change. He used Heclo’s (1974) social learning theory, which developed
the idea that most of the time policy change is only incremental. In this
model, policy communities dominate the field.
Second, Hall (1993) suggested that, like scientific knowledge, different
policy components can be identified and ordered in different levels. In
science there are hierarchical levels between principles and postulates,
methods, theories, and laws. Hall proposed distinguishing policy values,
principles, and instruments into three different hierarchical levels. As
later seen, distinguishing and ordering the components of a policy con-
tributes to producing an analogy with Kuhn’s (1962) theory.
Third, Hall (1993) made an analogy between enigmas in science and
problems in the policy field. While scientists must confront enigmas,
policymakers must solve problems. Following Kuhn’s (1962) proposal,
Hall suggested that the role of a paradigm is to help policymakers to find
a solution to a problem. If the paradigm is able to accomplish this, then
the period is stable and there is only marginal adaptation.
Fourth, also following Kuhn (1962), Hall (1993) considered there to be
a short link between paradigms and the social groups that support them.
In this way, paradigms and actors are inseparable. In this case, the level
is important because the group is firstly linked through the first level of
principles and values before they are attached to policy ideas or policy
instruments.
Fifth and most importantly is to build a theory of significant policy
change. The four previous characteristics that allowed Hall to describe
the ‘normal’ period, which is nothing more than the incremental model
invented by Lindblom and developed by Heclo, permit the conceptu-
alization of a ‘crisis’ period at which point a policy change is not only
incremental but also paradigmatic (Heclo, 1974; Lindblom, 1958).
Hall (1993) argued that there are three kinds of policy change that
echo the three levels of policy components. The first kind of policy
change is the most common and reflects only marginal change. Hall
suggested that in order to grasp this kind of policy change, Simon’s
(1947) model, in which actors are confronted with a problem to solve
and find a solution by choosing a satisfactory alternative, can be used.
Typically, the first type of change corresponds to a budget modification,
or normative adaptation, of a policy instrument. The second type of
policy change corresponds more to a change of policy instrument with-
out strongly modifying the goals and values with which the instrument
is associated. Replacing an instrument with another that belongs to the
same paradigm produces a more important change than the first type of
policy change discussed above. Changing, or replacing, a type of tax is an
example of this kind of policy change, which is still marginal. While the
first two types of policy change are not new in policy work, this analogy
with Kuhn’s (1962) theory allowed Hall to develop the concept of the
third type of policy change, which is the policy paradigm change. Using
the idea that a paradigm is weakened when it is incapable of solving
an enigma, thus opening space for the development of a new scientific
paradigm, Hall developed the idea that a policy paradigm becomes weak
when it cannot solve a policy problem, thus initiating a ‘crisis’ period.
To justify the adaptation process, Hall (1993) used a very relevant
example of economic policy. Drawing on the work he developed in
‘Governing the economy’ (Hall, 1986), he explained how a new mon-
etarist paradigm appeared in the United Kingdom and France in the
1980s after the Keynesian paradigm failed to solve the economic prob-
lems of the 1970s. It was true in the UK that Margaret Thatcher won the
election after arguing that classical Keynesian monetary policy failed to
solve economic crises, and it was also true in France when Mitterrand
changed orientation and developed a new paradigm in 1981 after trying
a Keynesian approach. Indubitably, this case of economic policy change
contributed to the success of the adaptation of the paradigm concept
because the model seemed to be relevant for grasping and explaining a
very important policy shift in two different countries.
The limits of adaption and the problem of the policy

paradigm concept
To use the vocabulary of Kuhn (1962), the problem with Hall’s (1993)
adaption of the paradigm concept is that there are different enigmas
that it is incapable of solving. While the adaption appears to explain
economic policy change, this example is more an exception that can-
not be generalized, nor can it be transformed into a general theory of
policy change. The main problem comes from the ontological difference
between the fields of science and policy. The difference between making
policy and producing knowledge makes Hall’s adaption of Kuhn’s model
impossible. In order to illustrate this, it is helpful to return to the five
points Hall made to justify the adaptation to see what he failed to take
into account.
First, in the academic field, there is only one scientific, identifiable
community with researchers and established rules for entry and rec-
ognition, like, for example, studying for a PhD in a specific discipline,
obtaining an academic position in a university, or publishing an article

(Latour, 1988; Merton, 1973). This social group is a closed group that
conditionally accepts members, and ideas expressed by members must
be recognized by the group. This entire social process is central and con-
tributes to the sharing of knowledge. After a paradigm is adopted by a
field, all individuals who want to become ‘scientists’ need to learn it. The
policy field is more complex. As suggested by many authors, a large num-
ber of actors contribute to defining a problem and the number of actors
who propose solutions is generally not the same (Howlett & Ramesh,
1995; Kingdon, 1995; Parsons, 2003; Peters & Pierre, 2006). There are
no stable ‘rules’ to authorize individuals to speak and there are numer-
ous examples indicating a surprising number of people who seem to be
authorized to give their opinion on a problem, or on a solution. While
the concept of a policy community shows that in some situations the
number of people who work on policy formulation is limited, the situ-
ation is not comparable with the field of science because, in most cases,
there are always some actors outside of the policy community, like depu-
ties, ministries, cabinets, lobbies, and the media, who play a role in the
process (Lindblom, 1980; Majone, 1989). In this way, the question of a
‘dominant’ group that shares the same knowledge in the academic field
cannot be compared to the policy field, even if there is a ‘dominant’
group.
Furthermore, there is rarely only one group observed in the policy-
making process. Conflict is often present not only because it is consti-
tutive of politics, which is always present in the policymaking process,
but also because, as observed in organizational studies, there are signifi-
cant conflicts inside government bureaucracies (e.g., between different
ministries or within the same ministry) and it is rare for new policies
to be managed by one government entity, whether ministry, depart-
ment, or office. Several examples from France illustrate this. First, as
observed through an analysis of the making of housing policy, there is
permanent conflict between the Ministry of Economy and the Ministry
of Housing, with significant dispute settling on the part of the first.
Inside the Ministry of Economy, the conflict between the Direction of
Treasury and the Direction of the Budget is also a permanent feature
of intra-institutional relations. Another example concerns the process
of deciding whether to ban fracking as a means of shale gas explora-
tion, in which there is also a permanent conflict – this time between the
Ministry of Economy and the Ministry of Environment. In each of these
cases, the concept of a dominant group with shared knowledge among
all members does not work.
Second, the hypothesis that you can logically order, and hierarchi-
cally structure, the components of policy from principles and values
to instruments is problematic. While there is a way to build hypoth-
eses and theories from a few principles in academic fields, the links
between values, general principles, and general instruments is very
complex. Robert Dahl and Charles Lindblom explained in their first
book, for example, that the capitalist system and the communist
system used the same instruments to fight against inflation (Dahl &
Lindblom, 1953). Research on French housing policy found that the
same instrument that was considered in the 1970s as the Trojan Horse
of the new liberal paradigm was then considered to be an instrument
to produce Keynesian policies in the 1980s (Zittoun, 2000, 2001). It
is also very interesting to note that practitioners and academics are
much divided as to whether the housing policy paradigm will change,
and if it does change, when the new paradigm will arrive. The 1977
reform, when instruments were introduced, was not considered by
some actors to be indicative of any change – relative nor paradigmatic.
Even the name of the policy paradigm and its definition were not
generally shared.
In Hall’s (1993) example of paradigmatic change in policy, it seems
easier to define the differences between a ‘Keynesian’ paradigm and a
‘monetarist’ paradigm. In this case, the main reason for this is that the
paradigm comes not from the policy field, but rather from the field of
economics. In this case, the existence of a paradigm is less problematic.
But the main question concerns whether the adaptation of the paradigm
concept from the field of natural science to the field of policy is neutral.
The case of Frederic Lordon’s (1997) study of French economic policy, in
which he found a considerable gap between economic theory and policy
ideas used in France by the Ministry of Economy, is very important. The
‘theory of competitiveness disinflation’ does not correspond with any
economic theory and sounds more like the random piecing together of
know-how rather than a coherent process.
Third, the analogy proposed by Hall (1993) between an enigma as
conceptualized by the natural sciences and a problem in policy is not so
evident. Returning to his example, his idea was to show that the decline
of the Keynesian paradigm was due to a problem that it could not solve.
In the field of science, an enigma is always something ‘new’; however,
in the case of economic policy, the problem of unemployment and infla-
tion are not new. What is new is that the old recipe that worked before
no longer functions. So, it is not the discovery of new phenomena that
challenges the theory.
More generally, in the policy process, the link between problem and
solution is more complex than it is for science. First, the link is built
not directly on the solution, but through the mediation of outcomes.
In the natural science domain, scientists create experiences to isolate
phenomena and study their reproducible effects. Through this process,
time, context, and contingency disappear to reveal a direct link. In the
policy process, however, while it is possible to study isolated experi-
ences, there are always unexpected effects depending on the context,
and different outcomes depending on time. The same policy can have
different outcomes during the same time period or in the same country,
and it is always difficult to identify stable links and transform a simple
correlation into causality.
Another difference comes from the incapacity of all actors to have the
same understanding of what constitutes a problem, or what defines suc-
cess and failure. Contrary to the field of science, problems are often not
always defined by the actors in charge of problem solving. In this way,
policy actors do not share the same perspectives on problem definition.
The struggle is precisely over defining problems. The same can be said
for success and failure in the policy field. While in the field of science
it is an accepted practice to present new scientific findings via publica-
tion in an academic journal, there is no equivalent to this in the policy
field. For example, there is an ongoing debate in France over whether
the reform of the work week to 35 hours was a success or failure. There
is no agreement between politicians, bureaucrats, experts, and scientists.
As Kuhn (1962) suggested when he spoke about the pre-Newtonian con-
ception of optics, when there is no common knowledge shared by the
entire community, it is not possible to use the concept of paradigm.
Fourth, the link between paradigms and their supporters is not so
easy to observe. Much of the time, a coalition of actors that support the
same reform can be identified, but at times the coalition does not share
the same ideology or values. Sometimes, some actors change their posi-
tion on a policy instrument. The fact that the link between values and
instruments is weak also implies that actors can agree and form a coali-
tion to support a proposal without necessarily sharing the same values.
Drawing again on the example of shale gas in France, the coalitions that
supported the law to ban the exploration of shale gas were very large
and included practitioners who did not share the same ideology, values,
or reasons to ban shale gas.
Fifth, and for all the previous points, the hypothesis that policy
paradigm change is provoked by the incapacity of an extant paradigm
to increasingly solve new enigmas does not really work. As discussed
above, actors do not share opinions about what degree of change indi-
cates a reform. For example, in France, a proposed law to give same-sex
couples the right to marry is considered by some of its promoters as a
marginal reform that does not have any additional impacts on society
nor challenges existing social values. For its opponents, the reform is
nothing more than a change of paradigm that introduces new values, a
new vision of society, and has a major impact on society.
The failure of a paradigm to solve an enigma is a major cause for scien-
tific revolution – an issue we will return to below. However, in policy, the
concept of paradigm failure is problematic. Taking into account the first
four points outlined above as to why the concept of paradigm cannot
be adapted to policy, the idea of policy paradigm failure does not work.
The examples described above contribute to creating a representation of
the policy field that is probably more akin to the scientific community
that studied pre-Newtonian optics than post-Newtonian optics. There
are no cases in which knowledge is shared by all actors or the social
group is more open – nor are the conditions upon which to determine
success or failure shared. In this case, Kuhn (1962) himself would argue
that the concept of ‘paradigm’ cannot be used. In the same vein, Popper
and Lorenz (1985), who critiqued Kuhn’s model, argued that in the field
of science, the opposition between two theories is more permanent than
argued by Kuhn. This is true not only for the social sciences, but also for
some natural sciences.
From paradigm to episteme and dispositif
The concept of a paradigm was constructed by Kuhn (1962) as a heu-

ristic concept to describe social reality in the natural sciences. Even if
this concept provokes a lot of debate between sociologists and science
historians who do not support it, it has been generally successful and
influential in the social sciences as well as in policy studies, more spe-
cifically. Its main advantage is that it highlights the role of ideas in the
policy change process. The concept’s problematic adaption to the social
sciences should not lead to the conclusion that ideas are not important.
Rather, other concepts are needed to help describe the role of ideas. Two
concepts developed by Michel Foucault, episteme and dispositif (Foucault,
1966, 2008; Zittoun, 2013a), are particularly useful. They can be used
interchangeably with the concept of paradigm, but they do not have
the same weaknesses.
Foucault’s ambition was much broader than that of Kuhn (1962)
in the sense that he did not limit his thinking to a specific field,
but rather theorized about society as a whole. Like Kuhn, Foucault

argued that there is some basic knowledge that plays a principal role in
the development of knowledge. Foucault spoke about the ‘condition of
possibilities’ for the emergence of new knowledge. In The Order of Things,
Foucault (1966) showed how there is some equivalence, comparison,
and hierarchization that is specific to a given period and to a society
and which further constitutes the basis for the emergence of new dis-
courses and knowledge. Foucault used the heuristic concept of episteme
to join together all these basis elements (Zittoun, 2013a). In this sense,
by linking some puzzling, disordered, and heterogeneous components,
the episteme is very different from the paradigm. The episteme proposes
to order things and distinguish what things are comparable from those
that are not, that which can be defined from that which cannot, and
what can be put into a category from that which cannot. The episteme
does not pretend to be as coherent as the paradigm nor is it relevant
only for building agreement. On the contrary, the episteme helps us to
break any presupposed logical and coherent link between components
to grasp the process of linking meaning and to better understand how
an episteme can more easily facilitate conflict rather than agreement.
With the concept of dispositif (device or apparatus), Foucault (1976)
delved deeper into his analysis by not only considering words, discourse,
and knowledge as heterogeneous components, but also practices, tech-
niques, objects, and institutions. Second, he argued that the concept of
dispositifs can be applied to relatively smaller situations than that of the
episteme, which is applied to society more broadly. Taking into account
these differences between episteme and dispositif, Foucault continued to
highlight the process of linking heterogeneous components, includ-
ing not just words and discourses but also ‘[i]nstitutions, architecture
planning, regulatory decisions, administrative measures, philosophical,
moral and philanthropic proposals’ (Foucault, 1976, p. 300).
With dispositifs, Foucault (1976) worked on the history of sexuality
by focusing on sexuality devices that existed at a given time to link spe-
cific disparate elements that were at the same time transforming as they
were being linked together. The dispositif reveals the intentions of actors
who construct it, which consequentially acts as ‘a machine to do see
and to do speak’ (Deleuze, 1986). Foucault took the example of impris-
onment, which he considered to be a dispositif, to highlight the way
that imprisonment brings together statements on criminality, courts,
prisons, laws, guardians, offenders, prisoners, and so on. Foucault sug-
gested focusing on the genesis and the evolution of this kind of dispositif
to better reveal both the incentives that are generated through the act
of bringing the disparate elements together during a specific period in

which the fight against criminality appeared on the agenda, and also
the unexpected effects in terms of socialization and professionalization
of delinquency. This is what Foucault called the ‘strategic filling of the
dispositif’ (Foucault, 1976, p. 300).
In more contemporary works, the concept of dispositif continues to
be regularly used and enriched. In the social sciences, Bruno Latour, for
example, developed the concept of ‘socio-technical dispositifs’ to desig-
nate a collection of technical instruments and how and with whom they
intertwine in the social world where they are made and have an impact.
Michel Callon (1986, p. 191) used the concept of an ‘incentive dispositif’
to speak about the assembling of social practices with a specific strategy
to recruit people. In this contemporary literature, there are other con-
cepts that refer to a very similar process of linking heterogeneous com-
ponents. For example, Latour (2006, p. 8) regularly used the concept of
‘association’ to designate something similar and to emphasize the pro-
cess of how these associations make, delete, and reconstitute phenome-
non in an ‘unexpected movement.’ John Law uses the term ‘network’ to
focus on this assembling of ‘heterogeneous, multiples, often unexpected
relations which link scientific knowledge, technical devices, making
unite, sellers and consumers’ (Law, 1989, p. 72). All of these concepts
further enrich the Foucaldian concept by focusing on the importance of
‘non-human’ actors and the process of how this assembling emerges and
transforms. For example, in a case study by Callon (1986), he analyzed
the problematization, interest-generating, enrollment, and translation
activities used by actors to network fishermen, some researchers, Jacques
scallops, fishing nets, and so on.
From policy paradigm to policy statement
The concept of policy statement can help explain the role of ideas in the
policy process without creating the problems provoked by the concept
of a policy paradigm. A policy statement is a discursive assemblage of dif-
ferent heterogeneous components without any presupposed coherence
that gives specific meaning to a policy proposal and is supported by some
policymakers who want to change policy (Zittoun, 2013b). The differ-
ent components of a policy statement can include a problem, solutions,
instruments, values, ideologies, goals, causality, consequences, outputs,
outcomes, the public, responsibilities, or guilt. Far from Hall’s (1993)
and Sabatier’s (1988) perspectives, which supposes a non-contextual,
preordering process between different components – from values to
instruments, the concept of a policy statement helps us to better grasp

the logical work done by policymakers to give meaning to the policy
proposal they support at a specific moment in time.
The policy statement concept is based on three hypotheses. First, the
relationship between a policy proposal and its value, the problem it must
solve, the consequence it might have, and the general policy it aims to
transform are not objective links that researchers and practitioners can
build independently and about which they can reach the same conclu-
sion, but these are rather subjective links that can be built in different
ways. Some actors can propose to link a proposal to a value and some-
one else can propose linking that same proposal to a different value. In
this way, the researcher cannot produce his own link without producing
a normative view that distinguishes the ‘true’ link from the ‘false’ link
but can observe the ways policymakers give meaning to their proposals.
As suggested by Wittgenstein, the main idea here is that the researcher is
not blinded by the content of a discourse and its meaning, but considers
this kind of assemblage as a game of language. Returning to Dahl and
Lindblom’s (1953) observation that the same instruments were used to
fight inflation in United States and in the former Soviet Union, the ques-
tion for the researcher is not to know if this policy instrument is onto-
logical or is rationally linked to a big ‘ism’ like capitalism or socialism –
and then later say that the ideology has no importance, but rather to
consider this link to be politically and socially constructed by actors, for
whom it is important that this link be studied (Dahl & Lindblom, 1953).
Second, the meaning of a policy statement is central for interaction,
communication, and appropriation. A Weberian perspective can be
integrated here that argues that behavior is always linked to the sub-
jective meaning a person gives to their action(s). To promote a policy
proposal, policymakers formulate proposals that are understood by both
themselves and those with whom they share the proposal. This second
hypothesis has two consequences that relate to the first hypothesis.
Even if a policy statement is considered to be a curious assemblage of
heterogeneous components that is not logically paired with a solution –
with the solution at times already existing even before the problem
itself emerges and the value linked to instruments sometimes existing
before the instrument itself is created, and so on – it must make sense to
promoters of the policy statement. This implies that all possible assem-
blages are not possible and a solution can only be attached to a problem
if the assemblage of the two makes sense. A second consequence is, even
if the assemblage is not chronologically logical, that promoters make a
real effort to reorder the policy proposal statement. The question is not
whether the problem or the solution, depending of context, came first,

but rather understanding how the promoter ordered the policy state-
ment, made it coherent, and developed arguments to transform the pro-
posal into a solution.
Third, on account of the first and second hypotheses, policy state-
ments are not separable from the policymakers who use them, the
moment it is employed, the audience to which it is addressed, and the
kind of activities that are mobilized as a result of presenting the policy
statement. The policy statement is not knowledge or an idea that just
floats in the air that researchers have difficulty capturing because it is
not visible. The policy statement is a discourse ‘in action’ developed by
some specific people at a specific time in a specific location. This idea
was also present in the paradigm concept when Kuhn (1962) explained
that the existence of common knowledge cannot be separated from
large numbers of people. But, in this case, the policy statement is sup-
ported by a few people who advocate the same policy proposal state-
ment. The consequence of this hypothesis is that ‘giving meaning’ to a
policy statement is not an objective activity, but always an intersubjec-
tive activity in which meaning is built between a speaker and interlocu-
tors in a contingent way.
Let us take an example to understand policy proposal statements and
these three hypotheses. This example is again taken from the case of
shale gas policymaking in France. This inquiry is based on data taken
from 20 qualitative interviews of stakeholders who participated in the
decision-making process as well as relevant documents. Four policy pro-
posal statements were identified that successively appeared in a short
period of time. The first one appeared around 2008 and stated a refusal
to distinguish nonconventional and conventional gas and oil licenses
for exploration and extraction. Two companies requested licenses from
the Ministry of Energy arguing that there was no change in policy by
asking for a shale gas and oil license and that the legal procedure can
work. This is the typical position of the office that manages licenses in
the Ministry of Energy. In this way, a shale gas license is considered to
be a continuation of energy policy.
The second policy proposal statement is more interesting. It appeared
in December 2010. Some people who lived in the vicinity for which the
State issued the license to extract shale gas joined together and began
constructing a new policy statement around a new proposal to ban shale
gas extraction in France. They developed an initial statement in their
discourse in which they acknowledged that authorizing a shale gas
license was a form of policy change. Second, they established a causal
link between a shale gas license and groundwater contamination, waste-

water pollution, landscape devastation, carbon footprints, and global
warming. Rejecting any NIMBY2 motivation, they broadened their dis-
course, risking increasing tension and contestation, by linking shale gas
to an apocalyptic future unwanted by citizens. Through this process of
linking elements of their discourse, they gave meaning to their proposal
to ban shale gas licenses as a way to preserve the future.
The third policy statement was juxtaposed against the second one.
The group that supported it also argued that authorizing a shale gas
license was a type of policy change, and they linked this proposal to its
consequences, like lowering the price of gas, making French companies
more competitive, modifying the trade balance, lowering dependency
on other countries, fighting against unemployment, and changing geo-
political tension. Through the linking of these arguments, they gave
meaning to the idea that authorizing the shale gas licenses in France is
a way to build a more prosperous future.
When these three policy statements are observed in the policy pro-
cess, it can be argued that they have continued to coexist in the policy
and political arena until now. While in science controversies between
different paradigms only exist in moments of crisis, the presence of
two or more competing policy statements in the field of policy is quite
typical, thus for policy the ‘normal’ period is characterized by crisis.
Furthermore, the three policy statements did not share a same view of
the importance of policy change provoked by the new policy. For the
first one, the change in policy was minor and for the last two the policy
change was major. This signifies that defining the importance of policy
change is also a subjective definition process.
The importance of recruiting to build a policy

statement coalition
One of the problems with the policy paradigm theory is the idea that,
during the ‘normal’ period, all scientists share the same ideas and learn
the same hypotheses and laws. In the case of the policymaking process,
there is always a diversity of approaches in politics and the process by
which to transform a proposal that is supported by very few to a pro-
posal supported by a large coalition is not an automatic and simple one.
As Olson (1985) suggested, for example, even if people have the same
interest, or share the same views or values, collective action is never
automatic and is always a complex process that fails much of the time.
When collective action exists, the black box of the coalition needs to
be opened by observing the process that transforms a small group of

actors that promote a policy proposal into a large coalition (Zittoun,
2011). The main hypothesis here is that this process is first the result of
the work of promoters who recruit actors to support and propagate the
proposal.
To grasp the process of recruiting support for the policy statement,
four kinds of activities are highlighted that contribute to encouraging,
or preventing, the propagation of a policy statement. These four activi-
ties are the product of two sets of concepts: (1) to critique and to enroll
and (2) to persuade and to convince. The first set of concepts desig-
nates which kind of incentives is supported by the actors. Recruiting
means that an actor persuades another to believe that a proposal is
‘good’ and that it should be supported. Critiquing consists of persuad-
ing another that the proposal is ‘bad.’ Drawing on the work of Perelman
and Olbrechts-Tyteca (1958), the second set of concepts helps to distin-
guish in which settings the speaker is in discussion with a small public
group and in which he communicates to a large group. Perelman and
Olbrechts-Tyteca (1958) suggested that speakers who use an argument
do not choose the argument without taking into account the kind of
audience for which the argument is best suited. They argued that the
arguments used for a large audience in public spaces are not the same
that the arguments used for a small public inside a private discussion.
He suggested abandoning the old rhetorical distinction between persua-
sion and conviction and rather adopting the concept of ‘persuasion’ in
a case of a small public and of ‘conviction’ in a case of a large public.
Mixing these two sets results in four different activities.
The first activity is to recruit an audience through conviction. This case
is similar to what Vivien Schmidt (2008, p. 308) called ‘communicative
discourse’ in which the discourse is constructed for a large audience. In
this case, the purpose of the activity is not only to present a policy state-
ment, but also to justify it by arguing on the solidity of the argument
embodied by the policy statement. For example, the actors who want
to convince others to support a proposal argue not only by explaining
which problem will be solved by the proposal, the kind of policy change
to which the proposal contributes, or the values taken into account by
the proposal, but also by developing some arguments to justify these
links. As Perelman and Olbrechts-Tyteca (1958) suggested, the argumen-
tation process is developed precisely because ‘pure’ logic does not work.
Because a solution is never deduced logically from the problem, many
times a solution exists before a problem. Actors cannot present already
existing solutions as evidence, but must justify them and answer to
other counter-arguments. Through these argumentative activities, the

promoters of a policy statement try to ‘attach’ the solution to a problem,
which is not a simple act of juxtaposing.
The second activity is to recruit the support of specific actors through
the act of persuasion. To do this, the actors promoting a policy proposal
need to act in two ways. The first is to prove to the audience that the
proposed policy statement has enough support to successfully recruit
the support of the audience through conviction and thus become a
communicative discourse. Second, the actors need to develop more spe-
cific arguments to explain why the audience should have an interest in
adopting this policy proposal. The main idea is to try to grasp the pro-
cess of propagation, which means that an actor changes their point of
view by supporting a proposal that they did not support before. It is this
change of opinion that assists in better understanding whether the suc-
cessful or failed circulation of a policy proposal and, more specifically,
how actors transform their view of their own interests.
The third and fourth activities consist of critique through convic-
tion and persuasion activities. The process of critiquing can also be a
‘communicative discourse’ (Schmidt, 2008) addressing a large public
or a more discreet discussion to a restricted audience. These activities
are supported by an argumentative process initiated against the policy
proposal statement, its linkages, and/or the interests of the audience.
‘Critical activities,’ include, for example, developing arguments to show
how a proposal is not a solution to the problem it must solve or how
the proposal does not produce the expected consequences and rather
provokes some unexpected consequence. Boltanski (2009) suggested
that critical activities, in general, must be taken very seriously in our
society, and especially in the policy process. They play a central role in
terms of testing, learning, and adjusting ideas and discourse. This means
that each policy statement that is developing in a public space is passed
through a process in which the statement and its arguments are tested.
For example, it can be regularly observed that arguments and policy
proposals move with the development of events and public debates and
adapt to critiques as promoters try to build a stable policy statement
through pragmatic adjustment.
Returning to the shale gas example, one of the most interesting aspects
of this example is the fact that no one had an opinion on this topic
before December 2010. The number of actors who knew of shale gas was
very small. In four months, this unknown topic became a major issue
with many stakeholders supporting, or opposing, the law to ban shale
gas and these two positions went across the traditional line in terms of
political parties and special interests. There was a division inside the
socialist party as well as a division inside the right-wing party. The posi-
tion of major stakeholders, like the Ministry of Environment, the prime
minister and the president of the French Republic, changed a few times.
The arguments used to promote and to critique are important in the
conviction process. Like a boxing match, as suggested by Mead (1967),
promoters and opponents are fighting to kick arguments, seeking to take
advantage of their opponent’s weaknesses, and taking shots at them
where it will hurt, while at the same time dodging their opponent’s
shots. It is indeed an interactive process in which arguments for and
arguments against interact.
One of the main arguments used in setting the agenda of the move-
ment against the shale gas ban was the case of the US and the problems
illustrated in the film Gasland. The first argument used by the Ministry
of Environment to promote its shale gas policy was ‘Dangerous tech-
niques for the environment and destructive are used in the U.S. It is
not about to engage France in this way.’3 Adapting to this new argu-
ment, those in favor of the ban explained that there was currently no
technology, other than the US technology, and only Americans wanted
to explore shale gas in France.4 The Ministry of Environment gradually
began to take into account the arguments of critics in order to adjust its
own policy statement. Facing difficulty in constructing a solid statement
that was strong enough to resist critical arguments, the Ministry of the
Environment slowly changed its position by promoting the ban two
months later. Surely, the adjustment process does not always finish with
a radical change of position, but it does always contribute to advancing
the position, policy statement, and arguments.
The redefinition of policy proposal as the cost

of recruitment
The purpose of describing the social activities discussed above was to under-
stand the complex process of constructing a policy proposal, the support
for which grows from a very small group of people to a larger coalition,
which eventually transforms the proposal into a decision. Because build-
ing collective action around a policy proposal is costly and also because
each policy proposal statement has to successfully endure critique and
opposition, it is argued here that the proposal resulting from this process –
which is typically a variation of the original – points to the role of ideas
in the policymaking process. Borrowing from Gusfield (1984), the concept
of a ‘proposal career’ assists in explaining the circumstances under which
the proposal is circulated or forgotten, which can be redefined during the

proposal’s ‘career’ as a result of the impacts of arguments and mobilization
in opposition to the proposal. Our hypothesis is that the policy proposal’s
career helps us to understand the impact of ideas, discourses, and argu-
ments, which transform a linear career into a complex one.
In the previous section, it was argued that a policy proposal is initially
supported by very few policymakers and the principal way that they
push their ideas onto the decision-making agenda is to transform their
isolated actions into collective action. Because collective action is never
an automatic or natural phenomenon, an important issue, as raised by
Crozier and Friedberg (1977), is under what conditions and at what price
collective action is possible.
The four kinds of activities that contribute to the successful propaga-
tion of a policy statement discussed above are costly for people who
support a proposal. First, they are costly in terms of time. Developing
strategies to identify people to persuade, arranging to meet these peo-
ple, and trying to recruit their support by developing convincing argu-
ments are time-consuming activities. As numerous scholars who study
the policymaking process have pointed out, a decision and a policy
change requires the support of many policymakers in order to be suc-
cessful. This is equally true for government institutions and agencies,
bureaucracies, interest groups, political parties, and so on – ideas must
have the support of other groups, and especially ‘specialists,’ in order to
be successful. Because support is not automatic, individuals often have
to engage themselves after initial discussions, and testing an idea and
recruiting support requires a lot of time and energy.
Second, these activities are costly in terms of identity. Because an idea
never exists without actors to support it, the link between the two is
complex and not neutral. As Mead (1967) pointed out, there is a com-
plex link between the identity and discourse of a speaker. In this case,
it means the link between the quality of policy and the quality of per-
sonality can be questioned. For example, in the case study of a tramway
in Paris (Zittoun, 2013c), it was clear the mayor considered it an effec-
tive strategy to emphasize the ‘modern’ and ‘ecological’ nature of the
tramway in order to transform and bolster his own image to be more
modern and environment friendly. So, the issue is not always one of
post-decision communication, but rather one of identity that was used
to persuade the mayor to support the tramway. In another sense, criti-
cizing a policy proposal as ‘not modern’ or not ‘rigorous’ is a means of
criticizing supporters of the solution.
Third, these activities are costly in terms of power and legitimation.
During the 1960s, the studies on influence and decision making focused
on identifying ‘who governs,’ ‘who decides,’ and ‘who has influence’

(Banfield, 1961; Dahl, 1965) by examining a very puzzling and very
complex distribution of power. To grasp power, they examine power
and influence ‘in action’ during the policymaking process. For example,
Dahl (1965) observed the path of ideas in urban policy, distinguished
what policy proposal arrives to become a decision, and grasped who
are at the origin of this proposal and, during the process, who support
the proposal and who are opposed to it. His idea is to grasp power of peo-
ple by its capacity to influence the decision-making process. He suggests
that important urban policy change in New Haven was generally linked
to the existence of important political man who chained his career to
this idea. This idea to link power to the success or the failure of policy
proposal and its bearer can be extended to the question of legitimation
and acknowledgment of a policymaker from others and can be consid-
ered by the first one as a risk. For a policymaker, supporting a policy
proposal is a risk. If he does not succeed in transforming the proposal
into a decision, he can be accused by others of being a policymaker who
does not have enough power and influence in the policymaking process
and his legitimacy can be challenged.
Several examples illustrate these three dimensions. Bruno Latour gives
a very interesting example that demonstrates the first point. He stud-
ied the development of a socio-technological innovation of transport
policy called ARAMIS (Latour, 1993). While this project seemed to have
many positive characteristics, its failure to become policy was explained
by Latour as a result of no one taking the time to support the idea.
Many actors argued that the idea was self-sufficient and could succeed
on its own, thus it was not prioritized. This idea was also echoed in
the study of the Paris tramway referenced above, in which the decision-
making process of the tramway was discussed (Zittoun, 2013c). While
some actors supported the idea for the tramway, they were largely on
their own at the beginning of the policymaking process. They had to
identify ways to strategically recruit other supporters. Even the mayor
was hesitant at the beginning to support the project. The hesitation and
difficulty faced by the mayor when he had to choose between two dif-
ferent tram layouts served as evidence for some actors that the mayor
was unable to govern effectively. The tramway is not, in this case, just an
example. It was like the secret weapon used to legitimate the critique of
the mayor. By taking a position on the tramway layout, the mayor used
this opportunity to prove that he was a ‘decision maker.’
These three dimensions help us to understand that it is not only costly
for a policymaker to choose a proposal to support, but the conditions
that they accept, as a result of supporting a proposal or being recruited
by other actors, can also be costly. While there is no one answer to this
question, one of the possible costs is the redefinition of the proposal
itself, which transforms over its career to incorporate the ideas, identi-
ties, and power of the actors who support it. This redefinition can serve
to create a new link between a proposal and a new component, like
another problem, value, audience, or ideology. The redefinition can also
serve to add or transform some characteristic of the policy instrument.
Conclusion: Following the complex career of a policy

proposal to grasp the role of ideas, identity, and legitimacy
This piece began by discussing the concept of the policy paradigm as a

way to grasp the role of ideas in the policymaking process. While the
concept of a paradigm appears relevant for the study of natural science,
it is not so relevant for the study of policy. Drawing on the work of
Michel Foucault, the concepts of episteme and dispositif are more appro-
priate for highlighting the heterogeneous aspects of policy statements
and putting the process of making meaning and producing coherence at
the center of the analysis. Using the heuristic concept of a policy state-
ment allows researchers to treat coherence as a problem and not as a fact.
The dynamic of a policy proposal statement, from its emergence to its
transformation into a decision, appears to be a complex path. Supported
by few actors at the time of its emergence, the development of a policy
statement depends on its capacity to resist to argumentative attacks that
question its supposed meaning and coherence as well as the solidity of
the argument it reflects, and to absorb the costs of redefining the proposal
that new actors require in exchange for their support, which further sug-
gest a policy statement’s supposed malleability and incoherence. These
two processes simultaneously build meaning and coherence into the pol-
icy statement while at the same time restructuring the policy statement
based on the requirements of actors recruited to support it, a process that
provokes much tension and produces important changes that ultimately
serve to make the career of the policy statement a complex one.
Notes
1 Author’s translation from the French version.
2 Not in my back yard (NIMBY).
3 ‘“Techniques that are both dangerous and destructive for the environment are
used in the U.S. This is not going to happen in France. [. . .] Is it possible to
exploit shale gas otherwise, not to increase gas consumption but, for example,
to substitute imports? This is the purpose of these explorations. Again, it must
be precise: an exploration license is not a license to operate,” says Nathalie

Kosciusko-Morizet, Ministry of Ecology,’ Agence France Presse, 26 January 2011.
4 José Bové, European deputy (member of European parliament) says: ‘“It is sur-
prising to hear the minister say that France will not do it, we will not use dirty
techniques, it is wishful thinking as patents and technology are American,”
says the old peasant leader. [. . .] “Besides,” he added, “the French groups
are backed by Americans on French exploration permit” to Nant (Aveyron),
Montelimar (Drôme) and Villeneuve-de-Berg (Ardèche). [. . .] It would be the
only area where the law cannot undo what he did,’ Agence France Presse,
27 January 2010.
References
Banfield, E. C. (1961). Political Influence. Glencoe, IL: Free Press.
Boltanski, L. (2009). De la critique; Précis de sociologie de l’émancipation. Paris:
Gallimard.
Callon, M. (1986). Éléments pour une sociologie de la traduction. La domestica-
tion des coquilles Saint-Jacques et des marins-pêcheurs dans la baie de Saint-
Brieuc. L’année sociologique, 36, 169–208.
Crozier, M., & Friedberg, E. (1977). L’acteur et le Système. Paris: Seuil.
Dahl, R. A. (1965). Who Governs? Democracy and Power in an American City. New
Dahl, R. A., & Lindblom, C. E. (1953). Politics, Economics, and Welfare: Planning
and Politico-economic Systems Resolved into Basic Social Processes. New York:
Harper.
Deleuze, G. (1986). Foucault. Paris: Les éditions de minuit.
Foucault, M. (1966). Les mots et les choses. Paris: Gallimard.
Foucault, M. (1976). Histoire de la sexualité. Paris: Gallimard.
Foucault, M. (2008). L’arché‚ ologie du savoir. Paris: Gallimard.
Gusfield, J. R. (1984). The Culture of Public Problems: Drinking-Driving and the
Symbolic Order. Chicago, IL: University of Chicago Press.
Hall, P. A. (1986). Governing the Economy: The Politics of State Intervention in Britain
and France. Oxford: Oxford University Press.
Hall, P. A. (1993). Policy Paradigms, Social Learning and the State. Comparative
Politics, 25(3), 275–96.
Heclo, H. (1974). Modern Social Policy in Britain and Sweden: From Relief to Income
Maintenance. New Haven, CT: Yale University Press.
Heclo, H., & Wildavsky, A. (1974). Private Government of Public Money. London:
Macmillan.
Subsystems. Oxford: Oxford University Press.
Kingdon, J. (1995). Agendas, Alternatives and Public Policies. New York: Longman.
Chicago Press.
Kuhn, T. S. (2012). The Structure of Scientific Revolutions (4th ed.). Chicago, IL:
University of Chicago Press.
Latour, B. (1988). La vie de laboratoire. Paris: La découverte.
Latour, B. (1993). Aramis ou l’amour des techniques. Paris: La découverte.
Latour, B. (2006). Changer de société – refaire de la sociologie. Paris: La découverte.

Law, J. (1989). Le laboratoire et ses réseaux. In M. Callon (Ed.), La science et ses
réseaux (pp. 117–48). Paris: La Découverte.
Lindblom, C. E. (1958). The Science of Muddling Through. Public Administration
Review, 19, 78–88.
Lindblom, C. E. (1968). The Policy-Making Process. Englewood Cliffs, NJ: Prentice-
Hall.
Lordon, F. (1997). Les quadratures de la politique économique. Paris: Albin Michel.
Majone, G. (1989). Evidence, Argument, and Persuasion in the Policy Process. New
Mead, G. (1967). Mind, Self and Society from the Standpoint of a Social Behaviorist
(2nd ed.). Chicago, IL: University of Chicago Press.
Merton, R. K. (1968). Social Theory and Social Structure. New York: Free Press.
Merton, R. K. (1973). The Sociology of Science: Theoretical and Empirical Investigations.
Chicago, IL: University of Chicago Press.
Olson, M. (1985). The Logic of Collective Action: Public Goods and the Theory of
Groups. Cambridge, MA: Harvard University Press.
Parsons, W. (2003). Public Policy: An Introduction to the Theory and Practice of Policy
Analysis. Northampton, MA: Edward Elgard Publishing.
Perelman, C., & Olbrechts-Tyteca, L. (1958). Traité de l’argumentation: la nouvelle
rhétorique. Paris: Presses Universitaires de France.
Peters, B. G., & Pierre, J. (Eds.). (2006). Handbook of Public Policy. Thousand Oaks,
CA: Sage Publications.
Popper, K. R., & Lorenz, K. (1985). Die Zukunft ist offen: das Altenberger Gespräch:
mit den Texten des Wiener Popper-Symposiums. Munich: Piper.
Schmidt, V. (2008). Discursive Institutionalism: The Explanatory Power of Ideas
and Discourse. Annual Review of Political Science, 11, 303–26.
Simon, H. A. (1947). Administrative Behavior: A Study of Decision-Making Processes
in Administrative Organization. New York: Macmillan.
Zittoun, P. (2000). La politique du logement, 1981–1995. Paris: L’Harmattan.
Zittoun, P. (2001). Partis politiques et politiques du logement, échange de res-
source entre dons et dettes politiques. Revue Française de Science Politique, 51(5),
683–706.
Zittoun, P. (2011). From Instrument to Policy: Observing the Meaning Process to
Make a Decision. Politička Misao – Croatian Political Science Review, 48(5),
106–24.
Zittoun, P. (2013a). Dispositif. In I. Casillo, L. Blondiaux, F. Chateauraynaud, J.-M.
Fourniau, R. Lefebvre, C. Neveu, & D. Salles (Dir.), Dictionnaire critique et inter-
disciplinaire de la participation. Paris: GIS Démocratie et Participation. Retrieved
from: http://www.dicopart.fr/fr/dico/dispositif
Zittoun, P. (2013b). La fabrique politique des politiques publiques. Paris: Presses de
Science Po.
Zittoun, P. (2013c). Entre définition et propagation des énoncés de solution. Revue
Française de Science Politique, 63(3), 625–46.
Zittoun, P. (2014). The Political Process of Policymaking: A Pragmatic Approach to
Public Policy. Basingstoke: Palgrave Macmillan.
8
Paradigms and Unintended
Consequences: New Public
Management Reform and
Emergency Planning in Swedish
Local Government
Daniel Nohrstedt
Introduction
What is puzzling about paradigms, according to Hall (1993) and others,

is primarily how and why policies change over time through processes
of social learning. While this perspective has been influential in revis-
iting state-centric theories of policymaking, it leaves important ques-
tions about policy paradigms unanswered. One of the questions that
have not yet received much theoretical and empirical attention is how
paradigmatic policy reforms are actually received by public managers
in their everyday work. Do managers embrace emergent paradigms by
complying with new directions, rules, and practices, or do they set on a
conservative course of alienation and resistance? And what factors may
account for these differences? These questions are at the center of this
chapter.
The chapter provides a ‘microlevel’ account of how emergent para-
digms influence public management at the local government level. The
empirical setting involves a case of new public management (NPM)
reform in the area of civil emergency planning in Sweden. Introduced in
2006, this reform shared many of the doctrinal components commonly
associated with the NPM paradigm, including managerialism, decentral-
ization, and performance measurement (Hood, 1991). The study sheds
light on how this reform has impacted on the work by local government
(municipality) managers. The reform introduced legislative require-
ments to be carried out by the municipalities in the area of crisis plan-
ning and preparedness and guidelines for evaluation of implementation
and compensation. Based on prior research on the NPM paradigm and
141
its effects, the analysis sets out from the assumption that NPM reform
in an area characterized by substantial ambiguity will breed uncertainty
and unanticipated responses by public managers (Diefenbach, 2009).
This assumption has been elaborated in length in the broader litera-
ture on unintended consequences of social intervention (Sieber, 1981;
Smith, 1995), which provides a conceptual framework for this study.
Focusing specifically on local-level consequences of auditing and per-
formance assessment systems, the study presents an empirical exami-
nation of six potential unintended consequences commonly associated
with NPM reform:
1. Monitoring costs – costs (personnel, knowledge, and money) associ-

ated with implementation and monitoring
2. Tunnel vision – focus on quantifiable phenomena at the expense of
unquantifiable aspects of performance
3. Ossification – organizational paralysis through impediment of inno-
vation and renewal
4. Functional disruption – overemphasis on certain objectives (perfor-
mance criteria) at the expense of others
5. Overcommitment – resources allocated to certain activities or meas-
ures do not match objectives
6. Suboptimization – focus on narrow local objectives at the expense of
the objectives of the organization as a whole.
Each variable is analyzed using data over a five-year period (2009–2013)

derived from an annual nation-wide survey on emergency preparedness
work at the local government level in Sweden.
The chapter proceeds in four steps. The first step discusses reform
implementation in intergovernmental systems, introducing the case of
local-level civil emergency planning in Sweden and the 2006 reform.
In the second step, the link between NPM policy reform and potential
consequences of reform at the level of individual managers is discussed
from a theoretical perspective, focusing specifically on the motivations
managers may have to either resist or comply with new public programs
and a range of potential unintended consequences. In the third step, the
study identifies and operationalizes a set of variables as a basis for exam-
ining potential unintended consequences. These variables are then ana-
lyzed using survey data derived from the Swedish Civil Contingencies
Agency (Myndigheten för Samhällsskydd och Beredskap – MSB). Finally, the
concluding section discusses implications from this case analysis for
future research on policy paradigms.
Daniel Nohrstedt 143
Paradigms and unintended consequences of NPM ideas
This chapter places the analysis of NPM reform impacts in the context
of policy paradigms. As highlighted in the introduction to this vol-
ume, prior research on policy paradigms spans several questions related
to the development of ideas in policymaking including, for example,
the relationship between ideas behind policy change and how these
ideas change and transform in the context of dynamic environments.
Increasing our understanding of these problems requires exploration
of complementary approaches to shed light on different dimensions
of policy paradigms (Cairney, 2013). Building from the literature on
unintended consequences, this chapter contributes to the literature
on policy paradigms by developing a framework to analyze unforeseen
consequences of paradigmatic reform. In contrast to the dominating
focus in the policy paradigm literature on the process of paradigmatic
change, this chapter seeks to add to our understanding of managerial
implications of emerging paradigms. It hereby challenges the normative
claim that policy paradigms facilitate problem solving (Carson, 2004)
and views outcomes as an open empirical question. The theoretical
approach is based on the insight that the consequences of paradigmatic
change require consideration of motivations and behavioral responses.
NPM techniques have been widely adopted across countries and lev-
els of government with the purpose of improving the quality of public
service and make it more efficient and effective. For Hood (1991), pub-
lic sector reforms undertaken to improve the quality of public service
delivery represent a paradigm shift from the traditional bureaucratic
model of public administration. NPM reforms of the 1980s were typi-
cally founded on ideas questioning monopolistic forms of service pro-
vision and favored a more market-oriented approach to management
(Stoker, 2006). Considerable scholarly work has been undertaken since
then to unveil the implications of this movement toward performance-
motivated administration. Prior research on the consequences of NPM
reforms is, however, biased toward structural explanations and implica-
tions focusing on the extent to which NPM reforms have been imple-
mented across countries – what Premfors (1998, pp. 145–6, cited in
Green-Pedersen, 2002) labels ‘the structured pluralism story.’ Less work
has been undertaken to assess how these reforms have actually impacted
the work of public managers (Andrews & Boyne, 2012). Pollitt (1995)
notes that the NPM paradigm has changed the working life for public
managers, but there is less evidence on what the implications actually
are. In return, there is a need for studies that explore how NPM reforms
are received by the actors who work with NPM ideas in specific fields
(Brodkin, 2011; Carstensen, 2011).
Comparative evidence suggests that NPM reforms sometimes lead to
outcomes that diverge from original intentions and objectives. In some
instances, reforms may even worsen the conditions originally addressed
(Hood & Peters, 2004). However, prior empirical research on the effects
of NPM reforms is constrained by the structural bias and focuses primar-
ily on macrolevel causes and implications. These studies typically focus
on how macroeconomic factors, party politics, administrative cultures,
and political institutions condition implementation of NPM reforms
(Green-Pedersen, 2002) as well as on the effects NPM reforms have on
trust (Dolmans & Leeuw, 1997), transfer of institutional power (Power,
2005), and increased politicization (Maor, 1999). In this research, less
attention has been paid to unintended consequences for individual
managers (Elliott, 2002).
NPM has been described as a ‘paradigm’ (Behn, 1998), a ‘loose collec-
tion of ideas’ (Christensen & Laegreid, 1999), and a ‘shopping basket of
measures’ (Pollitt & Summa, 1997), which in turn requires multidimen-
sional approaches to evaluation of reform impacts. Central elements of
NPM have been outlined in detail elsewhere and include, for example,
marketization, decentralization, private sector styles of management,
and performance measurement (Hood, 1991; Pollitt, 1995). Against this
background, this study is limited to auditing – focusing on ‘the transfor-
mation of existing, and the emergence of new, formal institutions for
monitoring’ (Power, 2003, p. 188) – and the potentially negative side
effects that performance assessment systems may have for local-level
public managers.
Case background and data
Prior studies offer detailed insight into the development of the NPM
paradigm in Sweden, focusing particularly on the welfare state and
broader developments toward privatization, marketization, decentrali-
zation, and output orientation (Pollitt & Summa, 1997). Taken together,
these policy changes represent a paradigm shift in how the public sector
was governed (Lane, 2002). These changes have also influenced inter-
governmental relations by increased decentralization (transfer of func-
tions from the state to the municipalities) and discretion (extension of
municipality power to regulate institutional structure and organization
of local authorities). In addition, the municipalities have been subject
to several NPM-inspired patterns of organizational change, particularly
outsourcing of provision of services and introduction of municipality-

funded vouchers to promote freedom of choice to citizens to buy ser-
vices (e.g. elderly care and education) ‘on the market’ (Wollman, 2004).
Another prominent trend in Sweden involves a growing emphasis on
performance measurement in the municipalities (Pollitt & Bouckaert,
2011). Many municipalities utilize their own self-developed models,
but various state initiatives have been undertaken to facilitate for local
governments to compare performance in a range of areas (Siverbo &
Johansson, 2006).
Reforms in the area of civil emergency planning have followed the
same trend. In the 1990s and early 2000s, the Swedish crisis manage-
ment system was increasingly decentralized, as civil defense responsi-
bilities were gradually transferred from the state to the local government
level. In this regard, reforms undertaken in 2002 are a vivid example
of a neo-bureaucratic approach to governance combining greater cen-
tralization with increased local autonomy (Pollitt, 1993). While civil
emergency planning became a formal policy area in 2002 placing coor-
dination responsibilities at the state level, the reforms were also based
on the idea that local-level capacity was the cornerstone in the new sys-
tem for civil emergency planning. At the time, the government specified
the guiding principles for the crisis management system, one of which
proclaimed that events should be managed at the lowest level possible.
This principle was founded on the idea that Sweden’s crisis manage-
ment capacity should be built in a bottom-up fashion, which implied
delegation of core tasks from the state to the municipalities. These tasks
were regulated by law in 2006 based on an agreement two years earlier
(June 2004) between the state and the Association of Local Authorities
(Kommunförbundet). The reform was based on three central properties,
namely (1) specification of a number of performance targets; (2) rules
for economic compensation to the municipalities (appropriation), and
(3) development of a methodology for performance assessment. The
reform was carried out within the framework of the ‘local government
funding principle,’ which prescribes that if the state makes a decision
that directly affects the activities of the municipalities the state should
provide financial compensation.
The reform is founded on doctrinal components commonly associ-
ated with the NPM paradigm (see Hood, 1991). Above all, it prescribes
explicit standards and measures of performance and a greater emphasis
on output controls. The law identified a range of predefined performance
targets for the municipalities, including knowledge about risks and vul-
nerabilities, planning, capacity to respond to extraordinary events, and
collaboration before and during extraordinary events. Through the law,

the state also imposed on the municipalities five main tasks: (1) risk
and vulnerability analyses, (2) coordination of preparedness and emer-
gency response activities, (3) education and training, (4) logistical main-
tenance, and (5) reporting. According to the reform, if a municipality
would fail to meet these requirements, the state could either reduce
appropriation or withdraw it entirely.
Prior research on the implications of the NPM paradigm, particularly
regarding the effects of performance assessment systems, suggests that
this is one area where one could expect to find unintended consequences
(Lonsdale, Wilkins, & Lang, 2011; Weets, 2008). This is also a recurrent
observation in studies on NPM reforms in the context of emergency and
disaster management, which suggest that problems associated with per-
formance measurement (primarily related to difficulties in developing
reliable measures of operational performance) is one common source of
negative unintended consequences (Almklov & Antonsen, 2014; Eakin,
Eriksen, Eikeland, & Øyen, 2011; Kloot, 2009; Labadie, 2008).
To study the occurrence of unintended consequences empirically, this
study uses data retrieved from a nation-wide annual survey targeting
emergency preparedness managers in Swedish municipalities. The sur-
vey was developed as part of the reform and has been administrated by
the Swedish Civil Contingencies Agency since 2006 with the purpose
of following up on the tasks imposed on the municipalities by the state
and to monitor goal attainment (focusing on the performance targets
identified above). Respondents include local government civil servants
with different responsibilities for coordinating emergency management
activities in the municipalities. Since local crisis preparedness systems
are structured differently across the municipalities, respondents include
managers with different formal roles and positions in the municipal
bureaucracy – including crisis preparedness coordinators, heads of secu-
rity and safety, heads of rescue service, and municipality planners. Due
to missing data for 2006–2008, this study uses survey responses for the
five-year period 2009–2013.
Unintended consequences: Motivations and measures
Motivations
Theorists remain somewhat divided on the issue of how paradigm
shifts will be received by local-level public managers. Some scholars
argue that reforms will meet resistance while others expect high levels
of compliance. Although underlying motivations associated with vari-

ability in implementation is beyond the scope of this study, behavioral
assumptions and incentives remain important to our understanding of
potential unintended consequences (Blom-Hansen, 1999). The literature
offers three approaches to reform implementation in intergovernmental
systems, which are detailed below.
In the first approach, NPM reform can be expected to be implemented
without any major resistance from local-level managers. Power (1997)
advances the thesis that the increase in auditing functions has been fol-
lowed by a ‘compliance society’ characterized, among other things, by an
increased tick-box mentality. This tendency is akin to what Frederickson
(1997) labels an ‘ethic of compliance.’ Although the substantive effects
of a compliance culture remains an open empirical question, one may
expect that the growth of a tick-box compliance mentality generally
increases the tendency among local managers to adhere to administra-
tive procedures, control mechanisms, and rules associated with the NPM
paradigm (Lapsley, 2009; Maesschalk, 2004). Economic incentives play a
role here too, since performance monitoring accompanied by financial
accountability mechanisms can be expected to feed a blame-centered
form of auditing, which is likely to foster compliance (Power, 2003).
The second approach is based on a set of assumptions attributing
noncompliance behavior to political institutions, professional culture,
and manager perceptions. For example, Christensen and Laegreid (1999)
argue that certain political and administrative cultural norms and values
constrain NPM implementation. In their view, civil service operating in
traditions ‘emphasizing the political-democratic context of the civil ser-
vice, Weberian values, corporatism and the equality and de-emphasizing
economic considerations’ – such as the Scandinavian countries – will
not easily adapt fully to NPM reforms (Christensen & Laegreid, 1999,
p. 172). NPM-influenced innovations can threaten administrative cul-
tures, which may feed obstruction, modification, and avoidance behav-
ior among local managers. Compliance in implementation may also
be constrained because managers feel estranged from the program as
such. Tummers, Bekkers, and Steijn (2009) note that the risk for policy
alienation is particularly high in relation to the NPM paradigm, due to
perceived dysfunctional focus on efficiency and result, which may cause
managers to obstruct. Professional boundaries are yet another source of
noncompliance and resistance. Accordingly, the presence of cadres of
strong professional groupings and bonds is likely to feed behaviors pro-
moting distinct professional ethos, which feed systemic conservatism.
Some public service professions can thus be seen as an obstacle to man-

agerialism and may actively oppose implementation of NPM reforms
(Lapsley, 2009; Thomas & Davies, 2005).
The third approach attributes difficulties in implementation to fuzzy
objectives, which may generate discrepancy between the policy objec-
tives set by reformers and the goals of the managers (Smith, 1995). In this
view, policy objectives are seen as ambiguous and nontangible, which
leaves room for managers’ subjective interpretations and, thereby, devi-
ations in policy implementation (van Thiel & Leeuw, 2002).
This study focuses on unintended consequences associated specifi-
cally with performance assessment systems. Borrowing from Hood and
Peters (2004, p. 269), unintended consequences is defined as a paradox
of social intervention, which involves ‘outcomes and developments
that were unexpected, unintended, or contrary to received belief, par-
ticularly but not only in the form of unanticipated negative side and
reverse effects.’ Previous research on the implications of performance
assessment systems has centered on examining a range of outcomes,
for example, end user utility, performance improvement, political dis-
cussion, economic effects, organizational learning, and social influence
(Lonsdale et al., 2011; Weets, 2008). All of these dimensions are impor-
tant and contribute to a broader understanding of the implications of
the NPM paradigm. However, assessing effects associated with specific
dimensions of NPM interventions requires more specific indicators.
Measures
Strict analytical application of the notion of unintended consequences
would require close examination of the level of congruence between
objectives and actual outcomes. For an effect to be ‘unintended,’ empir-
ical evidence should confirm that outcomes diverge from, or even
worsen, original intentions (Vakkuri & Meklin, 2006). However, most
studies examining the effects of performance assessment systems are
not as strict – rather, they typically focus on regressive effects associated
with interventions without necessarily documenting linkages between
outcomes and intentions. To identify a number of potential unintended
consequences, this study builds from prior work on potential effects of
performance assessment systems, including Smith’s (1995) typology of
dysfunctional behavior that may appear when quantitative performance
indicators are applied in the public service and Sieber’s (1981) conver-
sion mechanisms that convert intentions into opposite outcomes. In
summary, these outcomes include monitoring costs, tunnel vision, ossi-
fication, functional disruption, overcommitment, and suboptimization.
These are detailed below along with operationalizations based on the

survey material.
Monitoring costs
Perhaps one of the most commonly noted effects of NPM reforms, and
the increased focus on performance measurement, is increased monitor-
ing costs (e.g. Bouckaert & Peters, 2002; Hefetz & Warner, 2004; Olsson,
Humphrey, & Guthrie, 2001; Pollitt, van Thiel, & Homburg, 2007).
According to Pollitt (1995, p. 146), these include opportunity costs that
occur as managers devote time to ‘adjusting to the new arrangements
rather than with their immediate operational responsibilities.’ The argu-
ment is straightforward: increasing concern with policy reform and
associated performance measurement systems bring costs (personnel,
knowledge, and money) in terms of implementation and monitoring.
This study documents monitoring costs by focusing on the time devoted
by municipalities to coordinate actions and measures stipulated by emer-
gency preparedness statutes. This is done by analyzing survey responses
indicating the share of a full-time position devoted to coordination of
emergency preparedness measures at the local government level.
Tunnel vision
Tunnel vision refers to situations when performance assessment is
confined to quantifiable phenomena at the expense of unquantifi-
able aspects of performance. The latter includes, for example, aspects
of human, social, and organizational capital (Diefenbach, 2009). Thus,
tunnel vision may be detected by assessing procedural limitations focus-
ing on aspects of performance that are not covered by performance
assessment systems, that is, by using some objective criteria (Dolmans &
Leeuw, 1997; Smith, 1995; Vakkuri & Meklin, 2006). However, there is
another way to document tunnel vision using surveys employed within
performance assessment systems. Quantifiable performance standards
are often supplemented with open-ended questions that encourage
respondents to provide substantive accounts of specific dimensions of
performance. In this study, responses to such open-ended questions are
used as an indicator of the weight attached to unquantifiable aspects of
performance. Concretely, high rates of missing values on open-ended
survey questions are an indication of tunnel vision.
Ossification
Due to a strong focus on the monitoring of implementation of predefined
targets, performance assessment systems may encourage organizations to
exploit old routines instead of exploring new, or innovative, solutions.

In return, performance measurement can provide an incentive for manag-
ers to avoid innovation out of a fear of damaging measured performance
(Neely, 2007). In this perspective, ossification refers to organizational
paralysis through impediment of innovation and renewal (Smith, 1995).
In this study, evidence of ossification is provided by survey responses
indicating that managers are not deliberately adopting measures that
deviate from predefined targets. Ossification thus exists when managers
remain fixed on their ability to live up to predefined performance targets
while avoiding elaborating on local measures or actions that depart from
those targets. Conversely, evidence of innovation includes responses
questioning the premises of evaluation and/or legitimization of local
solutions. Whether or not a lack of innovation or renewal actually leads
to paralysis is beyond the scope of this study. In order to get to these dif-
ferences the responses were coded based on a simple distinction between
‘defensive’ responses (covering a range of ‘excuses’ as to why perfor-
mance targets have not been met or fully implemented, including, for
example, administrative routines, organizational reform, staff turnover,
lack of time, resources, and political support) and ‘obstructive’ responses
(respondents either question the performance targets as such or elevate
local solutions as viable alternatives to those targets). Thus, the analysis
makes a rough distinction between responses indicating a mechanistic,
tick-box attitude to performance targets (defensive) and a more critical
approach to performance assessment instruments (obstructive). Here,
high rates of defensive responses provide evidence of ossification.
Functional disruption
Functional disruption refers to an overemphasis on certain objectives
(performance criteria) at the expense of others, which may drive sys-
tems to malfunction (Sieber, 1981). Conditions for functional disruption
vary from one case to another and depend on the distribution of needs
relative to the overarching objective of performance assessment systems.
In this perspective, NPM-style evaluations tend to ‘fall back on the pri-
oritization of orthodox measures, such as efficiency and productivity,
costs and technical performance’ (Diefenbach, 2009, p. 899). Therefore,
argues Heinrich (2012, p. 33), ‘we need measures that inform and will be
used by public managers, not only “accountability holders” such as leg-
islators and oversight agencies, to guide them in improving service qual-
ity and results.’ In this context, Lapsley (2009) has hypothesized that
continuous use of financial incentives will result in dysfunctional conse-
quences for public managers as it will promote self-interests and conflict
with the ethos of professionalism. Since the performance assessment sys-

tem studied here is partially an instrument for the state to monitor how
the municipalities administrate state appropriation for civil emergency
planning, the weight attached in the survey to budgetary (use of state
appropriation) and accountability (reporting) aspects in relation to other
performance criteria is one way to assess functional disruption.
Overcommitment
According to Sieber (1981), overcommitment involves unintended con-
sequences that occur when the resources allocated to certain activities,
or measures, do not match objectives. This is manifested in two ways:
objectively (resources are exhausted while expectations remain stable,
or resources are inadequate to meet expectations) and subjectively
(expectations exceed the capacity to meet them, regardless of increased
resources). This study measures overcommitment based on actors’ per-
ceptions regarding the effects of state appropriation in terms of its con-
tribution to (1) vulnerability reduction and (2) increased capacity of
municipalities to cope with peacetime emergencies.
Suboptimization
One of the known side effects of increased performance monitoring is
the tendency among bureaucratic organizations to focus on a limited
range of objectives. Smith (1995, p. 287) labels this effect suboptimiza-
tion, which refers to a focus on ‘narrow local objectives by managers, at
the expense of the objectives of the organization as a whole.’ According
to Neely (2007), suboptimization is partly the result of the hierarchical
nature of bureaucratic organizations, which tend to encourage people
to focus on their own local concerns rather than the higher-level per-
formance of the organization. Operationalization of suboptimization is,
again, context dependent and should be adapted to the system being
investigated. With regard to crisis management systems, suboptimization
has been cited as a recurrent problem that constrains continuous and
active involvement of relevant actors. Issues about risk and crisis are com-
monly the concern of relatively narrow circles of expert-oriented manag-
ers while other personnel and leaders are generally less actively engaged
(Boin & ’t Hart, 2010). In return, one way to measure suboptimization
is to assess the extent to which personnel and leaders are continuously
and actively involved in emergency preparedness work. This is done here
by documenting whether or not survey responses suggest that personnel
and elected officials are engaged in education and training. Table 8.1 pre-
sents a summary of all measures included in the study.
Table 8.1 Overview of unintended consequences, operationalization, survey questions, and years covered
152
Consequence Description Operationalization Survey questions Years covered
Monitoring The resources managers Share of full-time What share of annual full-time 2009–2013
costs need to spend on position devoted to position is devoted to
implementation and implementation and local coordination of measures
evaluation increase over coordination work required to implement
time (Bouckaert & Peters, the law?
2002; van Thiel & Leeuw,
2002; Wholey & Hatry,
1992)
Tunnel vision Emphasis on quantifiable Responses leave out reports All open-ended questions in 2009–2013
phenomena in the on aspects of policy the survey
performance measurement implementation that are
scheme at the expense of difficult to measure:
Policy Paradigms in Theory and Practice
unquantifiable aspects of missing values in responses

performance (Dolmans & to audit questions about
Leeuw, 1997; Smith, 1995) substantive (non-
quantifiable) aspects of
performance
Ossification Organizational paralysis, Responses indicating no Open-ended questions about 2009–2013
innovation and renewal deliberate deviations from actors’ motivations for not
impeded (Hunt, 2003; policy – respondents do implementing performance
Smith, 1995) not pursue local solutions goals: (1) knowledge about
but remain committed to risks and vulnerability, (2) risk
targets set by policy reduction planning, (3) crisis
management capacity, and
(4) interorganizational
coordination capacity.
Functional Overemphasis on one Weight attached to Share of survey questions 2009–2013
disruption need at the expense of performance criteria targeting financial accounting
others (Sieber, 1981) related to spending issues (local governments’
schemes relative to administration of state
other aspects appropriation).
Overcommitment Resources are inadequate Responses indicating 1. To what extent has state 2011–2013
to complete a task that state appropriation funded measures reduced local
(Sieber, 1981) is insufficient to achieve vulnerability?
objectives 2. To what extent has state
funded measures increased the
capacity to deal with peacetime
emergencies?
Suboptimization Pursuit of narrow local Active involvement of 1. Have elected officials 2011–2013
objectives by managers personnel and elected and/or non-elected officials
at the expense of officials in crisis been trained to fulfill their
the objectives of preparedness activities, responsibilities?
the organization as a including education 2. Have elected officials and/or
whole (Smith, 1995) and training non-elected officials been
educated to fulfill their
responsibilities?
Daniel Nohrstedt
153
Results
Results for the six variables are summarized in Figure 8.1. Bars indicate
the percentage of responses for each variable and year respectively. With
respect to the variable ‘budgetary and accountability’ (functional dis-
ruption) the bars indicate the percentage of the number of questions
included in the survey. Note that each variable should be assessed sepa-
rately and that trends are not comparable across variables.
Overall, the results give a mixed picture regarding unintended conse-
quences in the wake of the reform. Starting first with monitoring costs,
the responses indicate that the resources that the municipalities have
spent on civil emergency coordination work since 2009 have decreased
over time. In 2009, 35 percent of the municipalities reported that they
allocated means to one full-time position or more to coordinate meas-
ures and activities to implement the measures stipulated by the 2006
law. This number decreased gradually – in 2013, 21 percent of the
municipalities indicated they allocated means to one full-time position
or more. Since the number and the nature of the tasks related to emer-
gency planning and preparedness imposed on the municipalities have
not changed in this period, these figures may be taken as evidence of
increased efficiency; over time, the municipalities have invested fewer
resources to perform the same set of tasks. This would imply that man-
agers adjust to new practices over time, reducing the costs associated
with implementation. However, any conclusions regarding efficiency
Figure 8.1 Summary of survey responses indicating unintended consequences

(as well as effectiveness) should be made with a great degree of caution

as it would require more careful analysis to establish what tasks have
actually been accomplished. Decreased monitoring costs may, for exam-
ple, be attributed to high initial transaction costs during the early stages
of implementation. Regardless, the trend toward decreasing monitoring
costs is interesting since it challenges the assumption that NPM-reforms
are associated with escalating costs.
Second, Figure 8.1 provides evidence of a relatively stable trend
regarding respondents’ attention to non-quantifiable dimensions of
performance. To reiterate, tunnel vision was operationalized as the miss-
ing values ratio related to open-ended questions about performance (as
opposed to questions about quantitative aspects of performance). In this
case, missing values on open-ended questions were taken as evidence
of a bias toward quantifiable performance measures, which is one of
the unintended consequences associated with performance assessment
systems identified by the literature. However, in this case the lack of
attention to non-quantifiable aspects of performance remains relatively
low over time; since 2010 the missing value rate is about 2 percent,
which suggests that 98 percent of the respondents pay attention to non-
quantifiable aspects of performance within the survey. While it should
be kept in mind that missing values are merely one way of assessing
tunnel vision, the evidence assessed here suggest that in this case tunnel
vision is a relatively negligible consequence of the reform.
The third variable involves ossification, which refers to the tendency
among respondents to remain fixed on their ability to live up to prede-
fined performance targets, while avoiding elaborating on local meas-
ures, or actions, that depart from those targets. Figure 8.1 summarizes
the share of respondents that either question the performance targets
as such or elevate local solutions as viable alternatives to those targets.
These data indicate an increase in ossification over time: the share of
respondents that are not adopting measures deviating from predefined
targets is increasing over time. In 2009, 13 percent of the respondents
indicated deliberate deviations from predefined targets. This number
declines gradually and in 2013 only six percent indicated deviations.
In aggregate, this evidence seem to suggest that managers have become
less critical toward the premises of the evaluation and/or less inclined to
elevate local solutions that deviate from predefined performance targets.
The fourth unintended consequence of interest here involves func-
tional disruption, which stems from an overemphasis on certain per-
formance criteria at the expense of others. Functional disruption was
examined based on coding of the survey questions as such, focusing
on the share of questions related to budgetary and accountability pro-

cedures relative to other performance criteria. Figure 8.1 shows a quite
unstable pattern with regard to the structure of the survey. There was a
dramatic drop in attention to budgetary and accountability procedures
between 2009 (43 percent of the questions) and 2010 (8 percent). The
numbers then suggest a dramatic increase in attention to these aspects
in 2011 (34 percent) and a downward trend since then (26 percent in
2012 and 22 percent in 2013). Taken together, these figures suggest that
functional disruption is not a major issue in this case. Although budg-
etary and accountability procedures received a great deal of attention
in 2009 the emphasis since then seems to have shifted toward other
aspects of performance. Based on the last three years, this appears to be
a downward trend where attention is increasingly given to other aspects
of performance. However, it should be kept in mind that these findings
are founded on a rough distinction between performance criteria related
to budgetary and accountability procedures. A more complete examina-
tion of functional disruption would require a more thorough analysis
of multiple survey items. In addition, supplementary analyses would be
necessary to document whether attention to some performance dimen-
sions actually contravenes other dimensions.
Due to changes in the survey instrument, the remaining two variables –
overcommitment and suboptimization – are based on data covering
the last three years (2011–2013). The fifth variable involves overcom-
mitment, which as discussed above is a potential problem that occurs
when the resources allocated to certain activities, or measures, do not
match objectives. To study overcommitment empirically, this study uses
an aggregated measure combining survey responses indicating whether
or not state appropriation has been sufficient to help municipalities to
(1) reduce vulnerabilities and (2) strengthen the capacity to respond to
crises. Figure 8.1 summarizes the share of responses indicating that state
appropriation ‘to a significant degree’ has contributed to these capaci-
ties. The most interesting observation regarding overcommitment is
that it seems to increase over time; while 81 percent of the respondents
in 2011 indicated that state appropriation was sufficient, only 52 per-
cent did so in 2013. This data thus indicate a growing gap between the
level of state appropriation and the tasks that the municipalities are set
to perform in the area of emergency planning, which may be taken as
evidence of increased overcommitment.
The sixth and final variable assessed here is suboptimization, which
refers to the tendency among managers to focus on a limited range of
objectives at the expense of objectives of the organization as a whole.
In this study, suboptimization is measured using an additive measure of

the extent to which personnel and elected officials are engaged in educa-
tion and training. The results summarized in Figure 8.1 (for 2011–2013)
show that this is the one area where most progress has been made. In
2011, 86 percent of the respondents indicated that municipality per-
sonnel and elected officials were engaged in education and training,
which increased to 89 percent in 2013. These numbers are relatively
stable and indicate that the majority of the municipalities are engaging
in activities serving broader organizational purposes. Although educa-
tion and training of personnel and elected officials is a narrow indicator
of broader organizational objectives, it provides one measure of emer-
gency planning activities that reach beyond the daily work of profes-
sional managers.
This analysis is constrained by the same set of methodological and
interpretive puzzles encountered in the broader literature on NPM
reforms (e.g. Pollitt, 2000). The obvious paradox here is that assess-
ment of unintended consequences is associated with the same type of
problems as are performance measurement systems themselves, includ-
ing difficulties in establishing means-ends relations in the context of
ambiguity (Noordegraaf & Abma, 2003). With these limitations in
mind, the findings reported above offer some insights regarding local-
level implications of the NPM paradigm and performance assessment
systems.
In summary, assessment of the survey material provides a mixed pic-
ture regarding unintended consequences in this case. One interesting
observation is that some of the negative side effects typically associated
with NPM reforms and performance assessment systems are not visible
in this case. These effects include monitoring costs, tunnel vision, func-
tional disruption, and suboptimization. By contrast, the data suggest
that ossification and overcommitment might be areas where the reform
generated some negative consequences.
Conclusion
Existing research on policy paradigms spans several topics related to

the emergence, development, and change of public policies. Theoretical
and empirical analysis into these topics has advanced our understand-
ing of how the policy process develops over time and across governing
systems and policy areas. While this research covers many of the most
fundamental dimensions of contemporary governance – including ideas
for creating public policy, the terms for involvement, and strategies
for influence – a comprehensive understanding of paradigms warrants

examination of complementary approaches and perspectives on public
policy. This chapter claims that in order to gain deeper insights into
policy paradigms and their implications, we need to expand the analyti-
cal focus beyond policy change and also take into account how para-
digms affect public service organizations and what the implications are
at different levels. Against this background, the chapter relies on the lit-
erature on unintended consequences of social intervention to examine
how emergent paradigms may affect public managers. This framework
is illustrated empirically by a case study of the implications of the NPM
paradigm for local-level emergency planning in Sweden.
The study demonstrates that some effects are more prominent and
more negative than others. These differences turn attention to the
broader question of how variations in unintended consequences can
be explained. Why do paradigms generate regressive effects in some
cases, but not in others? Why are some unintended consequences
more prominent than others? Answering these questions would require
more systematic efforts to develop and apply theories that can explain
the occurrence of perverse effects of paradigms through policy reform
implementation (van Thiel & Leeuw, 2002). Furthermore, this effort
should not be limited to cases of bureaucratic failure and management
paradoxes, but should also cover instances of creative behavior and
solutions to circumvent unintended consequences in practice (Birnberg,
Turopolec, & Young, 1983; Pollitt, 2003).
Examination of unintended consequences in the context of para-
digms brings several methodological and conceptual challenges. For
example, with regard to unintended consequences of NPM reforms,
some central challenges include: (1) developing approaches to sepa-
rate effects from reforms and other environmental variables; (2) devel-
oping baseline performance indicators; (3) allowing multiple solutions
to the same outcomes; and (4) developing performance criteria that go
beyond matching identifiable effects against stated objectives (Pollitt,
1995). Against this background, two caveats are in order regarding
this analysis. First, this analysis explores different types of unintended
consequences associated with performance assessment, which each
rests upon different theoretical underpinnings (see Vakkuri & Meklin,
2006). That is, each unintended consequence is founded on a spe-
cific causal logic connecting the design of performance assessment
system components, managers’ motivations, and outcomes. Second,
the analysis is based solely on managers’ subjective survey responses.
‘Outcomes’ hence refer to different dimensions of performance as per-

ceived and subjectively reported by local-level public managers, which
differ from objective measures of performance (cf. Andrews, Boyne, &
Walker, 2006).
From a theoretical perspective, this chapter seeks to expand the schol-
arly discussion about paradigms and policy change beyond the terms for
adoption, or revision, of ideas in the policy process. Clearly, there is still
much to learn about what happens with policy reforms after they are
enacted: why some reforms survive in the long run while other reforms
are reversed, or even eroded away (Patashnik, 2008). In this perspective,
the notion of unintended consequences adds one piece to the bigger
puzzle of paradigmatic policy change. By studying how policy change
is actually received by public managers, administrators, civil servants,
street-level bureaucrats, and other actors whose everyday working life
is directly affected by major policy reforms, we can advance our under-
standing of paradigms beyond policy enactment. Assessment of the
implications of policy reforms at this level contributes to a better under-
standing of the nature of policy change, as well as the process of change
itself, including the mechanisms that drive it (Yee, 1996). In this per-
spective, analysis of unintended consequences can inform future studies
of policy paradigms in at least two ways.
First, a microlevel perspective on the consequences of policy reform
enables insight into the ways paradigms actually affect public organiza-
tions. Such a perspective would contribute to a more complex picture
of policy dynamics than is currently found in the existing literature
(cf. Cashore & Howlett, 2007). Most theoretical frameworks on policy
change dynamics conceptualize policy change in terms of the level and
scope of change without accounting for substantive impacts (Howlett,
2000). Although the notion of unintended consequences is clearly not
the solution to the dependent variable problem in the study of policy
change (Capano, 2009; Howlett & Cashore, 2009), it certainly adds an
important dimension to our conceptualization of outcomes and pro-
cesses associated with policy change. Second, consideration of the
implications of emergent paradigms also adds to our understanding of
long-term policy dynamics. In theory, unintended consequences may
provide a potentially important feedback mechanism that expose per-
formance gaps and deviations from policy goals. Whether or not such
feedback is being used strategically by stakeholders in the policy process
in attempts to revoke policy reform is another important avenue for
future research on policy paradigms.
References
Almklov, P., & Antonsen, S. (2014). Making Work Invisible: New Public
Management and Operational Work in Critical Infrastructure Sectors. Public
Administration, 92(2), 477–92.
Andrews, R., & Boyne, G. (2012). Structural Change and Public Service
Performance: The Impact of the Reorganization Process in English Local
Government. Public Administration, 90(2), 297–312.
Andrews, R., Boyne, G., & Walker, R. (2006). Subjective and Objective Measures
of Organizational Performance: An Empirical Investigation. In G. Boyne,
K. Meier, L. O’Toole, Jr., & R. Walker (Eds.), Public Service Performance: Perspectives
on Measurement and Management (pp. 14–34). Cambridge: Cambridge University
Press.
Behn, R. (1998). The New Public Management Paradigm and the Search for
Democratic Accountability. International Public Management Journal, 1(2),
131–64.
Birnberg, J., Turopolec, L., & Young, M. (1983). The Organizational Context of
Accounting. Accounting, Organizations and Society, 8(2/3), 111–29.
Blom-Hansen, J. (1999). Policy-Making in Central-Local Government Relations:
Balancing Local Autonomy, Macroeconomic Control, and Sectoral Policy
Goals. Journal of Public Policy, 19(3), 237–64.
Boin, A., & ’t Hart, P. (2010). Organising for Effective Emergency Management:
Lessons from Research. Australian Journal of Public Administration, 69(4),
357–71.
Bouckaert, G., & Peters, G. (2002). Performance Measurement and Management:
The Achilles’ Heel in Administrative Modernization. Public Performance and
Management Review, 25(4), 359–62.
Brodkin, E. (2011). Policy Work: Street-Level Organizations under New Manageria-
lism. Journal of Public Administration Research and Theory, 21(2), 253–77.
Cairney, P. (2013). Standing on the Shoulder of Giants: How Do We Combine the
Insights of Multiple Theories in Public Policy Studies? Policy Studies Journal,
41(1), 1–21.
Carson, M. (2004). From Common Market to Social Europe?: Paradigm Shift and
Institutional Change in European Union Policy on Food, Asbestos and Chemicals,
and Gender Equality. Stockholm: Stockholm University.
Capano, G. (2009). Understanding Policy Change as an Epistemological and
Theoretical Problem. Journal of Comparative Policy Analysis, 11(1), 7–31.
Carstensen, M. (2011). Paradigm Man vs. the Bricoleur: Bricolage as an Alternative
Vision of Agency in Ideational Change. European Political Science Review, 3(1),
147–67.
Understanding Thermostatic Policy Change Dynamics in Pacific Northwest
Forestry. American Journal of Political Science, 51(3), 532–51.
Christensen, T., & Laegreid, P. (1999). New Public Management: Design,
Resistance, or Transformation? A Study of How Modern Reforms Are Received
in a Civil Service System. Public Productivity and Management Review, 23(2),
169–93.
Diefenbach, T. (2009). New Public Management in Public Sector Organizations:
The Dark Sides of Managerialistic ‘Enlightenment.’ Public Administration, 87(4),
892–909.
Dolmans, L., & Leeuw, F. (1997). Performance Auditing. In Vraagtekens bij de effec-
tiviteit van een onderzoekstraditie (pp. 81–8). Nar eel doelmatiger overheid. Goed
georganiseerd mensenwerk. Den Haag: Netherlands Institute of Registered
Accountants (NIVRA).
Eakin, H., Eriksen, S., Eikeland, P.-O., & Øyen, C. (2011). Public Sector Reform
and Governance for Adaptation: Implications of New Public Management for
Adaptive Capacity in Mexico and Norway. Environmental Management, 47(3),
338–51.
Elliott, J. (2002). The Impact of Intensive ‘Value for Money’ Performance Auditing
in Educational Systems. Educational Action Research, 10(3), 499–506.
Frederickson, H. (1997). The Spirit of Public Administration. San Francisco, CA:
Jossey-Bass.
Green-Pedersen, C. (2002). New Public Management Reforms of the Danish and
Swedish Welfare States: The Role of Different Social Democratic Responses.
Governance, 15(2), 271–94.
Hefetz, A., & Warner, M. (2004). Privatization and Its Reverse: Explaining the
Dynamics of the Government Contracting Process. Journal of Public
Administration Research and Theory, 14(2), 171–90.
Heinrich, C. (2012). Measuring Public Performance and Effectiveness. In G. Peters &
J. Pierre (Eds.), The SAGE Handbook of Public Administration (pp. 32–49). London:
SAGE.
Hood, C. (1991). A Public Management for All Seasons? Public Administration,
69(1), 3–19.
Hood, C., & Peters, G. (2004). The Middle Aging of New Public Management: Into
the Age of Paradox? Journal of Public Administration Research and Theory, 14(3),
267–82.
Howlett, M. (2000). Managing the ‘Hollow State’: Procedural Policy Instruments
and Modern Governance. Canadian Public Administration, 43(4), 412–31.
Journal of Comparative Policy Analysis, 11(1), 33–46.
Hunt, B. (2003). The Timid Corporation: Why Business Is Terrified of Taking Risk.
Leicester, UK: Wiley.
Kloot, L. (2009). Performance Measurement and Accountability in an Australian
Fire Service. International Journal of Public Sector Management, 22(2), 128–45.
Labadie, J. (2008). Auditing of Post-Disaster Recovery and Reconstruction
Activities. Disaster Prevention and Management, 17(5), 575–86.
Lane, J.-E. (2002). New Public Management. London: Routledge.
Lapsley, I. (2009). New Public Management: The Cruellest Invention of the
Human Spirit? ABACUS, 45(1), 1–21.
Lonsdale, J., Wilkins, P., & Lang, T. (2011). Performance Auditing: Contributing to
Accountability in Democratic Government. Cheltenham: Edward Elgar.
Maesschalk, J. (2004). The Impact of New Public Management Reforms on Public
Servants’ Ethic: Towards a Theory. Public Administration, 82(2), 465–89.
Maor, M. (1999). The Paradox of Managerialism. Public Administration Review,
59(1), 5–18.
Neely, A. (ed.). (2007). Business Performance Measurement: Unifying Theory and
Integrating Practice. Cambridge: Cambridge University Press.
Noordegraaf, M., & Abma, T. (2003). Management by Measurement: Public

Management Practices Amidst Ambiguity. Public Administration, 81(4),
853–71.
Olsson, O., Humphrey, C., & Guthrie, J. (2001). Caught in an Evaluatory Trap: A
Dilemma for Public Service under NPFM. The European Accounting Review, 10(3),
505–22.
Patashnik, E. (2008). Reforms at Risk: What Happens after Major Policy Changes Are
Enacted? Princeton, NJ: Princeton University Press.
Pollitt, C. (1993). Managerialism in the Public Service: The Anglo-American Experience.
Oxford: Basil Blackwell.
Pollitt, C. (1995). Justification by Works or by Faith? Evaluating the New Public
Management. Evaluation, 1(2), 133–54.
Pollitt, C. (2000). Is the Emperor in His Underwear? Public Management Review,
2(2), 181–200.
Pollitt, C. (2003). The Essential Public Manager. Maidenhead: Open University
Press.
Pollitt, C., & Bouckaert, G. (2011). Public Management Reform: A Comparative
Analysis of New Public Management, Governance, and the Neo-Weberian State
(3rd ed.). New York: Oxford University Press.
Pollitt, C., & Summa, H. (1997). Trajectories of Reform: Public Management
Change in Four Countries. Public Management and Money, 17(1), 7–18.
Pollitt, C., van Thiel, S., & Homburg, V. (Eds.) (2007). New Public Management in
Europe: Adaptation and Alternatives. Basingstoke: Palgrave.
Power, M. (1997). The Audit Explosion: Rituals of Verification. Oxford: Oxford
University Press.
Power, M. (2003). Evaluating the Audit Explosion. Law and Policy, 25(3),
185–202.
Power, M. (2005). The Theory of the Audit Explosion. In E. Ferlie, L. Lynn, &
C. Pollitt (Eds.), The Oxford Handbook of Public Management (pp. 326–44).
Oxford: Oxford University Press.
Premfors, R. (1998). Reshaping the Democratic State: Swedish Experiences in a
Comparative Perspective. Public Administration, 76(1), 141–59.
Sieber, S. (1981). Fatal Remedies: The Ironies of Social Intervention. New York:
Plenum.
Siverbo, S., & Johansson, T. (2006). Relative Performance Evaluation in Swedish
Local Government. Financial Accountability and Management, 22(3), 271–90.
Smith, P. (1995). On the Unintended Consequences of Publishing Performance
Data in the Public Sector. International Journal of Public Administration, 18(2),
277–305.
Stoker, G. (2006). Public Value Management: A New Narrative for Network
Governance? American Review of Public Administration, 36(1), 41–57.
Thomas, R., & Davies, A. (2005). Theorizing the Micro-politics of Resistance: New
Public Management and Managerial Identities in the UK Public Service.
Organisation Studies, 26(5), 683–706.
Tummers, L., Bekkers, V., & Steijn, A. (2009). Policy Alienation of Public
Professionals: Application in a New Public Management Context. Public
Management Review, 11(5), 685–706.
Vakkuri, J., & Meklin, P. (2006). Ambiguity in Performance Measurement:

A Theoretical Approach to Organisational Uses of Performance Measurement.
Financial Accountability and Management, 22(3), 235–50.
van Thiel, S., & Leeuw, F. (2002). The Performance Paradox in the Public Sector.
Public Performance and Management Review, 25(3), 267–81.
Weets, K. (2008). How Effective Are Performance Audits? A Multiple Case Study
within the Local Audit Office of Rotterdam. Paper presented at the 5th
International Conference on Accounting, Auditing and Management in Public Sector
Reforms, Amsterdam, 3–5 September 2008.
Wholey, J., & Hatry, H. (1992). The Case for Performance Monitoring. Public
Administration Review, 52(6), 604–10.
Wollman, H. (2004). Local Government Reforms in Great Britain, Sweden,
Germany, and France: Between Multi-function and Single-Purpose
Organisations. Local Government Studies, 30(4), 639–65.
Yee, A. (1996). The Causal Effects of Ideas on Policies. International Organization,
50(1), 69–108.
Part III
9
The Critical Role of Ideas:
Understanding Industrial Policy
Changes in Ireland in the 1980s
John Hogan and Brendan K. O’Rourke
Introduction
In the current context of economic crisis we examine how earlier

Irish governments, also confronted with challenging economic cir-
cumstances, sought to alter the country’s industrial policy. During
the second half of the 1970s the Irish economy performed relatively
well, after weaker performance following the first oil shock. However,
recovery proved transitory, as procyclical fiscal policies fed inflation. By
the 1980s, the economy shrank, and unemployment and emigration
returned. This led to a questioning of industrial policy in place since
the 1950s. What changes were made to this policy during the 1980s?
And what lessons might this hold for contemporary policy makers? We
use the Critical Junctures Theory (CJT) to investigate these questions.
According to Hogan (2006) a critical juncture is a multistage event that
sets a process of policy change in motion. A crisis can create a situation
where extant policies and associated ideas are called into question by
change agents. Any subsequent displacement of the extant paradigm
by a new set of ideas on how policy should operate can lead to radical
policy change. But, without ideational change, policy change will likely
be relatively minor – the hierarchy of goals underpinning a policy will
remain unaltered and extant policy will soldier on. Through using the
CJT we can gain a deeper understanding of the nature of the changes
that occurred in Irish industrial policy during the 1980s.
Case study: Testing for a critical juncture in Irish

industrial policy in the 1980s
From the 1950s onwards, Irish industrial policy moved away from
protectionism, seeking to attract foreign direct investment (FDI) as a
167
stimulus for growth and skills transfer (Girvin, 1994, p. 125). This fun-
damental policy change, prompted by dire economic performance,
emigration, and a precipitously falling population, constituted a critical
juncture in Irish industrial policy. Between 1981 and 1986 Ireland again
experienced severe economic difficulties. Once more Irish society, and
its policy makers, experienced great uncertainty (O’Rourke & Hogan,
2013). Would the state continue to rely on FDI as its engine for growth,
or would there be a radical change in industrial policy once more?
The critical juncture literature
Critical junctures result in the adoption of a particular arrangement

from among alternatives. Thereafter, the pathway established funnels
units in that direction (Mahoney, 2003, p. 53). For some, a critical junc-
ture constitutes an extended period of reorientation (Collier & Collier,
1991; Mahoney, 2003), while for others, it is a brief period in which one
direction, or another, is taken (Garrett & Lange, 1995; Hogan, 2006).
More recently, Flockhart (2005) used critical junctures to explain the
gap between Danish voters and their politician’s attitudes towards the
European Union, while Wolff (2012) examined the development of an
EU counter-terrorism policy through critical junctures. However, the lit-
erature is inconsistent in how it differentiates critical junctures from
other forms of policy change – such as incremental change. Also, the
literature often examines critical junctures from the perspective of crises
(exogenous shocks), emphasizing the tensions that precede the impor-
tant events that set policy change in motion. Hogan (2006) sought to
resolve these issues by setting out a revised CJT and developing a frame-
work capable of testing for critical junctures and producing consistent
findings.
According to Hogan and Doyle (2007), a critical juncture consists of dis-
crete, but interconnected, elements: crisis, ideational change (extant idea-
tional collapse, new ideational consolidation), and radical policy change.
Thus, CJT uses ideas in a form of ‘discursive institutionalism’ to over-
come the limitations of ‘traditional’ new institutionalist approaches (see
historical institutionalism) in explaining policy change – specifically their
static and overly determinist nature (Schmidt, 2008, 2010). Discursive
institutions are not rule following structures of the older institution-
alisms that serve as restraints on actors, but are internal to agents as
constraining structures and enabling constructs of meaning (Schmidt,
2010). Hogan and Doyle (2007) argue that, in the wake of a crisis,
outside influencers (public, media, NGOs, etc.), policy entrepreneurs
John Hogan and Brendan K. O’Rourke 169
(civil servants, technocrats, academics, economists, interest groups),

and political entrepreneurs (elected politicians) act, in the words of
Kleistra and Mayer (2001), as either carriers or barriers to policy change.
Discursive interaction (exchange of ideas) between these policy elites
and the general public generates the alternative ideas that may lead to
collective action (Schmidt, 2008). The attention to the entrepreneur-
ial agency and to the role of discourse of these researchers constitutes
an effort to ‘endogenize’ policy change (Schmidt, 2010) – making the
exogenous shocks, so important to historical institutionalism, less
significant.
Using the CJT to examine the nature of industrial policy change in
Ireland during the 1980s involves developing observable implications
to test for a crisis, ideational change, and radical policy change. These
observables incorporate aspects of societal/policy/political change that
we would expect to find. The stronger that the evidence supports the
observables, the greater the indication that a critical juncture occurred.
The framework (set out in detail below) has been applied to a variety
of policy areas: change in macroeconomic policy, the rolling back of
privatization policy, and the study of policy change in nondemocratic
states (see Donnelly & Hogan, 2012; Hogan, 2006; Hogan & Cavatorta,
2013; Hogan & Doyle, 2007).
Applying CJT
Testing for a crisis

A crisis implies extant policies are failing to address a problem (Boin,
’t Hart, Stern, & Sundelius, 2005) and as a result can unleash power-
ful forces for change that can have a lasting impact (Haggard, 2003).
Economic crises are more common in modern democracies than wars
or revolutions. Hogan and Doyle’s (2007, p. 888) CJT, recognizing that
identifying a macroeconomic crisis is difficult, involving subjective and
objective deliberations, uses 12 encompassing observable implications
that draw upon the currency crisis, recession, and policy reform work of
Garuba (2006), Kaminsky, Reinhart, and Végh (2003), Pei and Adesnik
(2000, p. 139), and Yu, Lai, and Wang (2006, p. 439). These observables
identify changes in nominal economic performance and perceptions of
economic health (see Appendix A).
Testing for ideational change

Ideational change can result in a transformed policy environment, but
understanding how ideas influence policy is something theorists have
long grappled with. The failure of extant policies to resolve a crisis pro-
vides a window of opportunity for change agents to contest the via-
bility of the underlying paradigm (Kingdon, 1995). These agents can
gain power for their ideas by setting the agenda for reform in the policy
sphere (Schmidt, 2010).
To address the question of why ideas underlying failing policies
sometimes change, resulting in radical policy change, whereas at other
times they remain unaltered, resulting in minor policy change, Hogan
and Doyle (2007), drawing on Dahl (1963), Kingdon (1995), and Legro
(2000), argue that significant policy change depends upon a broad
range of change agents (outside influencers and policy entrepreneurs)
perceiving the extant paradigm as inadequate (collapse) and coalescing
(consolidation) around a set of new ideas, championed by a political
entrepreneur. Political entrepreneurs can act as a bridge between coali-
tions advocating new policy ideas and the institutions implementing
them (Hogan & Feeney, 2012). Thus, once the new policy idea becomes
accepted amongst policy entrepreneurs and the political elite, a new
policy monopoly, and stasis, is instituted (Meijerink, 2005). As Blyth
(2002, p. 37) argues, ‘ideas facilitate the reduction of . . . barriers by
acting as coalition-building resources among agents who attempt to
resolve the crisis’. Ideational change constitutes the intermediating
factor between a crisis and policy change. Based on Hogan and Doyle’s
(2007) CJT framework, we set out nine observable implications for
identifying extant ideational collapse and new ideational consolidation
(see Appendix B).
However, ‘even when ideational collapse occurs, failure to reach con-
sensus on a replacement could still produce continuity, as society reflex-
ively re-embraces the old orthodoxy’ (Legro, 2000, p. 424). Thus, even
in the wake of a crisis, policy failure and ideational collapse, there is
no guarantee new ideas will become policy. This is because in addition
to policy viability, policy ideas must have administrative and political
viability (Hall, 1989).
Testing for policy change

The CJT leads us to expect significant policy change after there is political
entrepreneur-led consolidation around a new idea (ideational change)
in the wake of a crisis. Thus, the CJT’s final stage employs Hall’s (1993)
concepts of first, second, and third order change to develop observable
implications to enable us identify and differentiate normal and funda-
mental shifts in policy (see Appendix C). The observables incorporate
Hogan’s (2006) notions of swift and enduring change. This addresses
the problem in policy dynamics of defining and operationalizing the

scope and timing of policy change (Howlett, 2009). As Capano and Howlett
(2009) argue, when a policy is regarded as fundamental it is usually based
on a multiyear perspective. These observables enabled the differentia-
tion of policy changes, ranging from minor adjustments to the setting of
policy instruments to paradigm changes in policy goals (Hall, 1993).
Evaluation of the findings

To evaluate the evidence for a crisis, extant ideational collapse, new
ideational consolidation, and the type of policy change, the finding
for each observable implication was evaluated independently by each
author and assigned a score according to whether it indicated strong
(3), medium (2), weak (1), or no (0) support. The stronger the inter-
coder agreement indicated between the researchers’ findings, the more
confidence we have in the findings. This approach allows for a more
nuanced understanding of policy change. As interpretation plays a part
in divining meaning from codes, the reporting of findings involves
thick description of categories and contexts (Polgar & Thomas, 2008,
p. 248). We found inter-coder agreement to vary between 100 per cent
and 33 per cent, averaging 73 per cent, and Krippendorff’s alphas to
vary between 1 and 0.022, averaging 0.57, which Krippendorff (2004)
deems reliable (see Appendix D).
Identification of macroeconomic crisis
The Irish economy in the 1980s

During the late 1970s the Irish economy performed relatively well, hav-
ing recovered from the 1974 oil crisis. The high levels of inflation and
unemployment, which peaked in the middle of the decade, had begun
to fall, while growth returned (Leddin & Walsh, 1998, p. 26). We can
see from Table 9.1 that real gross domestic product (GDP) increased by
4.5 per cent on average between 1975 and 1979.
In 1977, a new Fianna Fáil government employed an expansionary fis-
cal policy at a time when the economy was growing at an unsustainable
rate (OECD, 1982, p. 10). This fiscal injection, in pursuit of a ‘dash or
growth’ that economists warned was unsustainable, led to a deteriora-
tion in fiscal balances, with the public sector borrowing requirement
increasing from 13 per cent of GNP in 1976 to 20 per cent in 1981
(FitzGerald, 2000, p. 43). This period saw record deficits in the current
external balance and the public sector accounts (OECD, 1983, p. 7).
As we can see from Table 9.1, inflation peaked at 20.4 per cent in 1981,
while interest rates remained high.
As the economy slowed, and then began to shrink and unemploy-
ment and interest rates rose, emigration increased (OECD, 1982, p. 10).
More people were unemployed in June 1981 than at any time in the
country’s history (The Irish Times, 1981a, p. 6). The balance of payments
deficit was 13 per cent of GNP (Central Bank of Ireland, 1982, p. 16).
The government’s spending was so high that the total amount budgeted
for 1981 had been consumed by midyear. Consequently, almost half
of exchequer borrowing for 1981 went to financing the current budget
deficit (Bacon, Durkan, & O’Leary, 1986, p. 6), which stood at an unsus-
tainable 7.3 per cent of GNP (Leddin & O’Leary, 1995, p. 167).
The debt to GNP ratio was on an unbroken upward trajectory from
1977 to 1987, surpassing 100 per cent by 1984. Between 1979 and 1986
the rate of increase of debt to GNP regularly exceeded 10 per cent per
annum. The fiscal deficit, intended in the late 1970s to be temporary,
became impossible to eliminate. Imports and exports fluctuated wildly,
as reflected in figures for trade openness in Table 9.1. Only inflation
Table 9.1 Ireland’s main economic indicators, 1974–1989
Government Growth rates

Unemployment Inflation Interest Economic
Year debt to GNP in real GDP
(%) (%) (%) opennessa
ratio (%)
1974 5.4 17.0 12 55.39 4.2 94.37706

1975 7.3 20.9 10 58.05 2.0 86.44022
1976 9.0 18.0 14.8 62.5 2.1 94.96721
1977 8.8 13.6 6.8 61.4 6.9 102.0045
1978 8.1 7.6 11.9 63.5 6.7 103.7894
1979 7.1 13.2 16.5 70.65 2.4 109.601
1980 7.3 18.2 14.0 71.91 1.9 106.4831
1981 9.9 20.4 16.5 77.45 1.1 105.0999
1982 11.4 17.1 14.0 86.53 −0.7 97.81183
1983 13.6 10.5 12.3 97.60 −1.6 101.7349
1984 15.4 8.6 14 106.28 2.3 112.7388
1985 16.7 5.4 10.3 108.60 0.8 112.172
1986 17.1 3.8 13.3 123.26 −1.1 101.1026
1987 17.7 3.1 9.3 124.07 4.6 104.8707
1988 16.4 2.1 8.0 117.35 4.4 109.5299
1989 15.1 4.1 12.0 106.84 7.0 117.0717
a
Measured by the trade to GDP ratio. This is acquired by adding the value of imports and exports
and dividing by GDP.
Source: European Commission (1997), Heston, Summers, and Aten (2002), Leddin and Walsh
(1998), Mitchell (1992), United Nations (2011).
improved after 1981. By 1986, the economy had been in continuous

stagnation since 1980, contracting for the third time in five years (see
Table 9.1). An Irish Independent poll found that a majority of citizens
were sceptical of the politicians’ ability to run the country properly
(O’Regan, 1981, p. 1).
The need to control public expenditure and prevent excessive reliance
on foreign borrowing dictated the adoption of tighter fiscal policies.
However, the catch-all nature of Irish political parties induced gov-
ernments to buy off short-term pressure from interest groups through
ad hoc policy concessions. This worked against the imposition of severe
economic policies and the formulation of enduring agreements between
the state and economic interest groups, like those in Continental
neo-corporatism.
By the mid-1980s there was unanimity in the domestic and foreign
media concerning the economy. Finlan (1987, p. 16), writing in The
Irish Times, described the economy as being ‘on the ropes’. The Economist
pointed out that by 1987 ‘the people of Ireland were deeply in debt
to the outside world, three times as much per head as Mexico’ (The
Economist, 1987, p. 53). The Irish Times (1987a, p. 10) noted that some
economic commentators were advocating debt repudiation due to the
scale of indebtedness. Overall, the general consensus in the newspapers
was one of stagnation and crisis.
Most economists were critical of economic policy and performance
during the 1980s (Bradley & FitzGerald, 1989, p. iii). While initially
boosting the economy, the government’s debt finance plan for rapid
development between 1977 and 1980 was a disaster, due to the depres-
sion and debt burden that followed (O’Rourke, 2010). Bacon et al.
(1986, p. 1) observed that ‘the first half of the decade of the 1980s,
taken as a whole, was a period of appalling economic performance’.
‘It is difficult to avoid the conclusion that Irish economic performance
has been the least impressive in Western Europe, perhaps in all Europe’
(Kennedy & Conniffe, 1986, p. 288). The Central Bank of Ireland
(1987, p. 7) foresaw no immediate prospects for an improvement in
growth or employment. More worryingly, it argued that the situation
did not permit for increases in welfare benefits to the disadvantaged.
An increasing level of poverty was eroding the lives of a growing seg-
ment of society (O’Morain, 1987, p. 7). As the OECD (1987, p. 77) put
it: ‘by the mid-1980s a number of acute imbalances confronted the
Irish economy’. These imbalances were also making the business com-
munity worried. The Small Firms Association noted steadily declining
business confidence (The Irish Times, 1987b, p. 6). Leading business-
man Tony O’Reilly warned of the dangers of International Monetary
Fund (IMF) involvement in running the country if the crisis was not
resolved (Keenan, 1987, p. 8).
After 1982 all the major parties agreed on the need to stabilise the
debt/GNP ratio (Mjøset, 1992, p. 381). The state changed its overall pol-
icy from focusing on employment to balancing budgets, export growth,
and international competitiveness. A member of Fine Gael, Alexis
Fitzgerald, remarked that in just 4 years Fianna Fáil had doubled the
national debt it had taken 57 years to accumulate (The Irish Times, 1981a,
p. 6). Subsequently, Prime Minister Garrett FitzGerald acknowledged
that the national debt and interest payments, rising faster than national
income, constituted a vicious circle, each year consuming a larger share
of taxation (The Irish Times, 1987c, p. 10). The opposition leader Charles
Haughey remarked that ‘the economy is at a total stand-still’ (Cooney,
1987a, p. 9). Public consensus held that the country was in the midst of a
serious financial crisis (Cooney, 1987b, p. 1). ‘By 1987 the Irish economy
was universally seen to have reached nadir’ (The Economist, 1992, p. 6).
From Table 9.2, we see that both authors felt the majority of observ-
able implications support the argument that Ireland, during the 1980s,
experienced an economic crisis. The next sections test for ideational
change in industrial policy during this crisis and the nature of the policy
change that followed.
Table 9.2 Identification of macroeconomic crisis in Ireland in the 1980s
Identification of macroeconomic crisis Coder 1 Coder 2
O1. Stagnant or negative GDP growth 3 3

O2. Unemployment above 10 per cent 3 3
O3. Inflation and interest rates above 10 per cent 3 3
O4. National debt, as a percentage of GDP, increasing 3 3
at more than 10 per cent, annually
O5. The level of economic openness declining 3 2
O6. Public perceives economic crisis 2 2
O7. Media perceive economic crisis 3 3
O8. Economic/political commentators perceive 3 3
economic crisis
O9. Central bank perceives economic crisis 3 3
O10. OECD perceives economic crisis 3 3
O11. Elected representatives perceive economic crisis 3 3
O12. Government pronouncements consistent with 3 3
crisis management approach
Strong Strong
(3) Strong support | (2) medium support | (1) weak support | (0) no support | (N/A) not
available.
Identification of ideational and policy change
The ideas underlying industrial policy

After opening its economy in the 1950s, Ireland sought to attract labour
intensive industries based around technically mature products (Lee,
1992). In the context of the weakness of indigenous industry, this seemed
a realistic alternative capable of delivering industrial development (Lee,
1992). By the beginning of the 1970s FDI accounted for the majority of
new manufacturing jobs and exports. However, this approach led to the
sidelining of indigenous industry in the policy process (Girvin, 1989).
It also failed to create high value jobs and the foreign companies were
footloose (O’Malley, 1985). In response, by the early 1970s the Industrial
Development Authority (IDA) started to look at attracting more sophis-
ticated overseas producers in pharmaceuticals and electronics (Wrigley,
1985). However, the economic problems of the 1970s led to a question-
ing of industrial policy by society, the media, and policy entrepreneurs.
This questioning was marked by a number of reports.
The Cooper–Whelan report

The Cooper–Whelan report (1973), co-authored by Noel Whelan, a civil
servant, who later became Secretary of the Department of the Taoiseach
(prime minister), was sceptical of the long-term benefits of FDI, as
opposed to indigenous enterprise (Lee, 1992). This report questioned
the weak links between FDI and indigenous industry. By 1979, it was
clear that the success of the FDI sector was not finding its way into the
rest of the economy (Lee, 1992).
This situation impelled the National Economic and Social Council
(NESC), then under Whelan’s chairmanship, to commission an exam-
ination of Ireland’s industrial programme (Brunt, 1988, p. 32). The
objective was ‘to ensure that the Irish government’s industrial policy is
appropriate to the creation of an internationally competitive industrial
base in Ireland which will support increased employment and higher
living standards’ (Telesis, 1982, p. 3). This examination encompassed
a number of reports, the most important of which was Telesis’s (1982)
examination of industrial policy.
The Telesis report

A Review of Industrial Policy, referred to as the Telesis report, after praising
the clarity and consistency of industrial policy, highlighted the prob-
lems of FDI, its impact on job creation, and its failure to create linkages
with indigenous industry (Telesis, 1982). Pointing out that sustained
economic development and a high income economy rely on native

entrepreneurship, the report queried the policy emphasis on foreign
investment (Lee, 1992, p. 531) and a culture of dependence on state
aid (Sweeney, 1998, p. 103). It argued that the IDA had not been as suc-
cessful in attracting FDI as it suggested (Telesis, 1982). Of jobs approved
between 1972 and 1978 ‘only 20 percent were in place at the end of
the period’ (Coogan, 1987, p. 6). The report encouraged a shift ‘towards
building strong indigenous companies in the export and sub-supply
business sector’ (Sweeney, 1998, p. 127) and that the proportion of
funds given to domestic industries (one-third of all funding available) be
doubled by the end of the decade (Telesis, 1982). The Telesis report ‘sent
shock waves through the policy establishment’ (Lee, 1992, p. 532). This
questioning of industrial policy challenged its intellectual coherence.
Reaction to Telesis – Ideational collapse

In light of the criticisms of the IDA in Telesis, it disputed the findings
of the report (O’Brien, 1982, p. 1). The IDA had become ‘the centre
of policy-making, a point strongly criticised by Telesis’ (Sweeney, 1998,
p. 127). ‘The development effort aimed towards new indigenous indus-
try must be reorganised to emphasise the building of structurally strong
Irish companies rather than strong agencies to assist weak companies’
(Telesis, 1982, p. 33). Deputy Prime Minister Dick Spring admitted that
Telesis pointed to failings in the overall industrial policy framework
(Dáil Debates 342, cols 861, 11 May 1983).
There was ideational collapse (see Table 9.3) as Whelan, the NESC, Telesis,
and a host of economists, constituting policy entrepreneurs, critiqued the
Table 9.3 Indication of ideational collapse
Indication of extant ideational collapse Coder 1 Coder 2
O13. Media questioning efficacy of current model 2 3

O14. Opposition critiques current model and propose 1 1
alternative ideas
O15. Policy entrepreneurs critique current model and 3 2
propose alternatives
O16. Civil society organizations critique current model 3 2
O17. Widespread public dissatisfaction with current paradigm 2 1
O18. External/international organizations critique current 2 2
model and actively disseminate alternatives
Extant ideational collapse Strong Medium
(3) Strong support | (2) medium support | (1) weak support | (0) no support | (N/A) not
available.
orthodoxy underlying extant industrial policy, while other change agents

in the media and trade unions supported the ideas proposed by Telesis.
Changes in industrial policy
Political instability – Policy drift

Telesis found that extant industrial policy ‘did not go far enough in devel-
oping native skills in technology and marketing, the key elements of
self-sustaining growth. The foundations of the industrial superstructure
therefore lacked depth’ (Kennedy, 1986, p. 49). The scale of the policy
changes required by Telesis would take time to both implement and take
effect (Kennedy, 1983, p. 34). The Economic and Social Research Institute
(ESRI), predicting the economic crisis would worsen, insisted that the
necessary policy changes could not be delayed (O’Brien, 1983, p. 12).
[However] the period from December 1979 to December 1982 was one
of the most remarkable periods in modern Irish history. There were
four changes of Taoiseach in that period, six Ministers for Finance,
three changes of government and the Irish economy declined pro-
gressively to a level unprecedented for decades. (Browne, 1983, p. 5)
Apart from crisis-induced cutbacks, a general commitment to austerity,

and a determination to maintain the trade unions at a remove from
the policy-making process (Roche, 1994), no coherent macroeconomic
policies emerged, as the governments were of such unstable character.
A coherent adoption of the neo-liberal ideas and polices, then current
in the US and UK, was to wait almost another decade in Ireland, until
the dawning of the Celtic Tiger period (Allen, 2000). A knock-on con-
sequence was that changing extant industrial policy in a coherent fash-
ion, despite the collapse of its underlying ideas, would prove difficult.
Thus, ‘not only economically, but also politically, instability peaked in
1981–1982’ (Mjøset, 1992, p. 381).
By November 1982, when the majority Fine Gael–Labour coalition
came to power, with the national debt exceeding GNP and the cur-
rent budget deficit out of control, a coherent set of corrective policies
had become essential. In this respect, the state of the public finances
restricted the government to austerity measures. Nevertheless, the coali-
tion experienced difficulty in devising an effective strategy (O’Byrnes,
1986, p. 219) and initially sought to avoid expenditure reductions
through increased taxation (McCarthy, 1999, p. 5). It is against this
background that we examine the government’s response to Telesis.
The political response to Telesis

For Snoddy (1982), Telesis marks the ending of the phase of reliance on
FDI and the beginning on a new phase focused on indigenous industry.
However, reality was more complex. The Minister for Industry, John
Bruton, insisted that it was critical not to lose sight of the contribu-
tion of 800 foreign firms, employing 80,000 workers (Dáil Debates 342,
col. 906, 11 May 1983). Although reports from the ESRI and NESC
acknowledged that the Irish economy was in crisis and that change was
required, the official response to Telesis was slow in coming.
For some, the government’s White Paper on Industrial Policy ‘repre-
sented a pivotal document in the re-evaluation of the philosophy and
strategic thrust of industrial policy’ (Boylan, 1996, p. 196). There were
indeed ‘several changes in the content of industrial policy’ in response
to Telesis (Kennedy, 1995, p. 59). It was acknowledged that ‘industrial
policies which had clearly served Ireland well in the 1960s and 1970s are
now having less success’ (Government of Ireland, 1984, p. 3) and rec-
ognized that economic ‘flexibility, creativity and growth were all being
thwarted by the dependence on foreign investment’ (Munck, 1993,
p. 158). Yet, in concert with Telesis (1982), the White Paper stated that
there would be no radical change to incentives for FDI – ‘consistency
and stability over many years of our policies for industrial develop-
ment have been a major source of strength’ (Government of Ireland,
1984, p. 7). There was a radical change in that employment creation
was no longer the sole objective and that attention should focus on
the international competiveness of the whole industrial sector (Brunt,
1988, p. 32). Wealth creation now took precedence over job creation,
as this permitted the development of the capacity to create jobs. There
was to be greater selectivity, the White Paper seeking to advance the
process of ‘picking winners’, with the aim of developing domestic com-
panies with export potential (Jacobsen, 1994, p. 169), and a promotion
of indigenous enterprise (Carr, 2000). The report also saw a shift away
from manufacturing towards services.
Alterations to industrial policy

Under the coalition government, incentives were to be more selectively
deployed, while indigenous enterprises were given greater attention
through the creation of a National Development Corporation (NDC).
The objective was to foster an increase in the number of internationally
traded companies, and to that end an export development scheme was
set up. There was some change in relation to FDI, with more of a focus
on foreign companies with R&D functions and the potential for linkages
with domestic enterprises. The National Planning Board pointed to a

lack of a manpower policy and argued that the appropriate ministers,
not the IDA, ought to direct policy (Jacobsen, 1994, p. 168).
The year 1984 also saw the introduction of the Company Development
Programme directed at indigenous companies and designed to assist
planning (Bielenberg & Ryan, 2013, p. 29). Policy change could be seen
in the National Linkage Programme in 1985 to achieve a more inte-
grated development pattern between indigenous and foreign enter-
prises (Brunt, 1988, p. 32). It would focus on upgrading local suppliers
by improving their technical know-how and being selective in concen-
trating on companies ‘which had the potential to succeed’ (Murdoch,
1985, p. 14). These approaches sought to achieve the new objectives
of industrial policy: a greater focus on indigenous enterprises; address-
ing weaknesses in management and marketing; achieving better value
for money; and creating better linkages with FDI enterprises (Kennedy,
1995). By 1985, the Irish Export Board (Córas Tráchtála) was seeking to
address the information needs of exporters, as identified in the White
Paper, through the publication of booklets entitled Guides for Exporting.
Nevertheless, the IDA continued to attract FDI and in 1990, at unprece-
dented costs, secured a major investment from Intel (Bielenberg & Ryan,
2013, p. 29).
‘The Telesis report led to some measure of industrial policy reform,
though this was less interventionist than the report envisaged’
(Bielenberg & Ryan, 2013, p. 29). Thus, industrial policy gradually
responded to the critiques of Telesis (1982) and the NESC (1982), who
criticized the failure of state support towards indigenous enterprises
(Ó’Gráda, 1997). However, as Hall (1989, p. 11) points out, states tend
to be predisposed towards those policies with which they already have
some favourable experience. Thus, to regard the policy changes arising
from Telesis, and the subsequent White Paper, as a break with extant
industrial policy would be incorrect. Rather, they constituted a form of,
what Streeck and Thelen (2005) refer to as, policy layering. These moves
represent learning effects, an effort to make extant policy work. Despite
Telesis, grants to indigenous firms increased by only 3 per cent between
1985 and 1989 (O’Hearn, 2001, p. 105). Industrial policy was adjusted in
1984, not transformed, due to a reluctance to break with past successes
in attracting FDI.
The nature of the policy change

Although the ideas underlying extant industrial policy might have col-
lapsed in the wake of the economic crisis, and provided a window of
opportunity for radical policy change, this did not occur. Whelan, the
NESC, and Telesis, along with many other economists and commentators,
acting as policy entrepreneurs, proposed alternative ideas to those under-
lying industrial policy. However, in the absence of these change agents
clustering around a political entrepreneur to champion their alternative
paradigm in the policy-making environment, new ideational consolida-
tion could not occur. Despite the growing recognition that industrial
policy was failing to produce the desired results, the Irish political estab-
lishment was reluctant to abandon a policy prescription that had, at least
during the 1960s and early 1970s, ended a century of depopulation and
stagnation. No political entrepreneur emerged during this period of idea-
tional contestation. Politicians only seemed interested in variations on
the existing paradigm. This was partly down to the fact that Ireland, at
the time, experienced a period of weak and unstable governments as the
economic crisis gripping the country deepened (The Irish Times, 1981b,
p. 12). In these circumstances, the ideas underpinning extant industrial
policy endured. The collective mindset failed to disengage from a reliance
on FDI and shift the focus of industrial policy to indigenous enterprise.
During the 1980s the ideas underpinning industrial policy’s focus
on FDI had collapsed. However, from Table 9.4 we can see that, in the
absence of a political entrepreneur willing to champion alternative ideas,
the change agents failed to consolidate around replacement orthodoxy.
The observables indicate that policy change was only of the first order
Table 9.4 Indication of (1) new ideational consolidation and (2) level of policy
change
1. Indication of new ideational consolidation

O19. Clear set of alternative ideas 3 3
O20. Political entrepreneur injecting new ideas into policy 0 0
arena
O21. Political entrepreneur combines interests to produce 0 0
consensus around a replacement paradigm
New ideational consolidation No No
2. Indication of level of policy change

O22. Policy instrument settings changed 3 3
O23. Instruments of policy changed 1 2
O24. Goals behind policy changed 0 0
Critical juncture in policy No No
(3) Strong support | (2) medium support | (1) weak support | (0) no support | (N/A) not available.
(Hall, 1993). Thus, there was an economic crisis and ideational col-
lapse, and no new ideational consolidation, and there were only minor
changes to industrial policy in 1984. From this we can conclude that
there was no critical juncture in Irish industrial policy at this time.
Addendum: The Culliton report 1991
‘The attempt to follow these new directions of policy scarcely had time
to prove themselves before a new review of industrial policy was initi-
ated in June 1991’ (Kennedy, 1995, p. 60). The Industrial Policy Review
Group’s focus, set out in the Culliton report (named after its chairman)
was again on the indigenous sector (O’Gráda, 1997, p. 119). The report
called for a reduction in grants and improving competiveness more gen-
erally, rather than picking winners (Newman, 2011, p. 241).
The central message was that the policy for industrial development
goes beyond industrial policy as traditionally conceived (Kennedy, 1995).
It advocated the breakup of the IDA. The result was ‘the creation of three
agencies, with IDA-Ireland specialising in promoting foreign invest-
ment, Enterprise Ireland focusing on assisting indigenous enterprise,
and ForFás concentrating on policy advice’ (White, 2001, p. 223).
The continued overdependence on foreign capital was a major con-
cern, with this report advocating the development of linkages between
domestic and foreign firms (O’Hearn, 2001, p. 105). This report’s find-
ings were similar to Telesis, despite the intervening decade.
Conclusion
According to theory, a critical juncture consists of crisis, ideational

change, and radical policy change, with ideational change linking crisis
and policy change. Following a crisis, policy failure, and extant ideational
collapse, significant policy change depends upon actors, led by a political
entrepreneur, reaching consensus upon, and consolidating around, new
ideas. It is in the discursive interactions between the various actors that
ideas are generated along with the potential for radical policy change.
Employing a range of observable implications, we did not find a criti-
cal juncture in industrial policy. Although the economy was in crisis,
undermining confidence in prevailing industrial orthodoxy, neither of
the main political parties was willing to challenge the status quo. While
ideational collapse occurred and alternatives were put forward, a politi-
cal entrepreneur willing to champion a new set of ideas on industrial
policy failed to emerge – extant orthodoxy endured.
In Ireland, since the 1950s, an outward-orientated economic policy has

remained unaltered, due to the recognition that a large number of jobs
and a high percentage of FDI depend upon membership of the EU. The
economic crisis of the 1980s did not provoke a rethink of this overarch-
ing economic vision. However, policy entrepreneurs questioned the focus
of industrial policy. There was growing scepticism of the long-term ben-
efits for the economy of a reliance on FDI and in particular its weak links
with indigenous industry. Following Telesis, industrial policy was altered to
attract more sophisticated foreign companies that would base R&D func-
tions in Ireland, and thereby create higher value jobs, and to provide greater
supports to indigenous industries which had previously been sidelined.
Now, 30 years after the events discussed, Ireland has become a mod-
ern knowledge-based economy focusing on services and high-tech
industries – many of which are foreign multinationals. The interven-
ing years have witnessed fiscal austerity, followed by the Celtic Tiger
and then more austerity following the banking collapse (O’Rourke &
Hogan, 2014). The economy has also gone through widespread deregu-
lation and privatization, all of which impacted upon industrial policy,
but without shifting its underlying paradigm. Thus, the attraction of FDI
is still crucial. Negotiations over the country’s recent IMF–EU bail-out
saw a key element of industrial policy, the 12.5 per cent corporation tax
rate, asserted as non-negotiable in the face of Franco-German pressure.
Foreign companies’ significant presence can also been seen in the dif-
ference between Ireland’s GNP and GDP figures. GNP, as a percentage of
GDP, has ranged from a high of 85.6 per cent in the third quarter of 2007
to a low of 79 per cent in the first quarter of 2012 (CSO, 2013).
Appendix A
O1. There is stagnant or negative GDP growth (Pei & Adesnik, 2000).
O2. Unemployment is above 10 per cent (Pei & Adesnik, 2000).
O3. Inflation and interest rates are above 10 per cent (Pei & Adesnik, 2000).
O4. National debt, as a percentage of GNP, exceeds 100 per cent and is
increasing at more than 10 per cent, annually.
O5. The level of economic openness declines.
O6. Public perceives an economic crisis.
O7. National/international media perceive an economic crisis.
O8. Economic/political commentators perceive an economic crisis.
O9. Central bank perceives an economic crisis.
O10. Organisation for Economic Cooperation and Development (OECD)
perceives an economic crisis.
O11. Elected representatives perceive an economic crisis.
O12. Government pronouncements on economy are consistent with a crisis
management approach.
Appendix B
Indication of extant ideational collapse

O13. Media question efficacy of the current model and/or specific policy areas.
O14. Opposition parties critique the current model and propose alterna-
tive ideas – at election time their platform will be built around these
alternatives.
O15. Civil servants, technocrats, academics, and economists (policy entre-
preneurs) critique the current model and propose alternatives.
O16. Civil society organizations, for example, labour unions, employer
organizations, and consumer groups (policy entrepreneurs), critique
the current model, reflecting Hall’s (1993) coalition-centred approach.
O17. There is widespread public dissatisfaction with the current paradigm,
observable through opinion polls, protests, and so on.
O18. External/international organizations (policy entrepreneurs) critique the
current model and/or actively disseminate alternative ideas.
Indication of new ideational consolidation

O19. A clear set of alternative ideas are developed by policy entrepreneurs.
O20. A political entrepreneur injects new ideas into the policy arena.
O21. The political entrepreneur combines interests, including policy entre-
preneurs, to produce consensus around a replacement paradigm.
Appendix C
Indication of level of policy change

O22. Policy instrument settings changed (swiftly and for longer than one
government’s term of office).
O23. The instruments of policy changed (swiftly and for longer than one
government’s term of office).
O24. The goals behind policy changed (swiftly and for longer than one gov-
ernment’s term of office).
Appendix D
Inter-coder agreement scores
Krippendorff’s α Agreement (%)
Crisis 0.635 92
Ideational collapse 0.022 33
Ideational consolidation 1 100
Policy change 0.615 66
References
Allen, K. (2000). The Celtic Tiger: The Myth of Social Partnership in Ireland.
Manchester: Manchester University Press.
Bacon, P., Durkan, J., & O’Leary, J. (1986). The Irish Economy: Policy and Performance
1976–1986. Dublin: Economic and Social Research Institute.
Bielenberg, A., & Ryan, R. (2013). An Economic History of Ireland since Independence.
London: Routledge.
Boin, A., ’t Hart, P., Stern, E., & Sundelius, B. (2005). The Politics of Crisis Management:
Public Leadership under Pressure. Cambridge: Cambridge University Press.
Boylan, T. (1996). Rural Industrialisation and Rural Poverty. In C. Curtin,
T. Haase, & H. Tovey (Eds.), Poverty in Rural Ireland: A Political Economy Perspective
(pp. 173–210). Cork: Oak Tree Press in association with Combat Poverty
Agency.
Bradley, J., & FitzGerald, J. (1989). Medium-Term Review: 1989–1994. Dublin:
Economic and Social Research Institute.
Browne, V. (1983). Three Turbulent Years. In B. Trench (Ed.), The Book of Irish
Politics 1983 (pp. 5–6). Dublin: Magill.
Brunt, B. (1988). The Republic of Ireland. London: Paul Chapman.
Capano, G., & Howlett, M. (2009). Introduction: The Determinants of Policy
Change: Advancing the Debate. Journal of Comparative Policy Analysis, 11(1),
1–5.
Carr, P. (2000). Understanding Enterprise Culture: The ‘Fashioning’ of Enterprise
Activity within Small Business. Strategic Change, 9(7), 405–14.
Central Bank of Ireland. (1982). Annual Report. Dublin: Stationery Office.
Central Bank of Ireland. (1987). Quarterly Bulletin, Summer. Dublin: Central Bank
of Ireland.
Central Statistics Office [CSO]. (2013). What Are GDP and GNP? Retrieved from
http://www.cso.ie/en/studentscorner/nationalaccounts/whataregdpandgnp/
Collier, R. B., & Collier, D. (1991). Shaping the Political Arena: Critical Junctures, the
Labour Movement, and Regime Dynamics in Latin America. Princeton, NJ:
Coogan, T. P. (1987). Disillusioned Decades: Ireland 1966–87. Dublin: Gill and
Macmillan.
Cooney, J. (1987a, February 4). Haughey Attacks ‘Fairy Tales’ of PDs. The Irish
Times, p. 9.
Cooney, J. (1987b, February 13). Haughey Denies Any Intention of Altering Pact.
The Irish Times, p. 1.
Cooper, C., & Whelan, N. (1973). Science, Technology, and Industry in Ireland: Report
to the National Science Council. Dublin: Stationery Office.
Dahl, R. A. (1963). Modern Political Analysis. Englewood Cliffs, NJ: Prentice-Hall.
European Commission. (1997). Eurostat Yearbook. Luxembourg: Office for Official
Publications of the European Communities.
Finlan, M. (1987, February 20). Why Haughey Has Fewer Options This Time. The
Irish Times, p. 16.
FitzGerald, J. (2000). The Story of Ireland’s Failure – And Belated Success. In
B. Nolan, P. J O’Connell, & C. T. Whelan (Eds.), Bust to Boom? The Irish
Experience of Growth and Inequality (pp. 27–57). Dublin: Institute of Public
Administration.
Flockhart, T. (2005). Critical Junctures and Social Identity Theory: Explaining the
Gap between Danish Mass and Elite Attitudes to Europeanization. Journal of
Common Market Studies, 43(2), 251–71.
Garrett, G., & Lange, P. (1995). Internationalization, Institutions, and Political
Change. International Organization, 49(4), 627–55.
Garuba, D. S. (2006). Survival at the Margins: Economic Crisis and Coping
Mechanisms in Rural Nigeria. Local Environment, 11(1), 17–36.
Girvin, B. (1989). Between Two Worlds: Politics and Economy in Independent Ireland.
Dublin: Gill and Macmillan.
Girvin, B. (1994). Trade Unions and Economic Development. In D. Nevin (Ed.),
Trade Union Century (pp. 117–32). Cork: Mercier Press.
Government of Ireland. (1984). White Paper on Industrial Policy. Dublin: Stationery
Office.
Haggard, S. (2003). Economic Crisis and Corporate Restructuring in Korea: Reforming
the Chaebol. Cambridge: Cambridge University Press.
Hall, P. A. (1989). The Power of Economic Ideas. Princeton, NJ: Princeton University
Press.
Heston, A., Summers, R., & Aten, B. (2002). Penn World Table Version 6.1.
Philadelphia, PA: Center for International Comparisons of Production, Income
and Prices at the University of Pennsylvania.
Hogan, J., & Cavatorta, F. (2013). Examining Critical Junctures in Macroeconomic
Policy in Non-democratic States: Algeria and Jordan in Comparative Perspective.
Digest of Middle Eastern Studies, 22(1), 13–38.
Hogan, J., & Doyle, D. (2007). The Importance of Ideas: An A Priori Critical
Juncture Framework. Canadian Journal of Political Science, 40(4), 883–910.
Hogan, J., & Feeney, S. (2012). Crises and Policy Change – The Role of the Political
Entrepreneur. Risk, Hazards & Crisis in Public Policy, 4(2), 1–16.
Howlett, M. (2009). Process Sequencing Policy Dynamics: Beyond Homeostasis
and Path Dependency. Journal of Public Policy, 29(3), 241–62.
Jacobsen, K. (1994). Chasing Progress in the Irish Republic. Cambridge, MA:
Kaminsky, G. L., Reinhart, C. M., & Végh, C. A. (2003). The Unholy Trinity of
Financial Contagion. Journal of Economic Perspectives, 17(4), 51–74.
Keenan, B. (1987, February 14). It’s Easier to Run a Big Company Than a Small
Country. Irish Independent, p. 8.
Kennedy, K. A. (1983). Symposium on Industrial Policy in Ireland – The Design of
a New Industrial Strategy. Journal of the Statistical and Social Inquiry Society of
Ireland, 24(5), 34–41.
Kennedy, K. A. (1986). Industry: The Revolution Unfinished. In K. A. Kennedy

(Ed.), Ireland in Transition: Economic and Social Change since 1960 (pp. 152–66).
Cork: Mercier Press.
Kennedy, K. A. (1995). Irish Enterprise in the 20th Century. In A. E. Burke (Ed.),
Enterprise and the Irish Economy (pp. 53–69). Cork: Oak Tree Press in association
with Graduate School of Business, University College Dublin.
Kennedy, K. A., & Conniffe, D. (1986). Employment Policy in Ireland. Dublin:
Economic and Social Research Institute.
Kingdon, J. W. (1995). Agendas, Alternatives, and Public Policy (2nd ed.). New York:
Harper.
Kleistra, Y., & Mayer, I. (2001). Stability and Flux in Foreign Affairs: Modelling
Policy and Organizational Change. Cooperation and Conflict, 36(4), 381–414.
Krippendorff, K. (2004). Content Analysis: An Introduction to Its Methodology (2nd
ed.). Beverly Hills, CA: Sage.
Leddin, A., & O’Leary, J. (1995). Fiscal, Monetary, and Exchange Rate Policy. In
J. W. O’Hagan (Ed.), The Economy of Ireland: Policy and Performance of a Small
European Country (pp. 159–95). Dublin: Gill and Macmillan.
Leddin, A., & Walsh, B. M. (1998). The Macroeconomy of Ireland (4th ed.). Dublin:
Gill and Macmillan.
Lee, J. J. (1992). Ireland 1912–1985: Politics and Society. Cambridge: Cambridge
University Press.
Legro, J. W. (2000). The Transformation of Policy Ideas. American Journal of
Mahoney, J. (2003). Long-Run Development and the Legacy of Colonialism in
Spanish America. American Journal of Sociology, 109(1), 50–106.
McCarthy, D. (1999). Building a Partnership. In B. Reynolds & S. Healy (Eds.),
Social Partnership in a New Century (pp. 4–20). Dublin: CORI Justice
Commission.
Meijerink, S. (2005). Understanding Policy Stability and Change: The Interplay of
Advocacy Coalitions and Epistemic Communities, Windows of Opportunity,
and Dutch Coastal Flooding Policy 1945–2003. Journal of European Public Policy,
12(6), 1060–77.
Mitchell, B. R. (1992). International Historical Statistics, Europe 1750–1988 (3rd
ed.). New York: Stockton Press.
Mjøset, L. (1992). The Irish Economy in a Comparative Institutional Perspective.
Dublin: National Economic and Social Council.
Munck, R. (1993). The Irish Economy: Results and Prospects. London: Pluto Press.
Murdoch, B. (1985, February 21). New Approach to Substitution of Imports
Planned. The Irish Times, p. 14.
National Economic and Social Council. (1982). Policies for Industrial
Development: Conclusions and Recommendations. NESC Report No. 66.
Dublin: National Economic and Social Council.
Newman, C. (2011). Manufacturing and Internationally Traded Services. In
J. W. O’Hagan & C. Newman (Eds.), The Economy of Ireland (pp. 173–210).
Dublin: Gill and Macmillan.
O’Brien, K. (1982, October 16). IDA Defends Policy after Telesis Attack. The Irish
Times, p. 1.
O’Brien, K. (1983, August 18). ESRI Now Forecasting 0.3% Fall in GNP This Year.
O’Byrnes, S. (1986). Hiding behind a Face: Fine Gael under FitzGerald. Dublin: Gill
and Macmillan.
Ó’Gráda, C. (1997). A Rocky Road. Manchester: Manchester University Press.
O’Hearn, D. (2001). Global Restructuring and the Irish Political Economy. In
P. Clancy, K. Lynch, S. Drudy, & L. O’Dowd (Eds.), Irish Society: Sociological
Perspectives (pp. 90–131). Dublin: Institute of Public Administration.
O’Malley, E. (1985). The Problem of Late Industrialisation and the Experience of
the Republic of Ireland. Cambridge Journal of Economics, 9(2), 141–54.
O’Morain, P. (1987, February 2). Threats and Promises: The Economics of the
Election. The Irish Times, p. 7.
O’Regan, G. (1981, June 2). Haughey, Gibbons, Feud in the Open. Irish Independent,
p. 1.
Organisation for Economic Cooperation and Development [OECD]. (1982).
Ireland 1982. Paris: OECD.
OECD. (1983). Ireland 1983. Paris: OECD.
OECD. (1987). Ireland 1987. Paris: OECD.
O’Rourke, B. K. (2010). Enterprise Discourse: Its Origins and Its influence in
Ireland. In J. Hogan, P. F. Donnelly, & B. K. O’Rourke (Eds.), Irish Business and
Society: Reflections on Governing, Participating and Transforming in the 21st Century
(pp. 69–89). Dublin: Gill and Macmillan.
O’Rourke, B. K., & Hogan, J. (2013). Reflections in the Eyes of a Dying Tiger:
Looking Back on Ireland’s 1987 Economic Crisis. In A. De Rycker & Z. Mohd
Don (Eds.), Discourse and Crisis: Critical Perspectives (pp. 215–38). Amsterdam/
Philadelphia: John Benjamins.
O’Rourke, B. K., & Hogan, J. (2014). Guaranteeing Failure: Neoliberal Discourse in
the Irish Economic Crisis. Journal of Political Ideologies, 19(1), 41–59.
Pei, M., & Adesnik, A. D. (2000). Why Recessions Don’t Start Revolution. Foreign
Policy, 118(1), 138–51.
Polgar, S., & Thomas, S. A. (2008). Introduction to Research in the Health Sciences
(5th ed.). Edinburgh: Churchill Livingstone.
Roche, W. K. 1994. Pay Determination and the Politics of Industrial Relations. In
T. V. Murphy & W. K. Roche (Eds.), Irish Industrial Relations in Practice
(pp. 126–205). Dublin: Oak Tree Press in association with the Graduate School
of Business University College, Dublin.
Schmidt, V. A. (2010). Taking Ideas and Discourse Seriously: Explaining Change
through Discursive Institutionalism as the Fourth ‘New Institutionalism’.
Snoddy, R. (1982, April 16). How the IDA Is Acting as a Matchmaker. Financial
Times, p. 31.
Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond Continuity.
University Press.
Sweeney, P. (1998). The Celtic Tiger: Ireland’s Economic Miracle Explained. Cork: Oak
Tree Press.
Telesis. (1982). A Review of Industrial Policy. NESC Report No. 64. Dublin: National
Economic and Social Council.
The Economist. (1987, January 24–30). How the Government Spent the People
into a Slump. The Economist, p. 53.
The Economist. (1992). Ireland: Country Profile 1991–92. The Economist Intelligence
Unit, p. 6.
The Irish Times. (1981a, June 5). Ryan Pledge on Neutrality. The Irish Times, p. 6.
The Irish Times. (1981b, June 15). Politicians’ Choice. The Irish Times, p. 12.
The Irish Times. (1987a, February 12). Bruton Wants Wider Base of Corporation
Tax. The Irish Times, p. 10.
The Irish Times. (1987b). Business Confidence Declining – Survey. The Irish Times,
p. 6.
The Irish Times. (1987c, February 11). Debt Still Outpacing Income – Taoiseach.
United Nations. (2011). United Nations Data: A World of Information. Retrieved
from http://data.un.org/Data.aspx?d=IFS&f=SeriesCode%3a60
White, P. (2001). The IDA Philosophy through the Decades. In R. Mac Sharry &
P. White (Eds.), The Making of the Celtic Tiger: The Inside Story of Ireland’s Boom
Economy (pp. 198–228). Cork: Mercier Press.
Wolff, S. (2012). The Mediterranean Dimension of the European Union’s Internal
Security. Basingstoke: Palgrave.
Wrigley, L. (1985). Ireland in Economic Space. In J. J. Lee (Ed.), Ireland: Towards a
Sense of Place. Cork: Cork University Press.
Yu, L., Lai, K. K., & Wang, S.-Y. (2006). Currency Crisis Forecasting with General
Regression Neural Networks. International Journal of Information Technology and
Decision Making, 5(3), 437–54.
10
The Bologna Process and the
European Qualifications
Framework: A Routines Approach
to Understanding the Emergence of
Educational Policy Harmonisation –
From Abstract Ideas to Policy
Implementation
Sharon Feeney and Conor Horan
Introduction
In this chapter we explain the background to the emergent nature of the

Bologna Process within a policy context. We identify the specific arti-
facts that were linked to how signatories attempted to meet the objec-
tives of the Bologna Process. We trace the selection and retention of
abstract ideas within the policy formation stage of the Bologna Process
and illustrate how these facilitated the emergence of an educational
policy whose goal was to harmonise qualifications within an agreed
European Qualifications Framework (EQF). In doing this, we utilise the
ostensive–performative theory of routines as our theoretical lens. By
using this approach we distinguish it from hegemonic approaches to
policy change such as the punctuated equilibrium framework (Jones &
Baumgartner, 2012) and advocacy coalition framework (Weible et al.,
2011). We illustrate the emergence of the harmonisation sub-routine as
it is supported by various declarations and communiqués. We trace pat-
terns of activities relating to both policy formation and implementation.
Our contribution is twofold. Firstly, we argue for use of routines theory
as a theoretical tool to understand the principles, objectives, and pro-
cesses undertaken in the creation and adoption of the EQF. Secondly, we
highlight the importance of abstract ideas, their connection to endog-
enous change, and how they inform performances. Finally, we conclude
this chapter with recommendations for further use of routines theory as
189
an alternative to the hegemonic theories and paradigms currently used

in policy studies in order to better understand historical and ongoing
policy change in an international context.
The Bologna Process from a routines perspective
A routines theory approach to understand harmonisation

The first of six objectives of the Bologna Declaration (European
Commission, 1999) aims to develop a harmonised approach to develop-
ing an EQF. The Declaration calls for the
[a]doption of a system of easily readable and comparable degrees,

also through the implementation of the Diploma Supplement,
in order to promote European citizen’s employability and the
international competitiveness of the European HE system. (Bold
text per original; European Commission, 1999, pp. 3–4 – see also
Table 10.1)
This objective is of particular importance due to its central role within

the overall Bologna Process. Without meeting this objective the over-
all process would be understood to fail. We focus on the processes of
policy formation and implementation from a routines theory perspec-
tive which has received significant attention within the field of organi-
sation studies. Routines theory, as a processual theory, defines routines
as ‘repetitive, recognisable patterns of interdependent actions, carried
out by multiple actors’ (Feldman & Pentland, 2003, p. 95; Pentland &
Feldman, 2005). A notable characteristic of routines is that they have been
shown to reflect continuous change (Feldman, 2000) which can also be
endogenous in nature (Feldman & Pentland, 2003, 2008; Parmigiani &
Howard-Grenville, 2011; Pentland & Feldman, 2005). This provides a
specific theoretical lens to understand, in broad terms, the processes
leading to the development and implementation of the Bologna Process
as it conceived to develop an EQF. In adopting this approach we were
cognisant of Kuhn’s (1962) understanding of paradigms which might,
for example, inform ontological perspectives of how we understand the
commensurability or incommensurability of policy change theories.
While considering theories reflecting the nature of change for our study
we noted Healy and Reynolds’ (1999, p. 2) view on the power of para-
digms in the context of ‘one’s “world view” . . . [and how] they under-
pin decisions concerning what constitutes a problem, how it should be
approached, [and] what action should be taken’.
Sharon Feeney and Conor Horan 191
For the purpose of this chapter, the developmental process towards

the design and adoption of the EQF is considered a ‘sub-routine’ to meet
the first objective within the overall Bologna policy process, under-
stood here as the focal routine consistent with the approach taken in
Feldman’s Residential Life research (Feldman, 2000, 2003, 2004). To
understand this sub-routine we bring to the fore the artifacts, including
declarations and communiqués, that were regularly produced between
1998 and 2005. We seek to understand how these particular artifacts
when produced influenced the actions taken by the signatories. We
understand these actions as performances that further policy formation
and implementation of the Bologna Process. We consider this policy
formulation and implementation to be analogous to performance of a
routine. This in turn influences how the signatories ostensibly under-
stood the abstract nature of the Bologna objectives and specifically the
goal of harmonising qualifications towards an agreed EQF.
In meeting the definition of routines, therefore, it is important to
recognise that the Bologna Process initiated a period of policy change
where a recognisable pattern of interdependent actions, which were car-
ried out by multiple actors during phases of policy formulation, were
followed by implementation. This recognisable and repetitive pattern
was captured and illustrated in declarations and communiqués that
were produced on a regular basis. We have chosen the signing of the
Bologna Declaration in 1999 as the starting point of the sub-routine
under scrutiny. A pattern of actions emerged as a consequence of the
bi-annual meetings with the resultant publication of declarations and
communiqués. These artifacts represented snapshots, or codifications,
of the process, allowing divergences of what is captured in the artifacts
to be reviewed and compared against multiple understandings of the
routine (i.e. the ostensive) and actions taken (i.e. performances). This
would direct attention to areas that needed further attention (i.e. policy
formation and/or implementation). The aim of harmonising qualifica-
tions across signatories contributes to an ultimate goal for a suitable
framework (i.e. EQF) that was ostensibly repeated in these artifacts,
thereby ensuring this goal remained a priority for the first decade of the
Bologna Process. This pattern of activity resulted in action plans being
specified and priorities being listed so that all of the interdependent
activities among the signatories, or members states, that were under-
taken within the Bologna Process were consistent with the broader aim
of creating a single higher education area (HEA) in Europe by 2010.
The Bologna policy process involved numerous signatories extend-
ing beyond EU member states between 1999 and 2005 (including
Azerbaijan, Kazakhstan, the Russian Federation, and Ukraine). Different

tasks and activities were undertaken by signatories to implement recom-
mendations across the various declarations and communiqués. Routines
theory is adopted as the actions were taken by multiple interdependent
actors in pursuit of a common goal: a single, approved, EQF to enable
recognition and comparability of higher education (HE) awards.
The ostensive–performative theory of routines

The ostensive–performative theory of routines, as a theoretical lens,
informs our understanding of the role of artifacts in supporting the
emergence of the Bologna Process as a routine. Within the broader
Bologna Process as a routine we are focused on the sub-routine relat-
ing to the first policy objective of harmonising qualifications towards
an EQF. Indeed, Hall’s (1993, p. 278) view of policy making as being an
ongoing dynamic between an interpretative framework of ideas and the
goals and instruments of policy cannot be ignored in this context.
Early conceptualisations of routines find their origins in Stene (1940)
and popularity in Nelson and Winters (1982). Here routines were
conceptualised broadly, using the metaphors of individual habits as
programs and/or as genes (Feldman & Pentland, 2003, p. 97). These met-
aphors suggest an ‘image of routines as being relatively fixed, unchang-
ing objects’ (Feldman & Pentland, 2003, p. 97). It is this assumption that
routines are inert, stable entities, or objects that come under scrutiny
with paradigmatic consequences for furthering analysis. More recently,
routines have been shown to continuously change (cf. Feldman, 2000)
and to endogenously change (Feldman & Pentland, 2003; Pentland &
Feldman, 2005). In their reconceptualisation of routines theory, Feldman
and Pentland present routines as being mutually constitutive ostensive
and performative aspects where ‘each part [of the routine] is necessary,
but neither part alone is sufficient to explain (or even describe) the
properties of the phenomenon’ (Feldman & Pentland, 2003, p. 95). Here
each part is necessary to appreciate routines as a source of flexibility
and change. In further developing this, the role of materials or artifacts
was included. How artifacts diverge from, and interact with, both the
ostensive and performative aspects is further explored (cf. Pentland &
Feldman, 2005).
The essence of the ostensive aspect of routines is that it allows actors
to be guided by the routine, and enables them to make reference to the
routine so as to account for their actions and describe what was per-
formed. This reflects the ‘know that’ aspect of the routine (Ryle, 1949).
However, multiple understandings of the routine, the tasks that cohere
to it from a set of possibilities, allow multiple actors to make sense of

the processes they are involved in from their own perspective. These
multiple understandings influence the process of policy formation as
abstract explanations of the objective, or goal, of a policy being adopted
and retained. This abstract understanding becomes a representation of
the policy objectives, or routine goal, as captured in various artifacts.
The performative aspects of a routine, on the other hand, are created,
maintained, or modified as they are performed or enacted. They also
reflect the ‘know how’ of the routine (Ryle, 1949). As tasks and activities
are carried out, often towards an ideal goal or objective, they in turn
influence what is ostensibly understood by the routine. Therefore, the
performance of a routine can be identified in the tasks and activities that
are enacted to implement a policy from a set of possibilities. However,
to compound this process, the development of policy (i.e. the selection
and retention of policy goals from a set of possibilities) also reflects a
sense of performance prior to traditional implementation, or enact-
ment, as we would typically understand it in an organisational setting.
The importance of artifacts in routines has received limited atten-
tion (D’Adderio, 2008, 2011). While tentatively connected with routine
maintenance, the artifacts in this context have brought about change
and provide a held expectation that results upon publication and guid-
ance for implementation. Artifacts such as communiqués provided
codified snapshots of the ‘abstract’ chosen from sets of possibilities and
reflect a record of policy formation and indeed provide blueprints for
policy implementation (i.e. guidance in relation to what should be done
next).
Ostensive policy formation and performance

as implementation
We loosely associate the ostensive aspects of the routine with policy
formation, while similarly equating policy implementation with the
performative aspects of the routine (Figure 10.1). Within this mutu-
ally constitutive theory artifacts play an important role in highlight-
ing the divergence between ostensive and performative aspects of
routines (Pentland & Feldman, 2005). We explore the mutually con-
stitutive relationship between policy formation (i.e. the ostensive) and
policy implementation (i.e. the performative), and bring into the fore-
ground, in a gestalt way, the integral role played by available artifacts
which are representative of the overall harmonisation sub-routine.
The ostensive–performative aspects remain important but in the back-
ground. While the artifacts produced within a routine are periodic, in
Figure 10.1 The ostensive–performative theory of routines
this context it would be incorrect to suggest they represent the total-

ity of the routine, as they always imperfectly capture varied ostensive
understandings of the actors involved. Indeed performances may not
always be guided by, or reflect, that captured within the artifacts them-
selves. This ostensive–performative view of routines enables us to gain
greater insights into the policy formulation phase relating to the first
objective of the Bologna Process. We make two main contributions in
relation to continuous change and the application of routines theory
in policy studies.
Continuous change
Firstly, by using routines theory we highlight the subtleties of continu-
ous change, which often reflect elements of both endogenous and pro-
cessual change over time in a more nuanced way, not provided for in
the more hegemonic theories commonly used in policy studies such
as punctuated equilibrium (Jones & Baumgartner, 2012; True, Jones, &
Baumgartner, 1999). This raises the question as to how we understand
change and whether change is episodic or incremental in nature, as
illustrated in punctuated equilibrium. Feldman and Pentland (2003)
draw the distinction between the two approaches by noting that
theories of episodic change, or ‘punctuated equilibrium’, use the

static conceptualisation of organisational routines as a basis for the
argument that organisations have structural inertia. According to this
perspective on change, routine create inertia and resist change as long
as possible, until exogenous forces overwhelm the structure. (p. 114)
Secondly, theories of continuous change tend to ignore these inertial

qualities. From this perspective organisations adapt their routines con-
tinuously to environmental pressures, while also accepting that change
can occur endogenously. Routine theory argues that processes change
from within and do so continuously as they are performed. The osten-
sive–performative ontology, outlined above, helps to explain why some
routines may or may not change (i.e. show inertia or not). Here con-
tinuous change is dependent on the ecological processes of variation,
selection, and retention (Feldman & Pentland, 2008) as they take place
between ostensive and performative aspects. This differs from seeing
processes as deliberately incremental in how they change and also dif-
fers from the crisis bias.
Paradigmatic & theoretical commensurability

Kuhn (1962) distinguishes between paradigms by focusing on the com-
mensurability of theories and whether the anomalies they reveal can
be explained within a particular theoretical viewpoint. Paradigm shifts,
for example, might occur within what he termed ‘normal science’.
However, a new paradigm is required when evidence continues to ‘over-
whelm the structure’ (echoing the phrase used by Feldman and Pentland
in the quote used in the previous section above [2003, p. 114]). By using
the ostensive–performative conceptualisation of routines theory we
force the question of theoretical commensurability. At first glance con-
tinuous change in routines theory is incommensurate with punctuated
equilibrium, with the latter focusing attention on observed incremental
changes, for example, the release of a policy artifact that might result
in critical junctures (Capoccia & Kelemen, 2007; Hogan, 2006; Vohora,
Wright, & Lockett, 2004) or locking implementation within a dependent
path (Pierson, 2000). This perspective suggests that change does not occur
between ‘punctuations’. Routines theory, however, reveals that how we
understand and perform routines continuously results in endogenous
variations. This would suggest that punctuated equilibrium is a blunt
instrument relying on observed performed change and ignoring change
which might occur within unobserved variations in the abstract osten-
sive understanding of policy formation. In relation to policy formation
and implementation we contend that at a higher level of granularity
the theories of punctuated equilibrium and the ostensive–performative
theory of routines are thus incommensurable. However, at a day-to-day
or practice level of granularity punctuated change can also and easily be
explained using routines theory. The implication of this is that by adopt-
ing a routine perspective we unpack the black box that is the routine and
can focus on continuous and endogenous aspects of change rather than

relying solely on change as a result of observable exogenous shocks. As
a consequence we argue that routines theory is a more holistic theory
capable of encompassing episodic change at macro levels of granularity
within data analysis as well as what might be understood as continuous
change at micro levels of granularity based on practice.
Similarly routines theory views stability and change as a duality rather
than as a dualism. Rather than opposites, stability and change are under-
stood as the two sides of a coin (Farjoun, 2010) and are mutually con-
stitutive. In implication, routines theory contributes to the discussion
of how policy formation and implementation unfolds but distinguishes
its contribution from hegemonic theories used within policy studies in
relation to its approach to how we understand and address change. Here
processes of policy formation and implementation – rather than being
punctuated, event based, inert, and path dependent – are regarded as
sources of endogenous and continuous change.
The inter-organisational nature of routines

In spite of calls to further explore the nature of inter-organisational rou-
tines (Tsoukas & Mylonopoulos, 2004) and attempts to consider rou-
tines across organisations (Zollo, Reuer, & Singh, 2002), little empirical
research has been conducted to consider the ecology of routines beyond
single organisations or cases (Pentland, 2004). We argue that the arbitrary
nature of teams, business units, and organisations as structures reflect
different levels of granularity across organisational settings. The Bologna
Process as a routine reflects definitional presence of multiple actors
involved in interdependent actions towards an emergent and common
goal. By elevating the role of artifacts, we take a particular perspective
of a policy formation and implementation routine towards harmoni-
sation of educational standards. The ostensive–performative approach
to routines theory accepts the plurality of goals, as multiple osten-
sive understandings of policy objectives may be distributed across the
actors involved (Feldman, 2000; Feldman & Pentland, 2003; Pentland &
Feldman, 2005). It is this variation of abstract understandings which in
turn influences and is mutually constituted by varied performances. The
first objective of the Bologna Process was to harmonise policy towards an
EQF. Within the overarching goals, or objectives, of the Bologna Process,
this specific goal to arrive at an EQF is the focus of this chapter.
This goal, or objective, could have been achieved in multiple ways due
to variations in both the ostensive understandings of the process and the
performance of signatories to achieve desired outcomes (Feldman, 2000).
To reiterate, while policy harmonisation represents the goal, it should

not be confused with the routine itself. To bring this about artifacts play
important roles in codifying ostensive policy understandings whilst at
the same time providing guidance in relation to policy implementation.
We describe this process and highlight how the production of the com-
muniqués, or artifacts, influenced and contributed to harmonisation
towards an agreed EQF in 2005.
The context of an emerging Bologna Process
The origin of the Bologna Process

The origin of the Bologna Process can be traced back to the Single
European Act in 1987 which stated that ‘the internal market shall com-
prise an area without frontiers in which the free movement of goods, per-
sons, services and capital is ensured’ (European Commission, 1987, p. 6).
The issue of developing a system of comparability of HE awards through-
out Europe is not a new one. The Memorandum on Higher Education
in the European Commission (1991) first established this as a principle
which was adopted by EU member states in the Maastricht Treaty in
1992 (Brennan, 1993, p. 10). This Memorandum stated that the internal
European market
depends to a large degree on the capacity of people to accept and

exploit the freedoms accorded to them, to be able, where necessary or
desired, to live and work in other Member States of the Community, to
trade and offer services across national boundaries and to co-operate
effectively with partners in other parts of the Community in agree-
ments, contracts and joint undertakings. (European Commission,
1991, p. 6)
The free movement of people, on an individual and collective basis,

was facilitated by a European Council Directive on ‘a general system
for the recognition of HE diplomas awarded on completion of profes-
sional education and training of at least three years duration’, and by
other European Directives governing the mutual recognition of quali-
fications for professional purposes (European Commission, 1991, p. 7).
In practical terms, therefore, the Memorandum on HE in the European
Community necessitated various stakeholders, or actors, making judge-
ments about the quality of HE programmes and the quality and com-
parability of qualifications between member states, with no apparent
criteria or information to assist or enable them to do so (Brennan, 1993,

p. 15). Consequently, it is not surprising that the Memorandum was
followed up with other policies and practices from the EU Commission,
particularly those that have occurred since the ‘Sorbonne Declaration’
in 1998 (Feeney, 2014, p. 21).
Similarly the use of Directives as a legal mechanism allowed mem-
ber states who might have similar laws in place to harmonise them to
fulfil their obligations under various EU treaties. The Directives, unlike
EU Regulations, are not directly applicable requiring secondary legisla-
tion (Byrne & McCutcheon, 2001, p. 662) and are therefore subject to
local interpretation which may be influenced by culturally specific and
institutional considerations unique to each member state and signa-
tory. It is this context of multiple ostensive understandings of policy
that leads to variations in how that policy is implemented across mem-
ber states. The Bologna Declaration (European Commission, 1999),
like the Sorbonne Declaration (European Commission, 1998) before it,
acknowledged that the objectives mentioned above would be under-
taken ‘within the framework of national competencies and taking full
respect of the diversity of cultures, languages, national education sys-
tems and of university autonomy’ (European Commission, 1999, p. 4).
The Bologna Declaration formally included follow-up processes, with
the written intention ‘to pursue the ways of intergovernmental co-
operation, together with those of non-governmental European organi-
sations with competence on HE’.
The Sorbonne Declaration

Before the Bologna Process was underway four ministers in charge of HE
(from France, Germany, Italy, and the United Kingdom) signed a joint
declaration on what they referred to as a ‘harmonisation of the architec-
ture of the European higher education system’ (European Commission,
1998, p. 1). These ministers acted as political entrepreneurs and ‘agents
of change’ in transforming the idea of a single, comparable framework
of HE awards into an agreed policy process (Hogan & Feeney, 2012). The
Sorbonne Declaration committed signatories to ‘a system, in which two
main cycles, undergraduate and graduate, should be recognised for inter-
national comparison and equivalence’ (European Commission, 1998,
p. 2). This Declaration was, by design, outside of any formal EU context
and is seen as being the precursor of the Bologna Process (Witte, 2006,
p. 124); however, it served to ostensibly narrow the focus on policy for-
mation around a two-cycle format for harmonisation. This was followed
up with the Bologna Declaration in 1999 which committed signatories
to develop a common ‘architecture’ of degrees and the promotion of

European cooperation with a view to developing comparable criteria.
The first objective, the subject of this chapter, provides closer guidance
to implementation by focusing on comparability without being overly
prescriptive (Table 10.1).
The Bologna Declaration

With the signing of the European Commission Sorbonne Declaration
(1998) the policy vehicle that has become known as the ‘Bologna Process’
commenced in earnest. Building on discussions of a general system which
acknowledged cycles of at least three years, the Sorbonne Declaration
committed signatories to developing ‘a system, in which two main
cycles, undergraduate and graduate, should be recognised for interna-
tional comparison and equivalence’ (European Commission, 1998, p. 2).
On 19 June 1999 the Declaration on a ‘European Space for HE’ was signed
in Bologna, Italy. This Declaration forms a written commitment by each
of the then 29 signatories to reform the structures of their own national
HE systems in a convergent way, while at the same time recognising
the fundamental principles of autonomy and diversity for HE in each
nation-state. To this end, the Bologna Declaration formulated a set of
broad aims, including the aim to construct a ‘European HE Area’, in order
to ‘promote citizens’ mobility and employability and the Continent’s
overall development’ (European Commission, 1999, pp. 1–2). In order
to meet the above aims, the Bologna Declaration affirmed the intention
of all signatories to ‘engage in co-ordinating our policies to reach in the
short term, and in any case within the first decade of the third millen-
nium’ a range of objectives, which have particular relevance to the estab-
lishment of a European area of HE (European Commission, 1999, p. 3).
A total of six objectives, action lines, or goals, to be completed by all
signatories by 2010 were presented (see Table 10.1).
The first objective identified above is the focus of this study, as it refers
to the need to be able to adopt a system of easily readable and comparable
degrees. A series of communiqués were then produced over regular time-
frames (bi-annually) in the hope of providing guidance towards imple-
mentation. Resources were employed in terms of policy meetings and
conferences, to support the production and agreement of their content.
This reflected a process of policy formation which enabled signatories
to be guided by the Bologna objectives towards policy implementation.
Implementing the Bologna objectives took on different forms in order
to account for the diversity among the signatories of differing cultures,
languages, education systems, and the nature of university autonomy.
Table 10.1 The objectives of the Bologna Declaration, 1999
Objective #1 Adoption of a system of easily readable and comparable

degrees, also through the implementation of the Diploma
Supplement, in order to promote European citizens
employability and the international competitiveness of the
European HE system.
Objective #2 Adoption of a system essentially based on two main cycles,
undergraduate and graduate. Access to the second cycle shall
require successful completion of first cycle studies, lasting a
minimum of three years. The degree awarded after the first
cycle shall also be relevant to the European labour market as
an appropriate level of qualification. The second cycle should
lead to the master and/or doctorate degree as in many
European countries.
Objective #3 Establishment of a system of credits – such as in the European
Credit Transfer and Accumulation System (ECTS) – as a proper
means of promoting the most widespread student mobility.
Credits could also be acquired in non-HE contexts, including
lifelong learning, provided they are recognised by receiving
Universities concerned.
Objective #4 Promotion of mobility by overcoming obstacles to the effective
exercise of free movement with particular attention to:
s for students, access to study and training opportunities and
to related services;
s for teachers, researchers and administrative staff,
recognition and valorisation of periods spent in a
European context researching, teaching and training,
without prejudicing their statutory rights.
Objective #5 Promotion of European cooperation in quality assurance with
a view to developing comparable criteria and methodologies.
Objective #6 Promotion of the necessary European dimensions in HE,
particularly with regards to curricular development,
inter-institutional cooperation, mobility schemes and
integrated programmes of study, training and research.
Note: Bold text is shown as per original document.

Source: Bologna Declaration (1999, pp. 3–4).
These multiple understandings reflected various perceptions of the

meaning of a common European Higher Education Area (EHEA), and
specifically the harmonising actions to be taken to implement and
achieve a system of easily readable and comparable degrees. We trace,
chronologically, the development of abstract ideas that facilitated policy
formation in this chapter. We focus our attention on available artifacts,
or communiqués, which reflect ostensive description which serves to
guide and inform this process of harmonisation (Table 10.2).
Table 10.2 Artifacts that relate to Objective #1 of the Bologna Declaration:
A system of easily readable and comparable degrees
Treaties and Legislative Implications for Refining Ostensive

Year Instruments Understandings
1987 Single European Act Facilitating the free movement of goods,

persons, services, and capital
1991 Memorandum on Higher A more open and accessible European market
Education (HE) in the for HE
European Union (EU)
1991 European Council A general system for the mutual recognition of
Directive qualifications of HE diplomas; cycles of at
least three years referred to as a loose criterion
1992 The Maastricht Treaty Freedom to trade and offer services across
national boundaries
1998 Sorbonne Declaration s A harmonisation of HE architecture
s A narrowing of focus around two main cycles
(undergraduate and graduate) to be recognised
for international comparison and equivalence
1999 Bologna Declaration s Six explicit objectives put into place to guide
implementation. Here we see clear action
paths to clarify the ostensive understanding
of the Bologna Process (see Table 10.1).
s Guidance without prescription
2001 Prague Communiqué s Six explicit objectives confirmed
s Three additional objectives were also
added in relation to lifelong learning, HE
institutions and students and promotion of
the European HE Area (EHEA).
s Resources added with the establishment of
the Bologna Follow-Up Group (BFUG)
2003 Berlin Communiqué s Reference is made to a need for a
framework of comparable and compatible
qualifications which describes qualifications
in terms of workload, level, learning
outcomes, competences, and profile.
2004 s Tighter guidance of the abstract
understanding around ‘learning outcomes’
and ‘competence’ as the way to go to
establish comparability – closer to policy
implementation (moving from policy
formation/formulation)
2005 Bergen Communiqué s Adoption of the overall framework for
qualifications in the EHEA (the European
Qualifications Framework)
s Commitment to elaborate national
frameworks for qualifications for
compatibility with the EQF by 2010, and to
have commenced the work relating to this
by 2007
The Prague Communiqué

The Prague conference was held on 19 May 2001, resulting in the Prague
Communiqué (2001) and confirming the six original objectives of the
Bologna Declaration, whilst also highlighting progress made to date
regarding many of them. An additional three objectives were added
which related to lifelong learning, HE institutions and students, and
promoting the attractiveness of the EHEA. In relation to the first objec-
tive of adopting a system of easily readable and comparable degrees,
the ministers encouraged universities and other HE institutions to take
advantage of existing legislation to meet this objective. In addition,
they called upon organisations and networks such as National Academic
Recognition Information Centres (NARIC) and European Network of
Information Centres (ENIC) to assist in promoting simple, equitable,
and transparent recognition criteria for comparison purposes. While
committing themselves to continuing with the Bologna Process, it was
decided that the next follow-up meeting, to be held in Berlin in 2003,
would review progress and would establish priorities for the next stages
of the process. In addition, it was decided that there should be a formal-
ised system of continuity between the conferences, and to this end a
Bologna follow-up group (BFUG) was established. This group would be
responsible for the future development of the process, as well as having
an ongoing monitoring role (Feeney, 2014, p. 26).
The period between the Prague and Berlin conferences proved to be a
busy time for the newly established BFUG with several seminars, stud-
ies, and position papers regarding progress in relation to the objectives
of the Bologna Declaration (Witte, 2006, p. 137). According to Zgaga
(2003, p. 7), ten official Bologna seminars took place during the period
(March 2002–June 2003) between the Prague and Berlin conferences.
These seminars, which were attended by over one thousand partici-
pants, ‘developed into a unique pan-European forum, which reflects
the “snowball effect” of the Bologna Process’ (Zgaga, 2003, p. 8). Much
progress was made regarding the objectives of the Bologna Process,
with many of the signatories of the Declaration preparing and imple-
menting significant reforms in their own national HE systems to meet
the challenges posed by the Bologna objectives. In addition, many
changes were also taking place in vocational education and training in
Europe as a result of other conferences, communiqués, and declarations
including such initiatives as the Lisbon Agenda promoting the idea of
a knowledge economy. The most relevant of these, for the purpose of
this study, was the Copenhagen seminar. Titled ‘Qualification Structures
in HE in Europe’, the seminar dealt with the issue of developing a HE
qualifications framework for Europe, focusing its attention on curric-

ulum planning, quality assurance, and recognition of awards (Zgaga,
2003, pp. 16–17).
Following the Prague Communiqué in 2001, two countries developed
national level frameworks of qualifications for their HE awards: Denmark
and Ireland. The issue of creating an overarching framework of qualifica-
tions was the focus of a ‘Joint Quality Initiative informal group’ which
met in Dublin on 18 October 2004. This informal group comprised
representatives from national and European agencies, including the
European Universities’ Association, the Belgian Ministry of Education,
the Dutch Ministry of Education, the UK Quality Assurance Agency, the
National Qualifications Authority of Ireland, the Irish Higher Education
and Training Awards Council, amongst other agencies. The group
focused on an ‘outcomes’-based approach which had been adopted by
the Danish and Irish agencies. This ostensibly clarified, out of the sets of
possibilities, an approach to inform performances which built upon the
approaches already used by Denmark and Ireland in their frameworks.
Previously, the multiple sets of possibilities of different approaches to
achieving mutual recognition of HE awards, which arguably stymied
performances before the release of the Prague Communiqué, served to
leave discussions in the realm of policy formation. However, the group
agreed a set of Award Descriptors, known as the ‘Dublin Descriptors’, to
clarify the outcomes that should be considered for each type of award.
This artifact brought ostensive understandings closer to policy imple-
mentation and an agreement framework. The framework of descriptors
provided a mechanism to recognise national competencies, whilst also
allowing for the diversity of cultures, languages, and national educa-
tion systems as mentioned in the Bologna Declaration. The agreed set of
qualifications descriptors are summarised in Table 10.3.
Berlin Communiqué
The European ministers of education from the 33 signatory countries
to the Bologna Process met in Berlin on 19 September 2003, for the
‘Realising the European HE Area’ conference. In addition, 7 new countries
were admitted to the Bologna Process in Berlin, giving a total of 40 coun-
tries which had committed to achieving the outcomes of the Bologna
Process (Berlin Communiqué, 2003, p. 8). The Berlin Communiqué
refers to the need of an overarching framework of qualifications for the
EHEA which should describe ‘qualifications in terms of workload, level,
learning outcomes, competences and profile’ (Berlin Communiqué,
2003, p. 4). The Berlin Communiqué outlined a work programme for
Table 10.3 Qualifications descriptors (known as the ‘Dublin Descriptors’)
Qualifications that signify completion of the higher education short cycle

(within the first cycle) are awarded to students who:
s have demonstrated knowledge and understanding in a field of study that
builds upon general secondary education and is typically at a level supported
by advanced textbooks; such knowledge provides an underpinning for a field
of work or vocation, personal development, and further studies to complete
the first cycle;
s can apply their knowledge and understanding in occupational contexts;
s have the ability to identify and use data to formulate responses to well-
defined concrete and abstract problems;
s can communicate about their understanding, skills and activities, with peers,
supervisors and clients;
s have the learning skills to undertake further studies with some autonomy.
Qualifications that signify completion of the first cycle are awarded to
students who:
s have demonstrated knowledge and understanding in a field of study that
builds upon their general secondary education, and is typically at a level
that, whilst supported by advanced textbooks, includes some aspects that
will be informed by knowledge of the forefront of their field of study;
s can apply their knowledge and understanding in a manner that indicates
a professional approach to their work or vocation, and have competences
typically demonstrated through devising and sustaining arguments and
solving problems within their field of study;
s have the ability to gather and interpret relevant data (usually within their
field of study) to inform judgements that include reflection on relevant
social, scientific or ethical issues;
s can communicate information, ideas, problems and solutions to both
specialist and non-specialist audiences;
s have developed those learning skills that are necessary for them to
continue to undertake further study with a high degree of autonomy.
Qualifications that signify completion of the second cycle are awarded to
students who:
s have demonstrated knowledge and understanding that is founded upon
and extends and/or enhances that typically associated with Bachelor’s level,
and that provides a basis or opportunity for originality in developing and/or
applying ideas, often within a research context;
s can apply their knowledge and understanding, and problem solving abilities
in new or unfamiliar environments within broader (or multidisciplinary)
contexts related to their field of study;
s have the ability to integrate knowledge and handle complexity, and
formulate judgements with incomplete or limited information, but that
include reflecting on social and ethical responsibilities linked to the
application of their knowledge and judgements;
Table 10.3 (continued)
s can communicate their conclusions, and the knowledge and rationale

underpinning these, to specialist and non-specialist audiences clearly and
unambiguously;
s have the learning skills to allow them to continue to study in a manner that
may be largely self-directed or autonomous.
Qualifications that signify completion of the third cycle are awarded to
students who:
s have demonstrated a systematic understanding of a field of study and
mastery of the skills and methods of research associated with that field;
s have demonstrated the ability to conceive, design, implement and adapt a
substantial process of research with scholarly integrity;
s have made a contribution through original research that extends the frontier
of knowledge by developing a substantial body of work, some of which
merits national or international refereed publication;
s are capable of critical analysis, evaluation and synthesis of new and complex
ideas;
s can communicate with their peers, the larger scholarly community and with
society in general about their areas of expertise;
s can be expected to be able to promote, within academic and professional
contexts, technological, social or cultural advancement in a knowledge based
society.
Glossary:
1. The word ‘professional’ is used in the descriptors in its broadest sense,
relating to those attributes relevant to undertaking work or a vocation and
that involves the application of some aspects of advanced learning. It is
not used with regard to those specific requirements relating to regulated
professions. The latter may be identified with the profile/specification.
2. The word ‘competence’ is used in the descriptors in its broadest sense,
allowing for gradation of abilities or skills. It is not used in the narrower
sense identified solely on the basis of a ‘yes/no’ assessment.
3. The word ‘research’ is used to cover a wide variety of activities, what the
context often related to a field of study; the term is used here to represent
a careful study or investigation based on a systematic understanding and
critical awareness of knowledge. The word is used in an inclusive way to
accommodate the range of activities that support original and innovative
work in the whole range of academic, professional and technological fields,
including the humanities, and traditional, performing, and other creative
arts. It is not used in any limited or restricted sense, or relating solely to a
traditional ‘scientific method’.
Source: Dublin Descriptors (2004).
the BFUG for the period from 2003 to 2005, which included a monitor-
ing role for the European Network for Quality Assurance (ENQA) pro-
ject on quality assurance. Following on from the Berlin conference the
BFUG established a working group in March 2004, to carry out a ‘stock-

taking’ exercise in order to complete an evaluation study (Reinalda &
Kulesza, 2005) regarding progress by individual countries in meeting the
Bologna objectives. The study was primarily based on information from
Eurydice (2005) and national reports from the signatory countries. This
process reviewed the variations in implementations, as well as issues of
particular importance to each signatory reflecting their interpretation of
the importance of specific tasks in turns of implementing the Bologna
objectives. It is these variations in the tasks and activities required to
meet multiple understandings of ideal goals, in the face of actual out-
comes, that illustrate the mutual constituent nature of policy formation
and implementation.
Again, the six original objectives of the Bologna Declaration were reaf-
firmed in the Berlin Communiqué (2003). Given the flurry of activity
between the Prague and Berlin meetings, it was agreed that there would
be a two-year action plan agreed in order to integrate all of the progress
up to that point. The Bologna Process was reviewed and reaffirmed for a
third time. Intermediate priorities were identified for action in relation
to quality assurance, degree structure, the adoption of a system based
on two main cycles, and the promotion of mobility. The emergence
of ‘intermediate objectives’ within the two-year cycles of the Bologna
Process suggested a more fine-tuned approach to implementation, as
operational aspects could be considered within a broader agreed frame-
work of understanding.
Bergen Communiqué
The European ministers met again in Bergen in May 2005, for the confer-
ence titled ‘The European HE Area – Achieving the Goals’. Five new partic-
ipating countries joined the Bologna Process bringing the total number of
signatories to 45. A mid-term review of the Bologna Process was conducted
to help set ‘goals and priorities towards 2010’ (Bergen Communiqué, 2005,
p. 1). Of particular relevance to this chapter is the fact that the conference
adopted the ‘overall framework for qualifications in the EHEA’ and recog-
nised that the framework was focused on ‘learning outcomes and compe-
tences’ (Bergen Communiqué, 2005, p. 2). These competencies could be
traced back to a ‘competences’-based approach adopted in Berlin 2003.
The ministers went on to commit themselves to ‘elaborating national
frameworks for qualifications compatible with the overarching framework
for qualifications in the EHEA by 2010, and to having started work on
this by 2007’ (Bergen Communiqué, 2005, p. 2). This is the critical period
for our study as the Bergen Communiqué effectively adopted the EFQ.
This artifact represents a single approach/benchmark against which all

national level frameworks would now be compared and measured, based
on the understanding of comparability within a two-cycle formation of
undergraduate and graduate degrees and based on the ostensive focus of
an outcomes- and competences-based approach.
Findings and discussion
Foundation
The Bologna Process, as described above, illustrates the emergent
nature of a policy formation and implementation routine towards
harmonising standards in an EQF which was accepted and adopted
in 2005. A portfolio of artifacts provided the early parameters for
meeting the first Bologna objective within the broad ostensive under-
standing that cycles should include at least three years (European
Commission, 1991) and a more focused conceptualisation of harmo-
nisation within the context of two main cycles as accepted within
the Sorbonne Declaration (1998). This gained legitimacy within the
Bologna Declaration (1999) (Table 10.1, Objective #1) where ostensive
understandings of multiple actors would be guided without being pre-
scribed (Table 10.2). It is at this point that we see the emergence of a
fledgling routine informed by tacit understandings guiding multiple
actors towards policy formation.
Emergence and convergence

From the Bologna Declaration a bi-annual process resulting in commu-
niqués began to emerge. It appears to be at this stage that the Bologna
Process, as a ‘routine’, was only emerging due to the significance of
repetition being built into the process. Repeated patterns of actions
involving multiple actors refining and developing a common osten-
sive understanding of the nature of harmonisation began to emerge.
The goal to adopt a harmonised system of easily readable and com-
parable degrees (Table 10.1, Objective #1) became attainable within
agreed abstractions. Six cycles of ostensive policy formation coincid-
ing with performative policy implementation with declarations and
communiqués at their centre can be seen (Table 10.2). Here we see the
emergence of a policy formation/implementation process and its for-
malisations with regular patterns of actions, as evidenced by meetings
such as those resulting in the Dublin Descriptors and the allocation
of resources to the BFUG to monitor progress. In this context that we
have glimpses of the emergence of a routine as resources are allocated,
actors recruited, tasks legitimately established culminating in agreed

pattern of actions. While multiple participant actors were recruited,
and additional signatories were added, progression towards imple-
mentation was codified in the routine’s artifacts (e.g. declarations and
communiqués).
We interpret the emergence of the policy formation and the policy
implementation routine as being supported by, and linked to, how mul-
tiple ostensive accounts meet the goal of harmonising systems of easily
readable and comparable degrees across multiple signatories. This is bol-
stered by a refining of how the goal is understood. Two ways to refine
the understanding of the goal were found in the data: firstly, by focusing
on an agreed three-year undergraduate cycle; and secondly, by recog-
nising both undergraduate and graduate cycles. These were ostensibly
embedded in available declarations and communiqués. It is within these
six cycles that we can consider how the harmonising routine emerges
and how multiple ostensive understandings begin to converge between
1999 and 2005 (when the EQF was adopted).
From formation to implementation

The recursive pattern of activities resulted in the production of bi-
annual communiqués, or artifacts, which in turn informed the multi-
ple actors’ interpretation of the routine towards the EQF (Figure 10.2).
A communiqué was produced every two years to the present. However,
these artifacts are of particular importance in relation to the sub-routines
to achieve the broad aim of creating a single EHEA, and specifically to
approve the EQF. However, the abstract nature of the goal of the rou-
tine remained linked to concepts for framing the formation of policy.
As additional communiqués were produced, the performative nature of
the routine was informed more towards policy implementation. As the
routine emerged it guided the signatory actors towards greater clarity
for implementation rather than remaining in the abstract pertaining to
policy formation.
Firstly, the nature of performances in this policy context is that perfor-
mance is not just policy implementation, but also, and simultaneously,
policy formation (Figure 10.2). The performance, as described by tasks,
activities, or enactments as attempted by the signatories, constitute just
a part of the overall picture. Alongside policy implementation tasks and
activities are also policy formation activities which are linked to develop-
ing the abstract understandings, or the multiple ostensive understand-
ings, instigated by the Bologna Process itself and by the attempts to
make sense of what the objectives of the process mean for each country.
Figure 10.2 Convergence from policy formation to policy implementation
This ‘sensemaking’ (Weick, Sutcliffe, & Obstfeld, 2005) is also per-

formative. The implication for routines theory, as applied in a policy
(formation and implementation) context, is that the overlapping and
mutually constitutive nature of the ostensive and performative theory
is even more intertwined. The second implication is the importance of
continuous change as a lens to understand the Bologna Process, the sub-
routine of harmonisation towards a single agreed EQF. In the same way
that policy formation and implementation are continuously changing
rather than being incremental or punctuated in nature, the ostensive
understanding of the routine continuously guides how performances
are enacted. This process can be interpreted with the introduction of an

‘outcomes’- and ‘competencies’-based approach to achieving the goal of
the sub-routine.
What is of interest here is how this routine hands off the artifacts and
how the artifacts can enable, or prevent, action as the signatories seek
to have their individual implementation of the Bologna Process guided
by various artifacts as they are produced. The artifacts, as materials, are
elevated as the primary focus and starting point for the analysis of the
Bologna Process as a routine. Artifacts were created periodically to capture
the ostensive understanding of the goal, and of the Bologna Process in
creating a common EHEA. In so doing the ostensive understanding was
affirmed and reaffirmed over the period from 1999 to 2005, through the
process of codifying the goal of the routine and more refined understand-
ings of the process emerging from diverse beginnings. The plurality of
how signatories understand the overarching goal of the Bologna Process
is reflected in variation in the performances in different member states in
terms of the actions taken to meet the goal of the Bologna Process.
However, plurality of sub-routine goals was also evident, as signatories
placed greater emphasis on different aspects for implementation as they
interpreted the communiqués produced bi-annually. Consequently,
while a uniform goal for the Bologna Process existed, the perceived
understanding of actions to be taken to reach this goal reflected varied
performances. Indeed, whilst the communiqués of interest are not the
routine in and of themselves, they are arguably as close an example in
routines research as they play a significant role within the routine. The
core nature of the artifacts, as they are relied upon by all signatories, also
suggests their importance in reflecting the routine.
Routine enactment, or performance, occurs once the artifacts are
released, while performances were delayed in anticipation of the next
communiqué. Without the presence of the artifacts to guide actions,
little or nothing occurred. Only when released, these communiqués cre-
ated an impetus for enacting and performing the Bologna Process policy
routine, thereby illustrating the importance of the presence, or absence,
of the artefact. Local performances, reflecting national and cultural agen-
das, meant that same artifact would result in different local interpreta-
tions and understandings informing actions in various forms across the
signatories. Therefore, multiple interpretations of a single artifact lead
to variations in performances. It is interesting to note that little progress
was made by signatories to put into place frameworks for awards in the
absence of clearer guidelines. This suggests that the abstract nature of
the first objective of the Bologna Process remained vague and difficult
to implement, and its objective for harmonisation mirrored a distrib-

uted sense of responsibility, or ownership, which reflects task legitimacy
being contested and questioned. This would suggest that the lack of
clarity around the ‘how to implement’, or ‘how to perform’, constrained
progress towards the broader Bologna policy process objective of a com-
mon HE area, and specifically in relation to Objective #1, the harmoni-
sation of the qualifications.
The outcomes- and competencies-based approaches

Some reticence on the part of some signatories is evident in the delayed
development of national level qualifications and awards frameworks
being adopted. The Dublin Descriptors represented a performance that
resulted in two pertinent outcomes. Firstly, they served to clarify the
ostensive understanding of the first objective of the Bologna Process.
The outcomes-based approaches enshrined in the Dublin Descriptors
(2004) clarified, aligned, and galvanised the ostensive understandings of
the first objective of the Bologna Process to the exclusion of alternative
approaches. A pathway informed by an outcomes-based approach was
introduced into the process as a possible way forward, or as an approach
that could guide action. Secondly, as a performance this also modified
the fledgling/emerging routine as the performance and introduction
of an ‘outcomes’-based approach focused attention on a mechanism to
achieve a goal, and alternative ostensives could be excluded. This resulted
in understandings of the ‘sets of possibilities’ being clarified. The out-
comes-based approach was formally adopted in the Bergen Communiqué
(2005) and thus was selectively retained, thereby informing the ostensive
understanding of the harmonising routine. The implication of this was
to refine the steps towards the goal of the sub-routine. As routines emerge
variations are selectively retained and can be normative aspects of the
routine itself by informing subsequent performances (Table 10.3).
Similarly, the competencies approach within the Dublin Descriptors
also informed and narrowed the ostensive understandings of how to
move towards harmonisation. Both terms focused the harmonisation
routine in such a way as to inform not just policy formation, but also
to indicate a substantive way forward in policy implementation. Rather
than emergence linking to ‘know that’ the ostensive understanding
began to inform ‘know how’ and signatories could now begin to imple-
ment meaningful actions to harmonise national educational systems to
reflect comparability of degrees with two main cycles. It is this retention
that cumulatively built the ostensive, while refining it towards a clear
framework that was deemed to be acceptable to all signatories.
The Bergen Communiqué – A ‘stocktaking’ of the ostensive

This represents the consolidation of the ostensive, in that the EQF was
adopted and that a ‘single’ approach, or benchmark, was accepted. After
many meetings, seminars, and declarations, a convergence and singu-
lar approach to implementation was arrived at and consolidated osten-
sive understandings, within an agreement framework, were in evidence.
The title of the conference indicated the operational nature of putting in
place the ‘achieving of goals’ and it was this aspect of the routines as scripts
and goals (with outcomes) that is particularly interesting for this chapter.
Acceptance of a European Qualifications Framework

The Bergen Communiqué in 2005 incorporated the adoption of the EQF.
Within this the outcomes- and competencies-based approaches remained
intact as an accepted solution towards harmonisation facilitating the
EQF. It is only at this point that the wider goal of the Bologna Process,
as represented in the first objective, could be achieved by 2010. The
Bologna Declaration identified six objectives, or goals, that needed to be
met in order to create a single EHEA by 2010. The first of these objectives
is the focus of this chapter, as it deals with the ‘adoption of a system of
easily readable and comparable degrees, also through the implementa-
tion of the Diploma Supplement, in order to promote European citizens
employability and the international competiveness of the European HE
system’ (European Commission, 1999). This first objective therefore
focused, in conjunction with the other objectives, primarily on harmo-
nising qualifications across the member states and signatories towards
the EQF – as a mechanism towards meeting the Bologna Declaration’s
stated goal of establishing an EHEA. It is this objective, understood here
as a sub-routine of the broader Bologna Process, that is the focus of this
chapter. The EQF is an overarching framework which facilitates com-
parison of qualifications to be undertaken across national boundaries.
The EQF has eight levels of qualifications, each level having designated
learning outcomes. Within this context, each nation could now formu-
late their unique national level framework of qualifications and map it
onto the EQF. As an overarching ‘meta-framework’ different nations, or
signatories, can use this as a reference point to clarifying issues of com-
parability of all national level awards and qualifications.
Conclusion
Our chapter makes an important contribution to the discussion about

policy paradigms. By elevating artifacts to the foreground we provide
a perspective on the policy routine under scrutiny across what might

otherwise be argued as an inter-organisational context. We contend this
to be an appropriate approach as multiple actors were considered, the
interdependent nature of actions were accounted for in arriving at vari-
ous policy artifacts and communiqués towards an emergent and com-
mon goal. This is a departure from Hall’s (1993) understanding of the
importance of policy goals and instruments, which kept the interpreta-
tive ideas and goals in the foreground.
This chapter also contributed to the discussion on paradigms by argu-
ing for the use of routines theory within policy studies to further explore
policy change (i.e. formation and implementation).
First, we highlight the value of routines theory to explain continu-
ous change, as a more flexible and robust theoretical account of policy
change, in place of hegemonic approaches such as punctuated equi-
librium. While arguably commensurate at higher levels of granularity,
stark differences in the ability of the latter theory to reveal continuous
changes become apparent at an up close, day-to-day, level of granularity.
With the ostensive–performative theory in mind, we argue that punc-
tuated equilibrium, while focusing on observable changes, falls within
‘the performative’ aspects of the harmonisation routine. But punctuated
equilibrium fails to account for the continuous variations or changes
in the ostensive aspects of the routine, that is, how multiple ostensives
were continuously refined and ultimately galvanised by the outcomes-
based approach within a two-cycle framework for qualifications. This
suggests that punctuated equilibrium is limited as it fails to take account
of continuous change within the ostensive aspects of a routine. As dis-
cussed above punctuated equilibrium favours observable changes that
are performed. By using the theoretical lens of routines it highlights
that punctuated equilibrium is silent on continuous change within the
ostensive aspects of a routine (i.e. policy formation).
Secondly, routines theory is making strides in explaining the nature
of endogenous change. Variation and selective retention of tasks and
activities within a routine comes about through a mutually constitutive
relationship between ostensive and performative aspects sometimes cod-
ified in related artifacts. This paradigmatic view of policy formation and
implementation takes into account ‘the ostensive’ aspect which argu-
ably has been overlooked in hegemonic approaches in policy studies.
Thirdly, the adoption of routines theory allowed us to illustrate the
central role that material artifacts play in capturing ongoing policy
development and implementation within an international context such
as the Bologna Process. The presence of the communiqués bridged the
divide between the policy formation as the ostensive aspect with policy
implementation, or the performative aspect, of the harmonising rou-

tines. We illustrated how these aims and goals continued to influence
the Bologna Process by being performed in practice, through the policy
implementation process in relation to the EQF. The repetitive nature of
the routine is prominently illustrated mainly by the bi-annual intro-
duction of communiqués, which we contend are the products of co-
mingling ostensive routines, served to inform performances.
In conclusion the outcomes of the first objective of the Bologna
Process, which culminated in the adoption of the EQF was evaluated
in this chapter. The nature and integral role of a portfolio of artifacts,
including bi-annual communiqués, were assessed as they informed the
emergent aims, goals, and objectives of the Bologna Process and spe-
cifically the routine towards harmonisation of readable and comparable
degrees within an EQF. The nature of abstract or ostensible understand-
ings of policy formation and how they emerge and continuously change
might not have been accounted for within dominant theories used
within policy studies. By taking a routines approach we open up a para-
digmatic debate on how subtle and continuous change is accounted for
within policy studies. In so doing we highlight the ostensive or abstract
understanding of policy formation as an equally and mutually constitu-
tive aspect of policy formation and implementation.
References
Bergen Communiqué. (2005). The European Higher Education Area – Achieving the
Goals Communiqué of the Conference of European Ministers Responsible for Higher
Education, Bergen, 19–20 May 2005. Retrieved from www.bologna-bergen2005.
no/Docs/00-Main_doc/050520_Bergen_Communique.pdf
Berlin Communiqué. (2003). Realising the European Higher Education Area
Communiqué of the Conference of European Ministers Responsible for Higher
Education, Berlin, 19 September 2003. Retrieved from www.ond.vlaanderen.be/
hogeronderwijs/bologna/documents/MDC/Berlin_Communique1.pdf
Brennan, J. (1993). Higher Education Quality: A European Dimension. In
J. Brennan & F. Van Vught (Eds.), Questions of Quality: In Europe and Beyond:
Quality Support Centre, Higher Education Report No. 1. London: Open University.
Byrne, R., & McCutcheon, J. P. (2001). The Irish Legal System (4th ed.). Dublin:
Butterworth.
Capoccia, G., & Kelemen, R. D. (2007). The Study of Critical Junctures: Theory,
Narrative, and Counterfactuals in Historical Institutionalism. World Politics,
59(3), 341–69.
D’Adderio, L. (2008). The Performativity of Routines: Theorising the Influence of
Artefacts and Distributed Agencies on Routines Dynamics. Research Policy,
37(5), 769–89.
D’Adderio, L. (2011). Artifacts at the Centre of Routines: Performing the Material

Turn in Routines Theory. Journal of Institutional Economics, 7(Special Issue 02),
197–230.
Dublin Descriptors. (2004). Shared ‘Dublin’ Descriptors for the Bachelor’s, Master’s
and Doctoral Awards – A Report from a Joint Quality Initiative information group.
Retrieved from http://www.tempus.ac.rs/here/tl_files/Dokumenti/Dublinski%
20deskriptori.pdf
European Commission. (1987). Single European Act. Brussels: European Commission.
Retrieved from http://www.europa.eu/legislation_summaries/institutional_affairs/
treaties/treaties_singleact_en.htm
European Commission. (1991). Memorandum on Higher Education in the European
Community, Com (91) 340, Brussels: European Commission.
European Commission. (1998). Sorbonne Joint Declaration: Joint Declaration on
Harmonization of the Architecture of the European Higher Education System. Signed by
the four ministers in charge of education for France, Germany, Italy and the UK.
Paris, May 25, 1998. Retrieved from http://www.ehea.info/Uploads/Declarations
/SORBONNE_DECLARATION1.pdf
European Commission. (1999). The Bologna Declaration: Joint Declaration of the
European Ministers for Education Convened in Bologna on the 19th June 1999.
Luxembourg: Office for Official Publications of the European Community.
Eurydice. (2005). Focus on the Structure of Higher Education in Europe (2004/05):
National Trends in the Bologna Process. Unité d’Eurydice/European Commission.
Brussels: Directorate-General for Education and Culture.
Farjoun, M. (2010). Beyond Dualism: Stability and Change as a Duality. Academy
of Management Review, 35(2), 202–25.
Feeney, S. (2014). Institutional Quality Review in Higher Education in the Republic of
Ireland and Northern Ireland: A Comparison of Two Approaches. Doctoral thesis:
University of Sheffield.
Feldman, M. S. (2000). Organizational Routines as a Source of Continuous
Change. Organization Science, 11(6), 611–29.
Feldman, M. S. (2003). A Performative Perspective on Stability and Change in
Organizational Routines. Industrial and Corporate Change, 12(4), 727–52.
Feldman, M. S. (2004). Resources in Emerging Structures and Processes of Change.
Organization Science, 15(3), 295–309.
Feldman, M. S., & Pentland, B. T. (2003). Reconceptualizing Organizational
Routines as a Source of Flexibility and Change. Administrative Science Quarterly,
48(1), 94–118.
Feldman, M. S., & Pentland, B. T. (2008). Routine Dynamics. In D. Barry &
H. Hansen (Eds.), The SAGE Handbook of New Approaches in Management and
Organization (pp. 112–31). London: Sage.
Healy, S., & Reynolds, B. (Eds.) (1999). Social Policy in Ireland: Principles, Practice
and Problems. Dublin: Oak Tree Press.
Hogan, J., & Feeney, S. (2012). Crisis and Policy Change – The Role of the Political
Entrepreneur. Risk Hazards and Public Policy, 4(2), 1–16.
Jones, B. D., & Baumgartner, F. R. (2012). From There to Here: Punctuated

Equilibrium to the General Punctuation Thesis to a Theory of Government
Information Processing. Policy Studies Journal, 40(1), 1–20.
Chicago Press.
Nelson, R. R., & Winters, S. G. (1982). An Evolutionary Theory of Economic Change.
Cambridge, MA: Harvard University Press.
Parmigiani, A., & Howard-Grenville, J. (2011). Routines Revisited: Exploring the
Capabilities and Practice Perspectives. The Academy of Management Annals, 5(1),
413–53.
Pentland, B. T. (2004). Towards an Ecology of Inter-organizational Routines:
A Conceptual Framework for the Analysis of Net-Enabled Organizations. Paper pre-
sented at the HICSS.
Pentland, B. T., & Feldman, M. S. (2005). Organizational Routines as a Unit of
Analysis. Industrial and Corporate Change, 14(5), 793–815.
Pierson, P. (2000). Increasing Returns, Path Dependence, and the Study of Politics.
American Political Science Review, 94(2), 251–67.
Prague Communiqué. (2001). Towards the European Higher Education Area.
Communiqué of the meeting of European Ministers in charge of higher educa-
tion in Prague on 19 May 2001.
Reinalda, B., & Kulesza, E. (2005). The Bologna Process: Harmonizing Europe’s Higher
Education. Leverkusen: Barbara Budrich.
Ryle, G. (1949). The Concept of Mind: London: Hutchinson.
Stene, E. O. (1940). An Approach to a Science of Administration. The American
True, J. L., Jones, B. D., & Baumgartner, F. R. (1999). Punctuated-Equilibrium
Theory: Explaining Stability and Change in American Policymaking. In Paul A.
Sabatier (Ed.), Theories of the Policy Process (pp. 97–115). Boulder, CO: Westview
Press.
Tsoukas, H., & Mylonopoulos, N. (2004). Introduction: Knowledge Construction
and Creation in Organizations. British Journal of Management, 15(S1): S1–S8.
Vohora, A., Wright, M., & Lockett, A. (2004). Critical Junctures in the Development
of University High-Tech Spinout Companies. Research Policy, 33(1), 147–75.
Weible, C. M., Sabatier, P. A., Jenkins-Smith, H. C., Nohrstedt, D., Henry, A. D., &
de Leon, P. (2011). A Quarter Century of the Advocacy Coalition Framework:
An Introduction to the Special Issue. Policy Studies Journal, 39(3), 349–60.
Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (2005). Organizing and the Process of
Sensemaking. Organization Science, 16(4), 409–21.
Witte, J. K. (2006). Change of Degrees and Degrees of Change: Comparing Adaptations
of European Higher Education Systems in the Context of the Bologna Process. Doctoral
Thesis: University of Twente. Retrieved from http://www.bologna-bergen2005.
no/Docs/00Main_doc/010519PRAGUE_CUMMUNIQUE.PDF
Zgaga, P. (2003). Bologna Process between Prague and Berlin: Report to the Ministers of
Education of the Signatory Countries, Berlin, September 2003. Brussels: European
Commission.
Zollo, M., Reuer, J. J., & Singh, H. (2002). Interorganizational Routines and
Performance in Strategic Alliances. Organization Science, 13(6), 701–13.
11
The Role of Ideas in Evaluating
and Addressing Hydraulic
Fracturing Regulations1
Christopher M. Weible, Tanya Heikkila, and Jonathan J. Pierce
Introduction
Many policy scholars have promoted and studied the influence of

paradigms and ideas on politics and policy (Béland, 2009, 2010; Blyth,
2001; Campbell, 1998; Goldstein & Keohane, 1993; Hall, 1993; Jacobs,
2009; Parsons, 2002). The study of ideas is based on the observation
that most political debates involve arguments about which actions to
take and how these actions correspond to outcomes. Set within a domi-
nant paradigm or competing paradigms, such arguments are ideational
as participants depend on reasoning, utilize concepts, apply their values,
and examine evidence about alternative policy options (John, 2012).
In this chapter, we examine the effect of different types of ideas on
the evaluation of current regulatory policy as well as the perception of
future policy proposals.
Two of the limitations in the ideational literature are (1) the variation
in definitions of ideas (John, 2012) and (2) the relationship between
ideas and policy (Jacobs, 2009). This research attempts to overcome
these two limitations. First, it defines ideas by linking them to beliefs
and differentiates among categories of ideas by considering the hier-
archical nature of beliefs (Peffley & Hurwitz, 1985; Sabatier & Jenkins-
Smith, 1993). Second, it utilizes empirical methods to investigate how
ideas relate to policy evaluations and preferences of policy actors – or
those individuals inside and outside of government involved in policy
processes on a particular issue.
The context for this study is oil and gas development that utilizes
hydraulic fracturing in Colorado, United States. Hydraulic fractur-
ing (‘fracking’) is a technique for extracting oil and gas primarily from
shale formations. Combined with horizontal drilling techniques, this
217
technology has allowed for high volume drilling and the precipitous
expansion of oil and gas development over the past five years (US
Energy Information Administration, 2014a). Much of this expansion
has been in close proximity to population centers (Rahm, 2011). As the
visibility of oil and gas development has expanded, so too have debates
over the environmental, public health, and economic impacts of this
industrial activity (Brazilian et al., 2014). It is within this context that
we explore how the ideas held by policy actors relate to their evaluation
of and preferences for policies that have been devised recently to address
these debates. Specifically, we focus on two new regulations for hydrau-
lic fracturing in Colorado, US: (1) the disclosure of the chemicals used
in hydraulic fracturing fluid and (2) the setback distance from a well-
head to occupied buildings or natural features. This chapter reports on
recently collected data about these two new rules in Colorado based on
a 2013 survey of policy actors. The findings show that those who have
ideas in support of hydraulic fracturing and who do not see problems
with the industry tend to rate positively the new rules but are against
future regulations.
Literature review
As studies of the policy process have increasingly emphasized ideas

(Blyth, 1997), policy scholars have attempted to explain the relationship
between ideas and policy change or stasis (Béland & Cox, 2010). To con-
duct such research, scholars have to bound and define ideas. This has
not been a uniform effort (Cairney, 2012; Schmidt, 2008). Scholars stud-
ying ideas have defined them as strategies (Blyth, 2002; Jabko, 2006),
narratives (Roe, 1994; Shanahan, Jones, McBeth, & Lane, 2013), social
constructions (Schneider & Ingram, 1993), frames (Stone, 1989), memo-
ries (Rothstein, 2005), and traditions (Katzenstein, 1996), or switches for
interests (Goldstein & Keohane, 1993).
Given the diverse definitions and uses of ideas in studying policy
and politics, this chapter narrows the definition by equating ideas with
different types of beliefs, which is commonly done in the ideational
literature (Béland, 2010; Cairney, 2012). In doing so, this chapter
adopts a hierarchical belief system model as found primarily in the
Advocacy Coalition Framework (ACF) but also related to other litera-
tures (Campbell, 1998; Hall, 1993; Peffley & Hurwitz, 1985; Sabatier &
Jenkins-Smith, 1993). The broadest scope and generalization of beliefs
are termed deep core beliefs (Sabatier & Jenkins-Smith, 1993). Such deep
core beliefs are the base for more applied beliefs and act as underlying
Christopher M. Weible, Tanya Heikkila, and Jonathan J. Pierce 219
assumptions for understanding the world (Campbell, 2004). The sec-

ond level is less broad in scope and abstractness and concerns policy
core beliefs (Sabatier, 1988). These beliefs are structured around how
problems are identified and defined, and alternative solutions to such
problems (Schmidt, 2008).2 Whereas the belief system model that is
used in this chapter to depict ideas operates within individuals, belief
systems up-scaled to a community or political system level equate more
to policy paradigms or shared views of a collective (Hall, 1993). Thus,
while this chapter focuses on the role of ideas in the context of oil and
gas development in Colorado, it can also be interpreted as a narrow
operational reflection of one or more broader paradigms enfolding the
political context.3
One way to conceptualize the deep core beliefs is through Cultural
Theory (Douglas, 1966, 1970, 1990; Wildavsky, 1987). In Cultural
Theory, deep core beliefs can be understood by two principal dimen-
sions: the ‘group’ dimension and the ‘grid’ dimension. The group dimen-
sion is about how individuals understand themselves to be incorporated
into and defined by bounded collectives. The group dimension spans a
continuum from weak to strong associations with collectives. The grid
dimension reflects how people’s interactions are determined by exter-
nally imposed prescriptions, such as norms or laws. The grid dimension
spans a continuum from weak to strong perceptions of the power of pre-
scriptive influence of rules on their decisions. Individuals may possess
high or low perceptions of themselves on the group and grid dimen-
sions. This creates four distinct worldviews: egalitarianism (weak grid,
strong group), hierarchism (strong grid, strong group), individualism
(weak grid, weak group), and fatalism (strong grid, weak group) (Rayner,
1992; Thompson, Ellis, & Wildavsky, 1990; Wildavsky & Dake, 1990).
For the purposes of this study, we focus upon three worldviews: hier-
archism, egalitarianism, and individualism. Those who possess a hier-
archical worldview will have strong group and binding rule-oriented
identities, and believe that individuals will have defined roles in society.
They will place great emphasis upon procedures, authority, stability, and
social order. In contrast, individualists do not have a strong group affin-
ity or feel bound to structural rules. Individualists dislike the constraints
of authority and procedure, instead preferring a lack of formal rules
and structure. Those who have an egalitarian perspective are a mixture
of the two. They will demonstrate strong group connections, but will
prefer minimal external rules. They prefer a world of equality across the
group, rather than the structured authority system of the hierarchical
worldview.4
As defined through Cultural Theory, deep core beliefs are often too
general to offer much insight into how people relate to government;
instead, scholars have sought to understand perceptions and behav-
ior from more specific policy-related beliefs. The ACF, for instance,
employs policy beliefs as a key variable in explaining how people coa-
lesce in trying to influence policy (Sabatier & Jenkins-Smith, 1993).
Policy core beliefs exist in a hierarchical structure subordinate to
more generalized worldviews (George, 1969; Peffley & Hurwitz, 1985;
Sabatier & Jenkins-Smith, 1993). Policy core beliefs relate to the par-
ticular, relevant issue, such as hydraulic fracturing and oil and gas
development in Colorado. There are two typical ways that beliefs
relate to the particular issue. The first is policy preferences, which are
a person’s general normative position toward an issue. The second
involves a person’s understanding of the nature of the problems in
relation to an issue.
In considering the relative importance of deep core and policy core
beliefs, the expectation from the Cultural Theory literature is that deep
core beliefs will show the strongest association with an individual’s
evaluation of policy outcomes. From the ACF perspective, policy core
beliefs will show the strongest association with an individual’s evaluation
of policy outcomes.5 Therefore, in relation to our specific research ques-
tion we would expect that different belief types (pro- and anti-hydraulic
fracturing) operating at different levels of scope and abstractness identi-
fied as deep core or policy core beliefs may both affect evaluations of
recent regulations on hydraulic fracturing. Whether deep core or policy
core beliefs are more strongly associated with these evaluations, how-
ever, is open to debate in the literature.
This chapter also seeks to understand changes in the positions of
policy actors regarding their preferences for more or less government
regulation over hydraulic fracturing. A growing body of literature,
particularly in the environmental field, has begun to explore how poli-
cies are adapted and modified over time (Brunner, Steelman, Coe-Juell,
Cromley, & Edwards, 2005; Folke, Hahn, Olsson, & Norberg, 2005;
Gunderson, 2001; Gunderson & Light, 2006). Such processes of adapta-
tion and modification suggest an ongoing, cyclical interaction between
people and their interpretation of the effects of policies over extended
periods of time. One of the mechanisms enabling this adaptation and
modification is learning from new information and recent experiences
with a corresponding update to their beliefs (Heikkila & Gerlak, 2013).
Thus, we expect that people’s evaluation of government regulations has
the potential to change how they support or oppose future government

regulations.
Other parts of the literature, however, would predict that belief
systems, especially deep core and policy core beliefs, are resistant
to change (Sabatier & Jenkins-Smith, 1993; Thompson et al., 1990).
Therefore, from this literature we might expect that individual pref-
erences for future government regulations in the context of hydrau-
lic fracturing are likely to have stronger associations with deep core
or policy core beliefs than with an individual’s evaluation of current
regulations.
Case study background
Hydraulic fracturing, also called fracing, hydrofracking, and fracking,

has facilitated rapid development of shale gas and shale oil across the
US, including Colorado. Hydraulic fracturing is the technique of inject-
ing water, sand, and chemical additives thousands of feet underground
into shale rock or other porous formations. This mixture of compo-
nents makes up the hydraulic fracturing fluid that fractures the shale
rock releasing either oil or natural gas that comes to the surface via the
well. Colorado (among other states) has a vast supply of shale gas and
oil trapped in shale rock that may now be extracted because of devel-
opments in hydraulic fracturing and horizontal drilling technology
(US Energy Information Administration, 2014b).
A number of political debates have emerged around the practice of
hydraulic fracturing across the US, including Colorado (Davis, 2012;
Schafft, Glenna, Green, & Borlu, 2014; Warner & Shapiro, 2013). This
political controversy is being driven by the uncertainty of the impacts
of hydraulic fracturing (Brazilian et al., 2014) including the benefits
and costs to the economy, employment, energy independence, the
environment, and public health (de Melo-Martin, Hays & Finkel, 2014;
Entrekin, Evans-White, Johnson, & Hagenbuch, 2011; Jackson, Rainey
Pearson, Osborn, Warner, & Vengosh, 2011; Lustgarten, 2009; Maykuth,
2011). While the oil and gas industry as well as some government sup-
porters promote the benefits of hydraulic fracturing, those who oppose
the practice, including environmental and local organized citizen
groups along with their government supporters, focus on the burdens
to the environment and public health (Heikkila et al., 2014). Some of
the most visible debates recently have focused on the potential risks to
public health and the environment associated with the chemicals used
in hydraulic fracturing fluids (Mooney, 2011), as well as the public nui-

sance posed by wells placed near neighborhoods and other public areas
(Jaffe, 2012).
Colorado has been at the forefront of states attempting to address
such debates through state-level regulations. In the summer of 2011,
the Colorado Oil and Gas Conservation Commission (COGCC) (the
state agency that regulates oil and gas) began informal meetings with
representatives of the oil and gas industry and environmental groups
about a rule that would require the disclosure of the chemicals utilized
in hydraulic fracturing fluids also termed the disclosure rule. The discus-
sions developed into formal meetings in the fall of 2011 with various
interest groups being represented. Governor Hickenlooper played a key
role in initiating these regulations (Heikkila et al., 2014) and argued that
industry needed to be more transparent about the process in order to
reassure citizens that fracking fluids are not posing serious health risks
(Jaffe, 2011a). The oil and gas industry and the COGCC also recognized
that greater transparency about hydraulic fracturing is needed to build
public trust (Jaffe, 2011b; Williams, 2011).
On December 5, the COGCC held a formal hearing to consider the pro-
posed rule changes. This public hearing included public comments and
party presentations by interest groups. On December 13, the COGCC
deliberated and eventually adopted an agreement regarding the disclo-
sure of the chemicals used in hydraulic fracturing. According to the new
rule (2 CCR 404-1 Rule 205A) companies are required to disclose the
total volume of water used, chemicals in the additives, and chemical
concentrations, on the public website FracFocus.org. The disclosure rule
permits companies to claim that chemicals in the additives, concentra-
tions of chemicals, or both are trade secrets, though the chemical family
(a classification of chemicals with similar properties) must be posted on
FracFocus.org. Public health professionals and regulators can request
that companies disclose trade secret ingredients when the information
is necessary to treat exposed individuals or respond to spills, but the
ingredients must remain confidential from the public. The public dis-
closure regulations in Colorado were supported and championed as a
victory by both industry and environmental groups (Jaffe, 2011c; New
Fracking Rules, 2011). For industry, they maintained trade secret pro-
tections, but environmental groups gained mandatory disclosure on a
public website. The Colorado disclosure rule was even touted by the US
Congress as a model rule (Murrill and Vann, 2012).
Only months after the disclosure rule was adopted in December 2011,
the COGCC began a new review process concerning the setback distance
of existing well locations from occupied buildings. According to the rules

at the time, well pads were to be placed at least 350 feet from occupied
buildings in high density areas and 150 feet in low density areas. Also,
from schools, hospitals, and other such buildings the distance was to
be 1,000 feet. The COGCC review completed in March 2012 identified
over 5,000 wells had been established in Colorado since 2009 with about
12 percent within the distance of 500 feet and about 26 percent were
within 1,000 feet. At the same time the COGCC began a stakeholder
meeting process similar to the disclosure meeting process where com-
plaints and statements were heard by individuals representing various
interests. In September the COGCC proposed a new uniform distance of
350 feet for all setbacks and various mitigating actions. In the fall they
began a series of prehearing meetings with the formal hearings about
the setbacks issue being held in December 2012 and January 2013. The
COGCC heard presentations from four groups of stakeholders: commu-
nity and environmental organizations, local government, trade associa-
tions such as the Colorado Cattleman’s Association, and the regulated
community. Various arguments and statements were made about the
setback distance and mitigating actions at well sites.
Finally, in February 2013 the COGCC voted on increasing setback
distances to 500 feet for all occupied buildings and passed various miti-
gating actions such as enhanced drilling practices that must be taken
at well sites to limit the public nuisance (COGCC, 2013: CCR 404-1).
The new measures were opposed by environmental organizations who
wanted the minimum setback distance to be 1,000 feet. ‘The way it
stands homeowners have no certainty a well won’t be drilled close to
their houses’, stated Matt Sura representing Western Slope conserva-
tion groups (Jaffe, 2013). The rule was also derided by industry rep-
resentative Tisha Schuller: ‘We do not believe this new rule properly
acknowledges either the complexities or the impacts to a diverse array
of citizen stakeholders’ (Schuller, 2013). In other words, both environ-
mentalists and the industry indicated that the new setback rule was not
acceptable.
Methods
Given our research goals to examine the effect of different types of beliefs
on the evaluation of current regulatory policy as well as the perception
of future policy proposals, our empirical analysis assesses the following:
(1) whether different beliefs identified as deep core or policy core beliefs
affect evaluations of recent regulations for hydraulic fracturing, and
(2) whether variations in evaluations of recent regulations, or variations

in deep core or policy core beliefs affect changes in preferences of future
government regulations for hydraulic fracturing.
Our empirical analysis uses data from a survey of policy actors in
Colorado involved in hydraulic fracturing. To identify policy actors
either directly or indirectly involved in hydraulic fracturing regulation
in Colorado, we first collected names from lists of individual stakehold-
ers who attended the COGCC meetings on disclosure in 2011 and on
setbacks in 2012 and 2013. Added to this list were individuals or organi-
zations that made statements or were represented during the public
hearings in 2011 and during December 2012 and January 2013. In addi-
tion, we collected names of organizations and individuals who attended
public discussions, rallies, local government hearings, industry associa-
tion meetings, and other events where hydraulic fracturing was being
debated in Colorado from June 2012 to January 2013. We also con-
ducted an Internet search for documents commenting on the disclosure
or setback rules, recording the author and publisher. Throughout this
process we conducted 14 interviews in Colorado with various stakehold-
ers asking who we should include in our survey sample. From these dif-
ferent sources we developed a list of 398 stakeholders representing over
130 individual organizations operating in Colorado. This list includes
individuals representing the federal government, regional government,
state government, local government, oil and gas service providers and
operators, industry and professional associations, environmental and
conservation organizations, real estate developers and homebuilders,
agriculture organizations, organized citizen groups, academics and con-
sultants, news media, and other organizations.
After the sample of direct and indirect stakeholders was completed we
launched an online survey in April 2013. The survey was sent to all 398
individuals and had 142 responses, with a response rate of 35.68 percent.
From the survey we developed our measures of our first set of depend-
ent variables – which are evaluations or assessments of the COGCC
2011 disclosure rule and the 2013 setback rules. For these variables,
a series of evaluation questions were asked of respondents based on
the stated goals of the regulations by the COGCC presentation dur-
ing the public hearing. The operational measures, descriptive statistics,
and factor loadings from the series of questions on the evaluation of
the regulations are shown in Table 11A.1. Each of the two sets of ques-
tions for the evaluations of the disclosure and setback rules were then
combined into distinct scales with the factor loadings and Cronbach’s
alphas presented.
Our second set of dependent variables is the extent to which policy

actors changed their position on government regulation in relation to
(1) disclosure, (2) setbacks, and (3) government regulations in general.
These variables are drawn from our survey question that asked respond-
ents to rank the extent to which they changed their position on a num-
ber of issues. This question included an option for the issues of disclosure
and setbacks. For the variable representing government regulations in
general, we computed a scale that combined all of the reported issues, as
shown in Table 11A.2.
The explanatory variables used to assess stakeholder evaluations
of government regulations and changes in positions on regulations
include measures of deep core beliefs and policy core beliefs. Table 11A.3
provides the operational measures and means for these independ-
ent variables, including the creation of two scaled items for the deep
core beliefs. We found that the deep core beliefs’ questions for iden-
tifying hierarchists and individualists load into the same scale. Thus,
the respective two sets of questions (for hierarchists and individualists)
were combined into a single scale called ‘hierarchists-to-individualists’
scale (with factor loadings needing reversal to create the scale). In such
a scale, high numbers indicate strong hierarchists views (support of
rules and structures) whereas low numbers indicate strong individualists
views (support for autonomy). While these two concepts are theoreti-
cally distinct, they do lie on opposite corners of Cultural Theory and,
thus, there is some validity that the two would be negatively correlated.
In all scales, the alpha scores exceed 0.9 suggesting internal consistency
of the items comprising the scales. The first set of dependent variables –
assessments of the two Colorado regulations – also serve as independ-
ent variables in analyzing the factors that influence the second set of
dependent variables.
Although the theoretical thrust of this chapter is to analyze policy
actors’ beliefs in relation to policy evaluations, we also control for sev-
eral individual attributes. These controls include gender, level of educa-
tion, and years of involvement in the issue. These control variables were
collected as part of our survey instrument.
Results
We analyzed how ideas relate to our two sets of dependent variables

using OLS regression models. In examining the results from the OLS
models for policy actors’ evaluations of the extent to which the 2011
disclosure rule and the 2013 setback rules resolved policy problems,
Table 11.1 Explaining policy actor evaluations of problem resolution by the

disclosure rule and setback rule
‘Disclosure ‘Setback
rule resolved rules resolved
problems’ scale problems’ scale
Deep Core Beliefs

‘Hierarchist-to-individualist’ scale 0.01 0.05
‘Egalitarian’ scale −0.05 0.06
Policy Core Beliefs
Pro-hydraulic fracturing 0.43*** 0.47***
Severity of disclosure problem −0.39*** –
Severity of setback problem – −0.34***
Control Variables
Female 0.04 0.01
Level of education −0.05 −0.10
Years involved −0.03 −0.02
F-Statistic/Significance 23.0/0.000 12.2/0.000

Adjusted R2 0.58 0.41
N 113 112
Note: p d 0.01***, p < 0.05**, p < 0.10*. Coefficients are standardized.
Table 11.1 shows that ideas at the level of policy beliefs explain a signifi-
cant amount of variance in the evaluations of the two rules. Both the
variables representing a policy actor’s position on hydraulic fracturing
and the variables representing problem severity that make up an actor’s
policy core belief are significant, whereas the variables representing deep
core beliefs are not.
Specifically, we find that individuals who report being more ‘pro’-
hydraulic fracturing are more likely to report that both the disclosure
rule and the setback rules resolved the problems associated with each
set of rules. At the same time, policy actors who report being more
‘anti’-hydraulic fracturing are more likely to report that disclosure of
chemicals and setback distances are severe problems and are less likely
to perceive that the rules resolved these problems. However, our results
do not indicate that deep core beliefs are significantly related to percep-
tions that the rulemaking processes resolved the problems at hand.
Table 11.2 presents findings on the factors associated with changes
in policy actors’ positions on government regulation related to the
regulatory rules of chemical disclosure (model 1) and setback distances
Table 11.2 Explaining variance in changes in positions on government regula-

tions related to disclosure and setbacks
Changed position for support

of government regulations on
disclosure of setbacks of wells

chemicals from occupied
in hydraulic buildings or
fracturing fluids natural features
Deep Core Beliefs

‘Hierarchist-to-individualist’ scale 0.00 0.28**
‘Egalitarian’ scale 0.21* 0.15*
Policy Core Beliefs
Pro-hydraulic fracturing −0.28* −0.14
Assessments of Recent Regulations
‘Problems resolved by disclosure rule’ −0.31**
scale
‘Problems resolved by setback rule’ −0.32**
scale
Control Variables
Female −0.35** −0.10
Education 0.03 0.01
Years involved −0.30** −0.22**
Wald Chi2/Prob Chi2 50.3/0.000 60.0/0.000

Pseudo R2 0.38 0.35
N 114 113
Note: p d 0.01***, p < 0.05**, p < 0.10*. Coefficients indicate fully standardized coefficients.
Ordered logit models calculated with robust standard errors.
(model 2). Although the model evaluating chemical disclosure does

not find support for the relationship between the deep core belief
‘hierarchist-to-individualist’ scale and changes in positions on the role
of regulation, the model on setbacks does find a significant relation-
ship. What this shows is that as policy actors move up on the hierarchist
scale of deep core beliefs, they are more likely to change their position
in support of setbacks. Both models indicate that as policy actors rank
themselves higher on the egalitarian scale, they are more likely to report
changes in their position and support more regulations.
The analyses also show that policy core beliefs matter for chang-
ing positions of government regulations on hydraulic fracturing. Both
models show that individuals possessing ideas in support of hydraulic
fracturing (relative to those who are opposed) report being less likely
to change their position on the support of government regulations,
although this finding is only significant in the model on chemical
disclosure.
At the same time, how people evaluated the regulations can also shape
their propensity to change preferences on future regulations. Those pol-
icy actors who are more likely to see that the problems were resolved
by the new regulations are less likely to report changing their position
for support of government regulation on the two policy issues. In other
words, if people believe that there are no more problems to address, they
may be less likely to report that they have become more supportive of
future government regulations.
The control variables in the two models presented in Table 11.2 also
are significantly related to the dependent variables. First, we see that
women are less likely, relative to men, to report becoming more sup-
portive of future government regulations, although the variable is only
significant in the chemical disclosure model. Additionally, individuals
with more years of involvement with the issue of hydraulic fracturing
are significantly less likely to report becoming more supportive of gov-
ernment regulation on both the chemical disclosure and setbacks issues.
The level of education is not significant in the two models.
Table 11.3 presents a single model that uses the combined depend-
ent variable representing policy actors’ changes in support for govern-
ment regulation in general (based on the mean of nine questions, see
Table 11A.3). The findings show that even if respondents perceive rules
on disclosures and setbacks as solving problems, they still do not report
support for more government regulations. The results suggest that posi-
tive evaluations of government rulemaking do not necessarily correlate
with support for future regulations. We also find support for deep core
beliefs and policy core beliefs in explaining some of the variance in
changed positions for support or opposition of government regulations.
Strong hierarchical beliefs are associated with support for future rule-
making as are strong egalitarian beliefs. People in support of hydraulic
fracturing are not supportive of regulation of the industry. Among the
control variables, females and years of involvement are negatively asso-
ciated with changes in positions on government regulations.
Conclusions
Paradigms are comprised of ideas and this chapter explores the role of ideas
in policy processes surrounding one of the most contentious environmental
Table 11.3 Explaining changes in support for government regulations on

hydraulic fracturing issues
‘Changed position for support of

government regulations’ scale
Deep Core Beliefs

‘Hierarchist-to-individualist’ scale 0.20**
‘Egalitarian’ scale 0.18*
Policy Core Beliefs
Pro-hydraulic fracturing −0.24**
Assessments of Recent Regulations
‘Problems resolved by disclosure rule’ scale −0.18*
‘Problems resolved by setback rule’ scale −0.23**
Control Variables
Female −0.16**
Education 0.01
Years involved −0.15**
Wald Chi2/Prob Chi2 31.5/0.000

Pseudo R2 0.69
N 111
Note: p d 0.01***, p < 0.05**, p < 0.10*. Coefficients are standardized.
issues in the 21st century – oil and gas development inclusive of hydraulic
fracturing. This rapidly growing industry has spurred intense public debates
and various government responses. In Colorado, the government adopted
two new rules governing hydraulic fracturing processes in the state. The
first involved disclosure of chemicals used in hydraulic fracturing fluids.
The second involved procedures for reducing public nuisances for hydrau-
lic fracturing operations by using setback distances. Using cross-sectional,
quantitative data, this chapter sought to understand the relationship of
different ideas, as defined as beliefs, with current regulatory policy, as well
as whether these ideas including policy evaluations are associated with
variation in the perception of future policy proposals.
The theoretical approach of this chapter was to understand ideas oper-
ating as two types of beliefs: deep core beliefs and policy core beliefs.
In conceptualizing and measuring deep core beliefs, this chapter draws
from Cultural Theory. The results found that deep core beliefs are not
related to the evaluation that regulations solved problems. However,
people with strong hierarchical beliefs and egalitarian beliefs are more
likely to become more supportive of future regulation. The findings sug-
gest that people use their deep core beliefs for assessing future policy
preferences, but less so in evaluating current contexts.
Policy beliefs were measured as (1) support or opposition to hydraulic

fracturing and (2) the perceived severity of problems related to disclo-
sure of chemicals and setback distances. The results found that policy
beliefs in favor of hydraulic fracturing are significantly related to evalu-
ating positively the new rules, but against future government regula-
tions of the industry. One interpretation is that people possessing ideas
in favor of hydraulic fracturing did not see the need for future regula-
tions because they perceived only minor problems related to these issues
of hydraulic fracturing, and viewed these two rules as solving those
problems. A different interpretation is that people against hydraulic
fracturing continued to perceive such problems, viewed the new regula-
tions as not effective in solving those problems, and reported changes
in support for more regulations. Consistent with both perceptions, the
influence of policy core beliefs supports the literature that people inter-
pret events based on issue specific aspects of their belief system rather
than upon fundamental normative values as depicted by their deep core
beliefs (Sabatier & Jenkins-Smith, 1993).
We also examined how evaluations of the new rules influence changes
in support or opposition of future preferences of regulations of the indus-
try. This argument is inspired from the assumption that people can learn
and update their ideas over time. The findings provide some support
for this argument. Evaluations of the two rules do significantly relate to
changes in support or opposition for future preferences toward regula-
tions. Respondents, however, are less inclined to support future regu-
lations if they evaluate the two recent rules positively. This finding is
counterintuitive – one would expect that positive evaluations of govern-
ment regulations would direct people to be more supportive of future
government regulations. One interpretation of the counterintuitive find-
ing is that positive evaluations of recent regulations are associated with
no change in support or opposition of government regulations, whereas
negative evaluations of recent regulations are associated with more sup-
port. In other words, failure of a government regulation would lead a
person to want more regulations because problems remain. A positive
evaluation of recent regulations would lead a person to maintain the sta-
tus quo, which is not necessarily the same as wanting more regulations.
The implication is that positive experiences with government need not
necessarily lead to changes in preferences for more regulation.
Despite these findings, a critical question must be asked: what the-
oretical insights have been gained by taking a paradigmatic and idea-
tional perspective in the analysis of hydraulic fracturing regulations?
This chapter assumes that ideas are the operational manifestations of
one or more paradigms enfolding a political system and then equates

ideas with beliefs, which is one of the established definitions of ideas in
the literature (Béland, 2010; Cairney, 2012). Yet, this chapter could have
been written as an application of the ACF and Cultural Theory, which
raises questions about the benefits of the ideas concept. One advantage
of ideas is its use as an umbrella term that integrates different mental
constructs, from knowledge to norms, and, thus, connects disparate
parts of the literature together. However, the corresponding disadvan-
tage is the same: equating radically different mental constructs under
the same term creates situations of miscommunication and generates
false sense of agreements. Indeed, what is gained when different scholars
claim that ‘ideas matter’ and conceive the term differently? To move the
ideas literature forward, there must be a concerted effort to narrow the
conceptual definition of ideas and to develop original theory around it.
Appendix
Table 11A.1 Means and factor loadings for disclosure and setback variables and
scales
Factor
Means loadings
‘Disclosure rule resolved problems’ (from 1 = Strongly disagree to 5 = Strongly agree)

1. What chemical information must be disclosed? 3.3 0.93
2. Where chemical information should be made 3.3 0.93
available?
3. Accessibility of chemical information to the 3.2 0.93
public
4. Protection of trade secrets 3.3 0.77
5. Disclosure of chemical information in a 3.3 0.92
health or other emergency
6. When disclosure of chemical must be made? 3.1 0.94
Averaged scale (Cronbach’s alpha = 0.95) 3.2
‘Setback rule resolved problems’ (from 1 = Strongly disagree to 5 = Strongly agree)

1. Public nuisance impacts 2.8 0.88
2. A patchwork of local regulations on setbacks 2.5 0.71
3. Priorities of surface owners 2.8 0.85
4. Health impacts upon the population living in 2.5 0.92
proximity to well pads
5. Impacts from open pits of wastewater 2.7 0.86
6. Communications between oil and gas 2.9 0.75
operators and nearby communities
Table 11A.2 Measures, means, and factor loadings for changes in support for
government regulations
Factor
Means loadings
‘Changed position for support of government regulations’ (1 = Have become

less supportive of government regulations; 2 = No change; 3 = Have become
more supportive of government regulations)
1. Disclosure of chemicals in hydraulic fracturing 2.4 0.80
fluids
2. Setbacks of wells from occupied buildings or 2.3 0.86
natural features
3. Monitoring of water quality 2.5 0.82
4. Monitoring of air emissions 2.4 0.83
5. Designing and constructing wells 2.3 0.80
6. Disposing or treating produced water 2.5 0.85
7. Constructing well pads 2.3 0.80
8. Mitigating risks from induced seismic activity 2.1 0.72
9. Mitigating risks and nuisances to the general 2.3 0.82
public caused by truck traffic, noise, and light
from well site operations
Table 11A.3 Measures, means, and factor loadings for independent variables
Means Factor loadings
Hierarchists (from 1 = Strong disagree to 4 = Strongly agree)

1. Government should put limits on the choices 2.1 0.74
individuals make so they do not get in the way
of what is good for society
2. The government should do more to advance 2.1 0.80
society’s goals, even if that means limiting
freedom and choices of individuals
3. Sometimes government needs to make laws that 2.7 0.73
keep people from hurting themselves
Individualists (from 1 = Strong disagree to 4 = Strongly agree)

1. It is not the government’s business to try to 2.5 −0.70 (reversed)
protect people from themselves
2. The government should stop telling people how 2.7 −0.77 (reversed)
to live their lives
3. The government interferes far too much in our 2.5 −0.82 (reversed)
everyday lives
‘Hierarchist-to-individualist’ scale (Cronbach’s 2.9
alpha = 0.85)
Table 11A.3 (continued)
Means Factor loadings
Egalitarians (from 1 = Strong disagree to 4 = Strongly agree)

1. We need to dramatically reduce inequalities 2.7 0.95
between the rich and the poor, as well as
between men and women
2. Our society would be better off if the 2.5 0.95
distribution of wealth was more equal
‘Egalitarian’ scale (Cronbach’s alpha = 0.89) 2.6
Policy Core beliefs

1. Please indicate what comes closest to your 3.1
current position in relation to natural gas
development that uses hydraulic fracturing. It
should be . . . (1 = Stopped; 2 = Limited; 3 =
Continue at current rate; 4 = Expanded
moderately; 5 = Expanded extensively).
Problem perceptions (from 1 = Not a problem to 5 = Severe problem)

1. Contamination of ground and surface water 2.6
from chemicals in hydraulic fracturing fluids
2. Nuisance to the general public caused by truck 3.2
traffic, noise, and light from well site operations.
Controls
Female (0 = Male; 1 = Female) .34
Education (from 0 = Not a high school graduate to 4.7
4 = Bachelor’s degree to 6 = PhD or MD)
Years involved (from 1 = 0–1 years to 3 = 5–9 years 3.1
to 5 = 21 or more years)
Notes
1 We are grateful for the individuals in Colorado who volunteered their time
to participate in this study. This research was funded by the Alfred P. Sloan
Foundation, though the research design and results are the authors alone. We
wish to thank Deserai Crow, Sam Gallaher, and Jennifer Kagan for providing
support in collecting and analyzing data for this research study.
2 The third level of the hierarchical belief system within the Advocacy Coalition
Framework is secondary beliefs, which are not studied in this chapter.
3 An interpretation of Hall (1993) would depict paradigms as a system-level
construct, which are comprised of ideas. The ACF begins at the individual
level and their belief components and belief systems. In this respect, Hall
(1993) and the ACF (Sabatier, 1988) build from different levels of analysis.
In this chapter, we are assuming that ideas can be applied at the individual
level by equating them with specific belief components.
4 For a more in-depth description of these worldviews and group grid theory,
see Thompson et al. (1990).
5 We are not stating expectations regarding the direction of the effect of deep
core or policy core beliefs on evaluating rulemaking products in Colorado.
In part this is because the relationship is ambiguous. A person in favor of
hydraulic fracturing, for example, may evaluate the regulations as solving
the problems, a reassuring expression that the status quo is now acceptable.
However, a person in favor of hydraulic fracturing may equally evaluate the
problems as unsolved, as regulations trample upon oil and gas development
and entrepreneurship. Similarly, a person opposed to hydraulic fracturing may
truly support that the regulations solved problems or be so adamant about
their opposition that even the best regulations will not go far enough without
banning the technique. While these examples focus on policy beliefs, similar
logic applies to deep core beliefs and evaluating government regulations.
References
Béland, D. (2010). The Idea of Power and the Role of Ideas. Political Studies Review,
8(2), 145–54.
Béland, D., & Cox, R. H. (2010). Ideas and Politics in Social Science Research. Oxford:
Oxford University.
Blyth, M. (1997). Moving the Political Middle: Redefining the Boundaries of State
Action. The Political Quarterly, 68(3), 217–314.
Blyth, M. (2001). The Transformation of the Swedish Model: Economic Ideas,
Distributional Conflict, and Institutional Change. World Politics, 54(1), 1–26.
Brazilian, M., Brandt, A. R., Billman, L., Heath, G., Logan, J., Mann, M., Melaina,
M., Statwick, P., Arent, D., & Benson, S. M. (2014). Ensuring Benefits from
North American Shale Gas Development: Towards a Research Agenda. Journal
of Unconventional Oil and Gas Resources. Retrieved from http://www.sciencedi-
rect.com/science/article/pii/S2213397614000123
Brunner, R. D., Steelman, T. A., Coe-Juell, L., Cromley, C., & Edwards, C. (2005).
Adaptive Governance: Integrating Science, Policy, and Decision Making. New York:
Columbia University Press.
Cairney, P. (2012). Understanding Public Policy: Theories and Issues. London:
Palgrave Macmillan.
Campbell, J. (1998). Institutional Analysis and the Role of Ideas in Political
Campbell, J. (2004). Institutional Change and Globalization. Princeton, NJ:
Colorado Oil and Gas Conservation Commission. (2013). ‘Statement of Basis,
Specific Statutory Authority, and Purpose New Rules and Amendments to
Current Rules of the Colorado Oil and Gas Conservation Commission, 2 CCR
404-1’. Retrieved from http://cogcc.state.co.us/RR_HF2012/Setbacks/FinalRules/
Final_SetbackRules-StatementOfBasisAndPurpose.pdf
Davis, C. (2012). The Politics of ‘Fracking’: Regulating Natural Gas Drilling

Practices in Colorado and Texas. Review of Policy Research, 29(2), 177–91.
de Melo-Martin, I., Hays, J., & Finkel, M. L. (2014). The Role of Ethics in Shale Gas
Policies. Science of the Total Environment, 470–1, 1114–19.
Douglas, M. (1966). Purity and Danger: An Analysis of the Concepts of Pollution and
Taboo. New York: Routledge and Kegan Paul.
Douglas, M. (1970). Natural Symbols: Explorations in Cosmology. London: Barrie &
Rockliff.
Douglas, M. (1990). Risk and Blame: Essays in Cultural Theory. New York: Routledge.
Entrekin, S., Evans-White, M., Johnson, B., & Hagenbuch, E. (2011). Rapid
Expansion of Natural Gas Development Poses a Threat to Surface Waters.
Frontiers in Ecology of the Environment, 9(9), 503–5.
Folke, C., Hahn, T., Olsson, P., & Norberg, J. (2005). Adaptive Governance of Social-
Ecological Systems. Annual Review of Environment and Resources, 30, 441–73.
George, A. L. (1969). The ‘Operational Code’: A Neglected Approach to the Study
of Political Leaders and Decision-Making. International Studies Quarterly, 13(2),
190–222.
Goldstein, J., & Keohane, R. O. (1993). Ideas and Foreign Policy: An Analytical
Framework. In J. Goldstein & R. O. Keohane (Eds.), Ideas and Foreign Policy: Beliefs,
Institutions and Political Change (pp. 3–30). Ithaca, NY: Cornell University Press.
Gunderson, L., & Light, S. S. (2006). Adaptive Management and Adaptive
Governance in the Everglades Ecosystem. Policy Sciences, 39(4), 323–34.
Gunderson, L. H. (2001). Panarchy: Understanding Transformations in Human and
Natural Systems. Washington, DC: Island Press.
Heikkila, T., & Gerlak, A. (2013). Building a Conceptual Approach to Collective
Learning: Lessons for Public Policy Scholars. Policy Studies Journal, 41(3), 484–512.
Heikkila, T., Pierce, J. J., Gallaher, S., Kagan, J., Crow, D., & Weible, C. M. (2014).
Understanding a Period of Policy Change: The Case of Hydraulic Fracturing
Disclosure Policy in Colorado. Review of Policy Research, 31(2), 65–87.
Jabko, N. (2006). Playing the Market. Ithaca, NY: Cornell University Press.
Jackson, R. B., Rainey Pearson, B., Osborn, S. G., Warner, N. R., & Vengosh, A.
(2011). Research and Policy Recommendations for Hydraulic Fracturing and
Shale-Gas Extraction. Center on Global Change, Duke University, Durham,
NC. Retrieved from http://www.nicholas.duke.edu
Jacobs, A. M. (2009). How Do Ideas Matter?: Mental Models and Attention in
German Pension Politics. Comparative Political Studies, 42(2), 252–79.
Jaffe, M. (2011a, August 3). Colorado Plans Disclosure Rule for Fracking Fluids by
End of Year. The Denver Post. Retrieved from http://www.denverpost.com
Jaffe, M. (2011b, November 1). Colorado Fracking Audit Advises Rule-Tightening.
The Denver Post. Retrieved from http://www.denverpost.com
Jaffe, M. (2011c, December 14). Colorado Approval of Fracking Fluids’ Full
Disclosure Came after Long Negotiations and Nudge from Governor. The
Denver Post. Retrieved from http://www.denverpost.com
Jaffe, M. (2012, December 31). Colorado Proposes Bigger Oil and Gas Drilling
Setbacks from Homes. The Denver Post. Retrieved from http://www.denverpost.
com/breakingnews/ci_22289504/colorado-proposes-bigger-oil-and-gas-
drilling-setbacks
Jaffe, M. (2013, February 11). Colorado Oil and Gas Well Setback Fight Headed to
State Legislature. The Denver Post. Retrieved from http://www.denverpost.com/
breakingnews/ci_22566155/colorado-oil-and-gas-commission-votes-500-foot
John, P. (2012). Analyzing Public Policy (2nd ed.). New York: Routledge.
Katzenstein, P. J. (1996). The Culture of National Security: Norms and Identity in
World Politics. New York: Columbia University Press.
Lustgarten, A. (2009, November 9). Natural Gas Drilling Produces Radioactive Waste-
water. Scientific American. Retrieved from http://www.scientificamerican.com
Maykuth, A. (2011, August 30). Fracking’s Thirst for Water: A Delicate Dance
between Gas Industry and River Commission. Philadelphia Inquirer. Retrieved
from http://www.post-gazette.com/stories/local/marcellusshale/frackings-thirst-
for-water-a-delicate-dance-between-gas-industry-and-river-commission-
312485/
Mooney, C. (2011). The Truth about Fracking. Scientific American, 305(5), 80–5.
Retrieved from http://www.scientificamerican.com
Murrill, B. J., & Vann, A. (2012, June 19). Hydraulic Fracturing: Chemical Disclosure
Requirements. Congressional Research Service. Retrieved from http://www.fas.
org/sgp/crs/misc/R42461.pdf
New Fracking Rules: What They’re Saying. (2011, December 13). Denver Business
Journal. Retrieved from http://www.bizjournals.com/denver
Parsons, C. (2002). Showing Ideas as Causes: The Origins of the European Union.
International Organization, 56(1), 46–84.
Peffley, M. A., & Hurwitz, J. (1985). A Hierarchical Model of Attitude Constraint.
Rahm, D. (2011). Regulating Hydraulic Fracturing in Shale Gas Plays: The Case of
Texas. Energy Policy, 39(5), 2974–81.
Rayner, S. (1992). Cultural Theory and Risk Analysis. In S. Krimsky & D. Golding
(Eds.), Social Theories of Risk (pp. 54–82). Westport, CT: Praeger.
Roe, E. (1994). Narrative Policy Analysis: Theory and Practice. Durham, NC: Duke
University Press.
Rothstein, B. (2005). Social Traps and the Problem of Trust. Cambridge: Cambridge
University Press.
Sabatier, P. A. (1988). An Advocacy Coalition Framework of Policy Change and
the Role of Policy-Oriented Learning Therein. Policy Sciences, 21(2), 129–68.
Sabatier, P. A., & Jenkins-Smith, H. (Eds.) (1993). Policy Change and Learning: An
Schafft, K. A., Glenna, L. L. Green, B., & Borlu, Y. (2014). Local Impacts of
Unconventional Gas Development within Pennsylvania’s Marcellus Shale
Region: Gauging Boomtown Development through the Perspectives of
Educational Administrators. Society and Natural Resources, 27(4), 389–404.
Schneider, A., & Ingram. H. (1993). Social Construction of Target Populations:
Implications for Politics and Policy. American Political Science Review, 87(2),
334–47.
Schuller, T. (2013). ‘Tisha Schuller’s, President and CEO of the Colorado Oil and
Gas Association, Statement Regarding the Colorado Oil and Gas Conservation
Commission’s Setback Ruling’. Press Release from Colorado Oil and Gas
Association. Retrieved from http://www.coga.org/COGAContent/COGA
StatementFeb112013.pdf
Shanahan, E. A., Jones, M. D., McBeth, M. K., & Lane, R. R. (2013). An Angel on
the Wind: How Heroic Policy Narratives Shape Policy Realities. Policy Studies
Journal, 41(3), 453–83.
Stone, D. A. (1989). Causal Stories and the Formation of Policy Agendas. Political
Thompson, M., Ellis, R., & Wildavsky, A. (1990). Cultural Theory. Boulder, CO:
Westview Press.
US Energy Information Administration. (2014a). Annual Energy Outlook 2014.
DOE/EIA0383(2014). Washington, DC: Department of Energy.
US Energy Information Administration. (2014b). Natural Gas, Shale Gas in
Colorado. Retrieved from http://www.eia.gov/dnav/ng/ng_enr_shalegas_dcu_
sco_a.htm
Warner, B., & Shapiro, J. (2013). Fractured, Fragmented Federalism: A Study in
Fracking Regulatory Policy. Publius: The Journal of Federalism, 43(3), 464–96.
Wildavsky, A. (1987). Choosing Preferences by Constructing Institutions: A Cultural
Theory of Preference Formation. American Political Science Review, 81(1), 4–21.
Wildavsky, A., & Dake, K. (1990). Theories of Risk Perception: Who Fears What
and Why? Daedalus, 119(4), 41–60.
Williams, D. O. (2011, September 21). Frack Frenzy Continues Nationwide as
Colorado Officials Eye New Disclosure Rule by End of Year. The Colorado
Independent. Retrieved from http://coloradoindependent
12
Communications Frameworks
and the Supply of Information in
Policy Subsystems
Samuel Workman and JoBeth S. Shafran
Introduction
In this chapter, we develop a new perspective on policy dynamics based

on the generation and communication of information between the
bureaucracy, Congress, and other relevant actors at the federal level
in the United States. Our communications-based model of signaling
emphasizes the importance of problem definition and the strategic
manipulation of competition in the provision of information. In doing
so, we draw a distinction between information supply in incentive-based
systems versus competition-based systems. This distinction in the way
information is supplied represents the emergence of a new paradigm for
the study of policy dynamics with theoretical and practical importance.
Our chapter proceeds in two sections. First, we critique existing litera-
ture on signaling and the provision of information in incentive-based
systems. We relax the assumption that information in the policy process
is privately held, and examine how communications frameworks yield
considerable leverage in understanding the development of problem
definitions, agenda setting, and policy dynamics. We pay particular
attention to the strategic manipulation of competition among the gen-
erators of information in the policy process.
In the second section, we apply the lessons learned from examining
communications frameworks and information supply to policy subsys-
tems. Within these subsystems, information is generated and filtered in
the congressional hearing process. For a given subsystem, bureaucrats
and vested interest groups produce information in hopes of providing it
to Congress and steering subsequent policy debates. Congressional com-
mittees must filter the barrage of information, attending to some sources
of information at the expense of others. We discuss the relationship
239
between committee filtering and refining of the information supply.

In the conclusion, we examine how the ideas drawn from a commu-
nications perspective on information might be useful more broadly in
informing further research in the policy processes.
Paradigms are apt characterizations at the level of theoretical devel-
opment, and as a set of organizing ideas within policy communities.
Concerning the former, our communications perspective is paradigmati-
cally rooted in bounded rationality (Simon, 1979, 1983), especially those
developments emphasizing the importance of attention allocation for
individual and organizational decision making (Jones, 1994, 2001). Our
approach depends heavily on a depiction of organizations, and interac-
tions among them, as a system of cognition. We view governing institu-
tions as engaged in problem solving. This information processing approach
leads us to agree with Baumgartner (2014) in that policy dynamics,
founded in human and organizational cognition, should be the result
of a single process across issues, institutions, and magnitudes of change.
Ideas are fundamentally important to this process.
From the notion that governing is problem solving (Howlett &
Ramesh, 2003), we identify a key distinction in organizational infor-
mation processing—the parsing of problems into the definition of the
problem and the generation of solutions (Newell & Simon, 1972). We
see problem definition as fundamental to paradigm shifts associated
with the organizing ideas of policy communities. Problem definition
and redefinition form, for us, a necessary element of the critical junc-
tures leading to paradigmatic shifts in policy communities (Donnelly &
Hogan, 2012). We stress the communication of information about prob-
lem spaces—the parameters of choice—in explaining policy dynamics
and how these channels are ‘tuned’ in the course of problem solving.
Information: Politics and public policy
The pivotal role of information in politics and public policy is borne

out in the theoretical development of our understanding of the major
political institutions as well as policy dynamics. There are, however,
key junctures at which the two trajectories of thought depart, leading
to drastically different conceptions of the influence of information in
policy making. These points of departure begin with a conceptual dis-
tinction between the types of information prominent in the central
theories, but extend to how and why information is generated, how
political institutions process information, and the relationship between
information and public policy.
Samuel Workman and JoBeth S. Shafran 241
In the study of political institutions information is invariably asso-

ciated with political or policy preferences or attitudes. Two features of
information as preferences are relevant for understanding political insti-
tutions. First, preferences, if known or measurable, determine a range
of important features of political institutions from their organization
(Krehbiel, 1992; Lewis, 2003; Shepsle, 1978) to legislative and adminis-
trative outcomes (McCubbins, 1985; Shepsle & Weingast, 1987). Second,
uncertainty surrounding preferences (or more often, the bias associated
with them) undergird much of our understanding of how institutions
at the federal level relate to one another (Epstein & O’Halloran, 1999;
Huber & Shipan, 2002; Patty, 2009).
Given that preferences are often unknown or biased, this work sees
the central problem for individuals or institutions in processing infor-
mation as inducing agents, often adversaries, to reveal preferences, or if
known, reveal the bias connected to preferences. In this framework, the
withholding of information yields advantages in the political and policy
process as information asymmetries give leverage to actors holding the
information.
For scholars of public policy, a vastly different depiction of informa-
tion emerges, beginning with a more varied conceptualization of infor-
mation. For these scholars, information has meant technical expertise
and science (Weible, 2008; Weible & Sabatier, 2009), attention shifts to
new attributes of issues or new issues altogether (Jones & Baumgartner,
2005), changing frames for understanding an issue (Baumgartner &
Jones, 1993), organizing ideas in the imposition of policy regimes (May,
Jochim, & Sapotichne, 2011), a destabilizing force in existing govern-
ing arrangements (May, Sapotichne, & Workman, 2009a, 2009b; May,
Workman, & Jones, 2008), and regulatory agendas as supplying infor-
mation in the governing system (Workman, 2015).
What is common among these conceptualizations is an emphasis on
pluralistic governing arrangements and information generation in a
competitive environment. In these depictions, information is not held
privately, but provided freely in the competition for ideas as coalitions,
institutions, and other actors attempt to steer policy debates and ulti-
mately shape policy change.
Building on Baumgartner and Jones’ (2014) two types of search in
the policy process, we argue that these different bases for exploring
the influence of information in politics and policy making lead to key
distinctions in how information is generated and search occurs. In the
coming sections, we illuminate two different tracks of information
search (Baumgartner & Jones, 2014, pp. 79–80). We then move on to
develop our new perspective on policy dynamics based in communi-

cations frameworks and extend this theory to information search and
processing in policy subsystems. For us, the importance of information,
whether it is preferences or beliefs, science or technical expertise, or
shifts in issue attention or frames, demands a theory of communication
in policy making.
Central to our development of a communications framework is a
model of signaling rooted in communications perspectives rather than
political economy. This represents both a departure from the existing
paradigm for thinking about the operation of information in the policy
process as well as the linchpin for a new paradigm. The confluence of
the organization as a system of cognition, the nature of the information
supply, and signaling based in communication theory forms the base of
this emergent paradigm.
Before we begin to develop our framework, it is necessary to define
how we think about information, and in particular, how it is related to
the types of information generated in the political and policy process.
It is clear that information is generated in both incentive-based and
competition-based governing arrangements, but this distinction relates
strongly to features of policy problems, and to the types of search in
which actors engage.
Attention and search
A key insight of the literature on decision making is the separation of

choice into its component parts of which there is a ‘front-end’ and
‘back-end.’ Newell and Simon’s (1972) path-breaking work on problem
solving held that problem solving could be broken into the construction
of a problem space and a solution space. In the problem space, the deci-
sion maker must ascertain what dimensions of choice are relevant to the
problem. This allows for the generation of solutions.
The distinction between the problem space and the solution space
is important in understanding the generation of information and pro-
cesses of search. From this discussion, information may pertain to the
problem space—the parameters of choice from which solutions are
generated or to the solutions themselves. The processes for generating
information will also vary depending on whether a decision maker,
or organization, is constructing a problem space or choosing among
solutions.
Herein lies a key distinction in how we conceptualize information in
the policy process. Information pertaining to solutions, or alternatives,
bears directly on the likely effects, consequences, costs, or benefits of
a particular solution. It also engenders consideration of the congruity

between ideological and policy preferences and their likely outcomes.
For example, the debate in the US surrounding cap-and-trade emissions
standards and air quality regulation under the Clean Air Act1 are solu-
tions to a particular problem definition involving aspects of energy,
environmental, and economic policy problems.
In contrast, information pertaining to the problem space relates to
the dimensions of choice from which solutions are generated. A classic
example of information in redefining a problem is that of pesticides
(see Baumgartner & Jones, 1993). Early on, policies toward pesticides
were mostly favorable and aligned along a single dimension of choice—
agricultural production. Environmental and fisheries dimensions began
to structure the generation of solutions alongside agricultural produc-
tion as the dangers of pesticides were exposed. The shifting of emphasis
to balance environmental protection, fisheries management, and agri-
cultural production was the result of focused attention on alternative
dimensions of choice.
Jones (1994, 2001) develops two important extensions of the bounded
rational paradigm and its application to policy dynamics. First, atten-
tion is a limited resource, and shifting attention is key to understand-
ing choice in public policy. Second, organizations are efforts to expand
upon the attention limits of individuals, though they face similar limits
at some level of aggregation. The policy process often serves to focus
attention on particular problems, making attention limitations more
acute (Jones & Baumgartner, 2005; May et al., 2008).
It is during the construction of the problem space when shifting atten-
tion is perhaps most consequential. The end result is a working definition,
or representation, of the problem that structures the generation of solu-
tions or alternatives (Dery, 1984, p. 25; Jones, 2001, p. 66; Simon, 1979,
p. 147). Thus, shifting attention among various dimensions, or attributes,
of a problem may forestall entire sets of solutions and abet others.
The parsing of problem solving into the problem space and solution
space instigates a paradigmatic fault line in the base conceptualization
of information as a private good versus being freely supplied in the
competition for ideas. Where the generation of solutions or alternatives
is concerned, information is conceptualized as a private good. In the
solution space, the mechanism that fosters information generation is an
incentive system, or a contract. The process of search for information is
then governed by economy, efficiency, and marginal returns for search.
Where the construction of a problem space is necessary, information—
at least in politics—is supplied freely in competition to define a prob-
lem and steer resultant policy debates. The mechanism for information
generation in the problem space is pluralistic competition. Here, the

process of search for information is governed by uncertainty, noise, and
issue prioritization. The following sections discuss the two different
tracks for information generation and search.
Information as a private good
George Stigler (1961) was probably the first economist to consider the
economics of information. He set out to develop a theory of price dis-
persion and come to grips with the costs and efficiencies associated with
the search for information. In Stigler’s model, the search for information
was governed by decreasing marginal returns to search as price stabilized
at an equilibrium with additional search, and increasing opportunity
costs to continued search.
Stigler’s was more a theory of the economics of search than of the
economics of information supply. Nevertheless, that information was
governed by the standard principles of economics carried forward into
subsequent work in both economics (Arrow, 1984) and political sci-
ence, where it was best embodied in work using principal-agent models
(Mitnick, 1975).
Incentive-based information supply

Principal-agent models in political science are borrowed from transac-
tion cost economics. They are incredibly useful in understanding policy
dynamics when the problem is well defined and the critical informa-
tion pertains to the set of solutions. Information asymmetry lies at the
core of the principal-agent relationship and has two different compo-
nents. Adverse selection pertains to the choice of an agent—agents have
information concerning their ability to perform, and principals do not.
Moral hazard pertains to the inability or costliness of monitoring agent
performance.
Early on, studies employing principal-agent models in political science
focused almost entirely on the monitoring problem (McCubbins, 1985;
Moe, 1985). Studies developing theories of delegation also continued to
focus empirically on the monitoring problem, rather than the selection
of an agent (Epstein & O’Halloran, 1999; Huber & Shipan, 2002).
Just as a theory of control requires a theory of delegation; a theory of
information requires a theory of search. Principal-agent models implic-
itly assume that problem solving requires what Baumgartner and Jones
(2014, pp. 79–80) label expert search. In other words, it assumes that
the key information needed pertains to the relationship between a
policy and the desired outcome (and perhaps the probabilities attached
to these), or alternatively, to the information relating preferences to
particular policy alternatives and potential biases resulting from pref-
erences of actors over policies. When the conditions associated with
expert search hold, transaction costs approaches to public policy in gen-
eral, and principal-agent models in particular, are powerful analytical
tools for understanding government problem solving. In illuminating
the connections between information as a private good, expert search,
and signaling, it is worth considering these conditions in greater detail.
Expert search and solution definition

To leverage the power of expert search and incentive structures, certain
conditions concerning the nature of the problem, the principal’s goals
or preferences, and the principal’s knowledge of the agent are helpful.
The problem must be in some sense ‘known.’ There are two compo-
nents to this. Principals and agents alike need to have had a consider-
able amount of experience dealing with the problem. This implies that
both principal and agent have enough experience with the problem to
have well-defined preferences over available solutions.
There is, however, a much more important aspect of a ‘known’ prob-
lem—adverse selection is not so pernicious because the appropriate
agents are likely known, or at least reducible to a small pool. In our com-
munications perspective on policy dynamics, these interactions occur in
what amounts to a repeated game meaning principals accurately weight
information communicated by agents (Workman, 2015). This is impor-
tant because it relegates preference and policy bias to a nuisance rather
than a dire threat to democratic accountability. Known problems pre-
sent paradigms that are already in place.
The degree to which a problem is known is not entirely in the hands
of the policy makers, however. Many important policy problems have
characteristics that engender complexity either with regard to their poli-
tics or the substantive dimensions comprising the issue. Expert search
is most powerful when the problem has a simple dimensional structure,
or if not, at least it does not present tradeoffs among the dimensions.
Policy makers may at times be able to artificially reduce the complex-
ity of a problem by prioritizing certain types of information, certain
‘signalers,’ or choosing to focus on a particular dimension of the prob-
lem. Members of Congress have from time to time chosen to discuss
energy policy, for example, only in terms of productive capacity, ignor-
ing important environmental, infrastructure, and agricultural dimen-
sions. This will not always be possible. The case of immigration policy
illustrates the inability of policy makers—politicians and bureaucrats

alike—to reduce a complex problem in order to steer policy. In this case,
the politics of the issue mitigate less complex understandings of the
problem that would allow comprehensive immigration reform. Given
problem solving that allows for expert search, we can begin to frame the
general features of signaling and communication in the first track for
our communications perspective on policy dynamics.
Signals and incentives

With the nature of the problem and the search process in place, we
can begin to understand how information is communicated. There are
two types of uncertainty associated with expert search and well-defined
problems. The first is bias associated with ideological, or policy prefer-
ences. The second is the uncertainty with which proposed solutions to
the problem map onto given outcomes; or more accurately, what is the
conditional distribution of outcomes given the policy choice? The infor-
mation that is generated and communicated will pertain to these two
forms of uncertainty.
It is here among simple dimensional structure and repeated interac-
tions that the power of incentive structures shine. Policy makers, par-
ticularly presidents and legislators, are able to induce the generation of
information and steer policy dynamics across a range of institutions and
issue settings (May & Workman, 2009; McCubbins, 1985; Moe, 1985).
What does this mean for the dynamics of signaling? Gailmard and
Patty (2012) suggest that signaling of the type described above fall into
one of three broad categories: cheap talk and communication, commu-
nication and costly signaling, and communication with verifiable infor-
mation. These three broad categories essentially represent a continuum
of creditability (Gailmard & Patty, 2012, p. 363). A key insight in the
dynamics of signaling across institutions within this framework is that
the risk-averse nature of actors and the uncertainty surrounding the
meaning of a signal require the repetition of the signals for the desired
effect (Carpenter, 1996). There is then considerable adaptivity associated
with signaling and signal processing that is conditioned on the visibil-
ity, or verifiability, of credibility.
While these approaches to signaling and signal processing lay a foun-
dation for understanding policy dynamics under expert search, these
models rarely touch upon the nature of information generation. We
think understanding the influence of credibility and communication on
the quality and quantity of information generation among agents rep-
resents a fruitful direction for future research on signaling models under
expert search. The differences in the types and quantity of information

generated in incentive-based versus competition-based systems in par-
ticular promises to further our understanding of information supply and
policy change.
Competition-based information supply
If problems are not well known, or if simpler dimensional structure can-

not be induced, entropic search is necessary to define the problem and
generate solutions (Baumgartner & Jones, 2014, pp. 79–80). Entropic
search is beneficial for constructing problem definitions, and involves
a broader search for potentially relevant parameters of choice. Without
problem construction, solutions cannot be generated. Given problem
complexity, and interdependency, the economics of search break down
and the costs of failing to incorporate relevant dimensions rises dra-
matically. Further, complex issues with multiple relevant dimensions
make contracts and incentive structures difficult, as there will generally
be multiple principals and multiple agents operating within the issue
(Whitford, 2005).
Adverse selection makes incentive structures difficult to design in these
situations for two reasons. First, the principal must choose an agent with
whom to contract, which is made difficult in complex problem spaces
with multiple potential ‘signalers.’ Because contracts become difficult in
the presence of more than one agent (the signaler), this means foregoing
the relevant information from other signalers in the space (i.e. incur-
ring these opportunity costs). The second is that the principal, or policy
maker, may attempt to structure incentives across multiple substantive
dimensions, which is very difficult when dimensions are interdepend-
ent or present tradeoffs.
Scholars of the policy process do not find the characteristic settings of
privately held information, especially in the context of issue complex-
ity. In contrast to incentive-based mechanisms for information supply,
policy scholars depict information freely supplied in the competition
for ideas, to define policy problems, and to steer policy debates. Two of
the major perspectives on the policy process embody this approach to
information search.
The information processing perspective focuses on how attention shifts
among substantive issues, or dimensions of issues, shape policy agen-
das, the way issues come to be understood, and how these map onto
policy change (Baumgartner & Jones, 2014; Jones & Baumgartner,
2005; Workman, Jones, & Jochim, 2009). Within this perspective, the
information generated in the policy process is oversupplied and results

from endogenous processes of positive feedback in the government
system coupled with jurisdictional overlap (Baumgartner, Jones, &
MacLeod, 2000; Jones & Baumgartner, 2005; Workman, 2015).
Information processing leads to an institutional and substantive desta-
bilization of existing governing arrangements. As attention shifts, high-
lighting new issues or issue dimensions, subsystems destabilize bringing
in new actors, or shifting consideration of the issue, to the highest
reaches of government (Baumgartner & Jones, 1993). At the same time,
new issues and arguments destabilize existing problem definitions,
changing the way issues are understood and the dimensions of choice
that are relevant for policy making. The pattern of issue prioritization
and disproportionate information processing leads to the characteristic
fat-tailed policy change distributions, where both large and incremental
policy change appear in larger than expected frequencies both in the
US and comparatively (Baumgartner et al., 2009; Jones, 2012; Jones &
Breunig, 2007; Jones et al., 1998, 2009, 2011).2
For scholars working in the information processing perspective, infor-
mation is usually conceptualized in terms of the substantive dimensions
of an agenda, or an issue. But, it is not the only theory to incorporate the
notion of competitively generated information. The advocacy coalition
framework (ACF) also places information at the center of policy dynam-
ics, focusing in particular on the incorporation of scientific and technical
information in protracted coalitional struggles to shape resultant policy
change (Lodge & Matus, 2014; Weible & Sabatier, 2009). Policy change
for these scholars is connected to policy learning and belief change as
coalitions compete for influence. The provision of information is itself a
strategy of influence in this perspective.
Theories of the policy process thus depict competition-based infor-
mation supply. Whereas incentive-based information supply starts from
the assumption that holding information privately has advantages,3
competition-based information supply holds that the free provision of
information has advantages. In part, this is because these perspectives
place tremendous importance on agenda setting and problem defini-
tion. If an actor fails to provide information pertaining to the problem,
especially the construction of the problem space, the actor runs the risk
of another doing so. In these settings, structured incentives are not nec-
essary as the competitive environment fosters information supply.
While the study of the policy process has provided a second track for
information search, it is far less developed in understanding how infor-
mation is transmitted among political institutions. In what follows, we
develop a perspective on policy dynamics based in a communications

framework. We develop a model of signaling that is rooted in communi-
cations, and builds from the premise that information is competitively
supplied in the policy process. Our perspective is particularly useful for
understanding the dynamics of agenda setting and problem definition.
Communication, signaling and policy dynamics
Stigler’s (1961) economics of information sees a consumer shopping

potentially far and wide for price information, all the time facing
decreasing returns and increasing costs. The policy process is more akin
to two gas stations across the street from one another competing to
bring the price to the customer. The foundation of our communications
perspective is that information is competitively supplied in the policy
process. Further, we are particularly interested in information supply
under conditions of entropic search and problem definition.
Under conditions of entropic search, where problems are complex or
not well defined, policy makers face two problems. Bias in the informa-
tion supply remains a problem, except it now appears alongside signal
noise, as signalers compete for influence by oversupplying information.
Following Workman (2015), the central feature in our perspective is that
policy makers strategically manipulate the competition among actors in
supplying information. In doing so, policy makers are able to triangulate
bias in circumstances of extreme uncertainty. They are also able to ‘tune’
the supply of information emanating from bureaucrats, interest groups,
and other actors.
Information supply under pluralistic competition

Central to our communications model of policy dynamics is that pol-
icy makers can strategically manipulate the competition faced by other
actors in supplying information (Workman, 2015). It is important to
understand why information is oversupplied in the policy process,
especially in situations of entropic search. Problem space construction
occurs as a result of two configurations of problems. First, problems may
be complex, containing multiple substantive dimensions that are inter-
dependent of present tradeoffs. Second, problems may not be defined.
The oversupply of information is a result of problem definition,
where actors must define problems in such a way as to allow govern-
ment action. When viewed from the perspective of policy makers, par-
ticularly administrative officials, problems present what Dery (1984,
pp. 24–7) labels bridgeable discrepancies and opportunities. In the former,
deficiencies in the existing set of policies are observable. Bureaucracies

then define the gap in a way that makes their interjection possible. In
situations of a bridgeable gap, bureaucracies act out of risk averseness,
hoping to avoid the destabilization of other institutions (who may view
the same gap as an opportunity) interjecting in policy making. Where
problems are viewed as opportunities, bureaucracies define problems in
such a way that expands their responsibility, or jurisdiction. In both
cases, information is generated out of competition or the threat thereof.
Given a set of bureaucracies occupying a complex problem space, and
highly motivated to provide information, the adverse selection problem
facing policy makers may seem insurmountable. The key is that policy
makers in this situation need information about the problem space.
Given bridgeable gaps and problems viewed as opportunities, bureaucra-
cies generate a supply of information, meaning that search costs are low
compared to the deleterious effects of failing to incorporate a relevant
dimension of the problem.4
The adverse selection problem applied to problem spaces also proposes
a false dichotomy—delegation must be to one bureaucracy or another
for the implementation of a solution. In the case of problem definition,
we think that it is more accurate to think of policy makers, or principals,
as ‘tuning’ the information supply (Workman, 2015). Entropic search
costs less because actors ‘flood’ to problems, supplying information in
an effort to steer policy debates, secure jurisdiction, and make policy.
Under these circumstances, policy makers may choose to blend informa-
tion suppliers, or signals, effectively tuning the information supply to
account for complex dimensionality.
But, what of the problem of bias? In general, policy makers can weight
for bias in one of two ways. The first is through experience. In repeated
interactions over time concerning a given set of problems, principals
are likely well aware of agents’ preferences. The second is triangulation.
Because bureaucracies and other actors flood to problems, policy makers
are able to triangulate bias in one actor by weighting against the others.
This is particularly useful in a situation where relevant agents are not
clear. The quality of signals is shaped by the repeated setting in which
information is supplied coupled with the triangulation of information
channels in many types of governing arrangements.
Policy makers may stimulate or deaden competition regarding any
given problem. Importantly, they may do so without the need to design
an incentive structure. Policy makers are able to induce competition and
‘tune’ and information supply through processes of issue shuffling and
issue bundling (Workman, 2015, Chapter 5). This means that principals
are able to expand, or contract, the flow of information as dictated by

the complexity of the problem and costs of search without the need to
lock in structured incentives.5 With attention to information processing
and the nature of information provision, our theory depicts a dynamic
and adaptable process of information supply as an input to government
problem solving. This dynamism is plagued by the twin problems of
complexity and noise, both exacerbated by competition over the infor-
mation supply.
Complexity, noise, and signaling

Despite being able to undermine the adverse selection problem and tri-
angulate for bias, the twin problems of complexity and noise remain.
Policy makers have to deal with each of these problems in adjusting
the supply of information. Employing conflict expansion greatly aids in
adverse selection and the problem of bias; it exacerbates the complexity
and noise associated with the information supply. Yet, policy makers
face tremendous opportunity costs to closing off information along dif-
ferent dimensions of the problem because uncertainty surrounds the
nature of the problem itself.
The complexity of many important policy problems is both spatial
and displays dynamism. First, it is spatial because any subsystem repre-
sents a substantively distinct dimension of choice for the formulation of
a problem definition. Second, it is dynamic in that there are gradations
of stability through time within each of the dimensions (this is easily
seen in the ‘subsystem EKGs’ of May et al. [2009a, pp. 176, 184–5]).
Faced with considerable complexity, policy makers have two avenues
for reducing complexity in service of problem solving. First, policy
makers may shift attention amplifying one or a few dimensions of the
problem. This of course incurs the opportunity costs of ignoring poten-
tially relevant dimensions of the problem (May & Workman, 2009; May
et al., 2008). Second, policy makers may rely on signalers with a history
of activity on the issue.
There is evidence that existing suppliers of information have a privi-
leged position as suppliers in the face of problem complexity (May et al.,
2009a, pp. 802–8). This suggests policy makers faced with a complex
potential problem space have an inclination to rely on historically sta-
ble channels of information. This is especially true at the initial realiza-
tion of a complex problem (LaPira, 2014, pp. 242–3).
The dynamics of problem space complexity and strategies for dealing
with it leave us to question the effects of these processes on the genera-
tion of information in subsystems. In particular, our approach suggests
that policy makers may value consistency and stability alongside accu-
racy and quantity. Most of the examination of volatility in public policy
has occurred on the output side of the process. We argue that govern-
ment problem solving has implications for the generation of informa-
tion on the input side, especially in and among policy subsystems. In
addition, we argue that it highlights the importance of thinking about
policy analysis in theoretical treatments of policy dynamics. The follow-
ing section discusses our communications approach in the context of
emerging areas of research in public policy processes.
Relevance for policy processes
Our approach develops a signaling model whose general components

are based in a communications perspective. It is focused on the informa-
tion processing among political institutions that generate information
and signal one another about the nature of policy problems. The sys-
tem we describe is animated by the dynamics of competition in supply
information and defining policy problems. With multiple competing
channels of information, it is oversupplied creating both complexity
and tremendous amounts of noise in government problem solving. If
we are to understand paradigms as the organizing ideas for approaching
a policy problem, then much of the policy process involves conflicting
paradigms or efforts to impose paradigms.
Our approach has important implications for emergent subfields in
the study of policy dynamics. These include research addressing bound-
ary-spanning or trans-subsystem problems, the imposition or emer-
gence of policy regimes, and how policy process theories address the
role of policy analysis in the policy process. We pose these as fruitful
areas of research for exploring questions generated by our communica-
tions-based approach to signaling and information processing.
Boundary-spanning problems and policy regimes

Boundary-spanning problems present great difficulty for policy para-
digms precisely because these cross-cutting issues involve many dis-
parate paradigms, often in direct contention with one another. The
politics of boundary-spanning policy problems represent an area of criti-
cal importance for understanding information generation, complexity,
and noise. These are policy problems that intersect various features of
many different subsystems, including conceptual understandings of the
problem, actors involved, institutional jurisdictions, and partnerships
that span the public and private sector (Jochim & May, 2010, p. 304).
Examples of problems that are boundary-spanning include homeland

security (May et al., 2011), critical infrastructure (Koski, 2014; May &
Koski, 2013), food safety (May, Jochim, & Pump, 2013), and climate
change (Jones & Jenkins-Smith, 2009).
These issues span multiple subsystems and defy definition in terms
of a single understanding of the issue. Many of these problems exhibit
tradeoffs and other nonlinear dependencies characteristic of complex
systems. In this context, the noise inherent in generating and commu-
nicating information puts a premium on the ability of policy makers at
higher levels of government to prioritize and tune the supply. For our
purposes, it is worth considering what the politics of boundary-spanning
problems mean for information generation and resultant signaling from
the vantage point of information suppliers, particularly bureaucracies
and associated interests.
As suppliers of information, bureaucracies and other actors face a two-
tiered set of choices when confronted with a boundary-spanning prob-
lem. The first is whether to continue to generate information. This is
not a trivial decision. Without directives from above, a bureaucracy may
choose to define the problem out of, or into, its jurisdiction if possible.
The second choice pertains to whether an agency continues to generate
similar information as in the past, or alternatively, chooses to redefine
its definitions and pursuant information in light of the new dimensions
of the problem (Workman, 2015, Chapter 5).
These choices alter the mix of information that makes its way up
the machinery of government for consideration and decision. They
are also not easy choices when one considers the competition inher-
ent in boundary-spanning problems. Faced with a boundary-spanning
problem, many actors are drawn into a situation of competition over
information provision that is unfamiliar, particularly in that any actor
may compete against unfamiliar actors employing unfamiliar strategies
(Workman, 2015; see also Nohrstedt & Weible, 2010). We must note, too,
that the reverse is true. Institutions used to competition may suddenly
find themselves alone in the problem space. What do these dynamics
mean for the supply of information fueling policy dynamics? We see
this as an important question for policy scholars moving forward.
Policy regimes constitute one institutional attempt to address emergent
policy problems spanning the boundaries of multiple subsystems. Policy
regimes can be thought of as attempts to impose a paradigm from the top
onto boundary-spanning problems. Regimes emerge as top-down institu-
tional attempts to carve problem definitions from multiple institutions
and sources of information. Jochim and May (2010, pp. 322–3) argue that
policy regimes emerge under focused attention on the problem among

subsystems with permeability. When regimes emerge under these condi-
tions, their success, or failure, is largely dependent on whether existing
problem definitions and the institutions holding them weaken the status
quo or reinforce it. Again, we consider the implications for the supply of
information in policy regimes.
A regime striking the right balance between centralized direction,
coordination, and delegation of information processing stands to incor-
porate complexity, rather than ignore it. This is easier said than done.
We suggest that the imposition of what Jochim and May (2010) label
an ‘anemic’ policy regime alters the information supply in one of two
ways. First, the information supply may shift and focus attention on
one dimension at the expense of others. This results in policy bereft
of information on critically important dimensions of the policy prob-
lem (May et al., 2008). Second, the supply may become destabilized,
such that the relevant dimensions are present for consideration, but
the information supply is highly volatile (May et al., 2009a). We argue
for deeper development of the concept of policy regimes attentive to
their influence on information generation in boundary-spanning pol-
icy problems.
Feedback, policy analysis, and goal refinement

Another area where we see our approach contributing to understand-
ing policy dynamics concerns feedback, goal refinement, and policy
analysis. One of the central debates concerning policy paradigms is
over the nature of paradigm shifts and evolution. We see policy feed-
back as key to understanding paradigm shifts, evolution, or drift. The
information generated in subsystems forms a key source of system feed-
back in government problem solving (Workman, 2015, Chapter 6). The
information generated by federal bureaucracies especially provides a
foundation for problem definition and goal refinement or adjustment
into the future in an iterated process of information supply and tuning.
A particularly vital component of this feedback is policy analysis at the
federal level.
Policy scholars have attended to the role of policy analysis in policy
change for some time, particularly the incorporation of science and
technological information in policy debates (Lodge & Matus, 2014).
Early works attempted to situate policy analysis within the larger frame
of democratic politics and the role information should, or does, play
in informing policy (Jenkins-Smith, 1990; Williams, 1998). In general
these works tend to focus on how policy analysis shapes public policy
outcomes, or how it shapes institutional changes (Jenkins-Smith, 1988).
Less attention has been given to how policy analysis, as a feedback
mechanism in the policy process, shapes policy agendas and affects the
information supplied by other institutions. This is an especially impor-
tant area of study given the rise of large ‘analytical’ bureaucracies in the
US and their counterparts in other Western democracies (Workman, 2014,
p. 1). These are macro bureaucracies who exist solely to generate informa-
tion and analysis for the elected branches of government. We think it is
worth considering the substantive content of policy analysis, and what
that might signal to suppliers of information in policy subsystems.
We argue that the substantive focus of policy analysis impacts infor-
mation generation in at least one general way resulting from the over-
sight position many of these institutions hold. If policy analysis focuses
on internal management and matters of organizational efficiency, the
information generated by administrative units is likely to pertain to
matters of management and efficiency to the potential exclusion of
information pertaining to policy goals and outcomes. Alternatively, if
policy analysis focuses on policy goals and outcomes, administrative
units may sacrifice efficiency and internal management to meet these
goals or expectations. A focus on one general type of information obvi-
ously deprives the policy-making system of the other. This is especially
important given the propensity of organizations to process these signals
disproportionately (May et al., 2008).
We think of policy analysis as a mechanism for focusing attention.
In this way, policy analysis can be thought of as a signal that shapes, or
tunes, the information supply dependent on what bureaucracies learn
from the analysis. This conception of policy analysis as a powerful sub-
stantive signal is perhaps even more useful with the rise of powerful
institutional actors supporting, conducting, and disseminating policy
analysis in a wide range of fields and settings (Carlson, 2011). We argue
that a more robust conceptualization of the relationship between policy
analysis and problem definition is vital to understanding policy dynam-
ics with the increasing visibility and importance of these institutions.
Our approach to signaling turns the focus on how information is
generated in the system. Policy makers’ concern for both accuracy and
consistency in information supply leads us to expect some particular
patterns of information generation in policy subsystems. Our argument
highlights the central role played by bureaucrats in supplying informa-
tion and defining policy problems (Miller, 2007; Workman, 2015).
Information supply in policy subsystems
A policy subsystem consists of issue-specific experts from three sources:

congressional committees with jurisdiction, relevant bureaucracies,
and vested interest groups (Baumgartner & Jones, 1991; McCool, 1998;
Thurber, 1996). A simplified textbook example of a subsystem would
include the US House Committee on Energy and Commerce, the US
Department of Energy, and the American Coalition for Ethanol. An
actual subsystem consists of multiple committees, bureaucracies, busi-
nesses, and interest groups with actors entering and exiting over time.
The subsystem approach has remained popular over the years because it
helps to simplify bureaucratic politics, public policy, and policy imple-
mentation (McCool, 1998).
In essence, subsystems are information processors, best understood in
terms of the supply and prioritization of information (Workman et al.,
2009). Subsystem actors, such as bureaucrats, interest groups, and busi-
nesses, are motivated to produce and supply information to Congress
with the hope that it will influence problem definitions and policy mak-
ing in their particular issue area. These actors will not choose to withhold
information because doing so would give another actor a greater oppor-
tunity to influence policy making in their stead (Simon, 1983; Workman
et al., 2009). The result is that subsystems produce an oversupply of
information and congressional committees are tasked with prioritizing
it (Jones & Baumgartner, 2005). How the information is prioritized will
depend on a number of factors including the congressional committee,
the policy area, as well as the presence of problem uncertainty and crises
(all of which will be discussed in detail later in this chapter).
Thurber (1996) also suggests that subsystems can also be thought of
as open communication networks in which vested interests, bureau-
cratic agencies, and congressional committees share valuable informa-
tion. The power of a bureaucracy or interest group within a subsystem
is directly related to the quality and variety of the information it pos-
sesses (Thurber, 1996). The higher the quality and greater the variety of
information the actor possesses, the more power that it has to influence
policy making. Both bureaucrats and interest groups are aware of this
relationship and compete to provide Congress with quality information
(Esterling, 2004; Workman et al., 2009). The competition to supply infor-
mation is amplified by the actors’ needs to influence policy making and
the limited nature of the enterprise—access granted to one interest or
agency means access is less likely to be granted to another. Further, both
bureaucrats and interest group representatives are incentivized to supply
valuable information because the more useful the information is to the

committee, the more likely the committee will be willing to listen (see
Esterling, 2004). While research tends to focus on bureaucratic expertise
in the two areas previously discussed, federal bureaucrats are involved
throughout the policy process, especially during the problem definition
stage (Katzmann, 1989; Miller, 2007; Workman, 2015, Chapter 6).
The bureaucratic channel
Our communications approach expects that there is bureaucratic advan-

tage in supplying information, and further, that this advantage increases
with congressional uncertainty in the problem space of issues. To under-
stand why this is so, it is useful to consider what aspects of informa-
tion are valuable to policy makers, in our case, to member of the US
Congress. Further, it is also helpful to consider why bureaucrats possess
an advantage in these regards given that any issue at the federal level is
occupied by other powerful signalers (businesses, think tanks, citizens’
groups, etc.).
Where interest groups are concerned, bureaucratic influence runs up
against two very persuasive visions of policy making in the administra-
tive state. Theories of subsystem politics and regulation posit regulatory
capture (Stigler, 1971). Bureaucracies become the prisoners of those they
regulate. Federal agencies become an instrument of organized groups
and their powerful congressional allies in the provision of the benefits
of policy making.6 For proponents of regulatory capture theory, all reg-
ulation is in the end distributive politics. This is a view of regulation
very symbiotic with the subsystems literature of the late 1960s, which
saw ‘iron triangles’ holding the broader American public hostage while
providing particularized benefits to powerful organized interests by way
of public administration (Fiorina, 1977; Lowi, 1969).
Meanwhile, scholars working within the contractual paradigm were
positing theories of interest group influence within the structure of
legislation (McCubbins, Noll, & Weingast, 1987). The proponents of the
congressional dominance in control of the bureaucracy put forth the now
famous ‘deck stacking’ or ‘McNollGast’ thesis. Stated plainly, in an effort
to deliver the benefits of policy to favored organized interests, Congress
would construct administrative procedures so as to ‘stack the deck’
against federal bureaucratic agencies. Even more so, these administrative
procedures would be adaptable such that as the constellation of organ-
ized interests in the system changed, the administrative system would
accommodate this, seamlessly transferring benefits to the new interests.
In this way, Congress solved two problems: the delivery of policy ben-
efits to favored constituent interests and the reduction of the heavy costs
entailed in micro-managing the federal bureaucracy (police patrol over-
sight). Both capture theorists and those scholars working in the contrac-
tual paradigm saw the administrative system as a tool to favor organized
interests in the policy process.
A basic weakness common to these distributive approaches concerns
the nature of congressional decision making in situations rife with
uncertainty. In order for distributive theories of policy making to hold,
it must be true that the actors involved in the process (bureaucracies,
congressional committees, and interest groups) know what it is they
want. Absent a fairly clear notion of what it is the parties involved
desire from the process, politicians will find it difficult to design a sys-
tem geared toward the delivery of amorphous policy benefits. Again,
adverse selection is problematic in this context. The best course of
action to get what one desires from any decision-making process, espe-
cially one filled with uncertainty, is not always clear. Very often, the
actors involved are engaged in the refining of policy goals. Where the
bureaucracy is concerned, Dodd and Schott (1979, p. 2) have noted
that the bureaucracy engages in the ‘refining of congressional intent.’
In general, bureaucracies are a critical player in defining what it is
actors involved in the process want by way of their efforts to influence
problem definitions and adapt goals to divergent policy environments
(Miller, 2007; Workman, 2015).
Much of this debate also obscures what is true of the process of regula-
tory policy initiated by Congress in its own image (Rosenbloom, 2001).
Once bureaucracies set their agendas, the Administrative Procedures
Act of 1946 ensures the incorporation of group and citizen preferences
in the shaping of regulatory policy (Workman, 2015; Yackee, 2006).
In this sense, the regulatory process in particular serves to aggregate
and legitimate interest group preferences for particular definitions and
solutions.7
Interest groups tend to generate information as a by-product of other
activities. Federal bureaucracies generate information as a matter of
course. In part, they are required to do so as a result of oversight, and in
part, they do so in the course of problem solving as a result of proximity
(Workman, 2015). Further, many units in the US government are created
for the sole purpose of generating information and analysis (Carlson,
2011, pp. 21–3; Workman, 2014). Bureaucracies collect information that
benefits them not only in making decisions during the implementation
process, but also to use at other points throughout the policy process
(Katzmann, 1989; Workman, 2015).
The Energy Information Administration (EIA) is an excellent example

of exactly this type of congressional delegation as it is charged with
both policy surveillance (see Feldman & March, 1981) and identifying
and defining policy problems (see Katzmann, 1989; Workman, 2015). In
criminal justice policy, Miller’s study (2007) shows that bureaucrats are
thoroughly involved in defining policy problems and alternatives. The
EIA, for example, regularly publishes a variety of measures of the state of
energy policy, including the amounts of energy produced and consumed
by source and prices by source. In turn, this information can be used to
define emerging problems and redefine existing ones (Katzmann, 1989;
Workman, 2015). The per barrel price of crude oil, a common metric
generated by the EIA, is frequently used in discussions surrounding the
state of the US economy.
Bureaucrats are free to bias information in favor of increased agenda
control, but this tendency for biased information is in part mitigated
by two factors. The first is that bureaucrats, by virtue of longevity, are
engaged in a repeated interaction with Congress in an effort to define
policy problems (Aberbach, 1990). What this means is that while bureau-
crats may very well bias their provision of information to policy makers,
the latter are likely aware of the biases (detected over decades of inter-
action) and are able to weight these biases more accurately (compared
to groups) in their decision calculus. Even if provision of information
is biased, repeated relationships allow for the systematic adjustment of
stance based on even biased information if policy makers are able to
induce competition and triangulate.
Sometimes smaller variance trumps pinpoint accuracy, especially
where this accuracy often comes at the cost of some very bad misses.
Private information, put simply, is revealed in the competition for ideas.
The ideas and argument presented above lead to the general expectation
that bureaucrats will be influential in the generation of information,
processes of congressional search, and most consequentially, in problem
definition. The logic of bureaucratic influence offered here generates
some testable hypotheses concerning interest group participation rela-
tive to the bureaucracy in policy making.
Congressional prioritization of information

Congressional policy making requires information about the impor-
tance of problems (Baumgartner & Leech, 1998; Hilgartner & Bosk,
1988), the impact of solutions (Baumgartner & Jones, 1993; Burstein &
Hirsh, 2007), and reelection implications for officeholders (Burstein &
Hirsh, 2007). Information required to create public policy can be broken
down into two broad categories: policy and political.
Committee hearings act as the filter for the incoming information

because they are an efficient way to gather political and policy infor-
mation (see Burstein & Hirsh, 2007; Gormley, 1998; Kingdon, 1981).
Committees use hearings to inform a number of legislative activities,
including defining problems, limiting the solution space, evaluating
the merits of an existing bill, and drafting new pieces of legislation. By
inviting certain subsystem players to testify and not others, committees
prioritize some sources of information over others.
Do congressional committees prioritize bureaucrats in the informa-
tion supply? Energy policy provides an instructive example. Shafran
(2014) analyzes congressional committee hearings on energy policy
from 1995 to 2010. As a result of the collection and coding of 5,774
witnesses appearing across 828 hearings, Table 12.1 displays the prioriti-
zation of various types of witnesses in the hearing process.8
Table 12.1 Witnesses at congressional hearings on energy policy, 1995–2010
Frequency Percentage (%) Mean SD
Government
Federal bureaucracies 1,587 63.5
Members of Congress 306 12.3
State and local governments 603 24.2
Total 2,496 43.2 2.99 8.90
Business
Big business 852 52.5
Small business 771 47.5
Total 1,623 28.1 1.94 5.82
Interest group
Nonprofit 688 53.3
Trade association 514 39.8
Professional association 49 3.8
Labor unions 35 2.7
Other 4 0.3
Total 1290 22.3 1.54 4.10
Public institutions
Hospitals 258 99.6
Schools 1 0.4
Total 259 4.5 0.31 0.52
Others
Individual citizens 61 57.5
Native Americans 35 33.0
Foreign representatives 7 6.6
Others 3 2.8
Total 106 1.8 0.13 0.24
Overall total 5,774 100 6.91 28.18
Table 12.1 shows that government officials comprise 43.2 per cent, a
strong plurality, of all witnesses called to testify at congressional hearings
on energy policy. Of these, the vast majority were federal bureaucrats
at 63.5 per cent. The table suggests that, within federal energy policy,
administrative officials have a greater presence than any other group
represented. Businesses are the second most prevalent group, making
up 28.1 per cent, with an almost even split between large and small
businesses. The table provides descriptive support for our arguments
concerning the prominence and advantage enjoyed by bureaucrats as
suppliers of information.
The bureaucratic composition of the witnesses is also instructive.
Table 12.2 breaks down the witness composition of federal bureaucrats
into the major departments and independent agencies represented in
the data. We would expect that the vast majority of these witnesses
would come from bureaucracies whose job it is to monitor for problems
in energy policy.
Table 12.2 shows a majority, 58.7 per cent of federal bureaucrats, at
hearings on energy policy are from agencies whose job it is to moni-
tor for energy problems. What is perhaps more striking is that there
is considerable competition from bureaucracies in interdependent sub-
stantive areas. Public Lands and Water policy, along with Agriculture
and the Environment are issues intertwined with energy policy in the
US, especially with the onset of climate change and associated initiatives
for alternative energy sources. These bureaucrats make up 15.7 per cent
of the witnesses testifying at congressional hearings on energy policy.
Table 12.2 Bureaucratic composition of witnesses in energy policy, 1995–2010
Substantive area Bureaucracies Frequency Percentage (%)
Energy DOE, NRC, EIA 931 58.7

Land, environment, and DOI, EPA, USDA 248 15.6
agriculture
Research and coordination GAO, Pres Task Force 289 18.2
Defense, intelligence, and DOD, State, NSA 119 7.5
foreign affairs
Total 1,587 100
Note: DOE—Department of Energy; NRC—Nuclear Regulatory Commission; DOI—

Department of Interior; EPA—Environmental Protection Agency; USDA—United States
Department of Agriculture; GAO—Government Accountability Office; DOD—Department
of Defense; NSA—National Security Agency; Pres Task Force—multiple task force assembled
by the president to investigate/research a policy area, usually officially titled President’s
Task Force on ‘X’ or White House Task Force on ‘Y’.
Table 12.2 shows that there is competition in the supply of information,

even among bureaucracies. It also shows that the complexity inherent
in energy policy is reflected in the search for information in Congress.
Descriptively, energy policy in the US follows the pattern our commu-
nications perspective would expect with regard to the composition of
suppliers of information.
Conclusion
In this chapter, we have laid out a communications perspective on

policy dynamics that is based in information generation, processes of
search, and a model of signaling. Our theoretical point of departure
for developing a signaling model based in communications is the pars-
ing of organizational problem solving into its component problem
and solution spaces (Newell & Simon, 1972). From here, we build on
Baumgartner and Jones’ (2014) characterization of two distinct pro-
cesses of search in order to depict how signaling occurs under an over-
supply of information.
Our signaling model departs from standard approaches based in eco-
nomics and privately held information. The general contours of our
model are defined by bureaucratic competition to supply information,
which results in an oversupply instead of the undersupply envisioned
in economic approaches. In our argument, bias is triangulated in the
process of inducing competition among actors in a repeated setting.
This oversupply results from the process of problem definition, where
actors supply information and ply their peculiar definitions of problems
in order to steer the policy debate. The classic approaches emphasize
search processes under conditions where the main concern is uncover-
ing the bias in the stream of information associated with ideological and
policy preferences. In contrast, our approach emphasizes the complexity
inherent in defining policy problems and the tremendous noise associ-
ated with signaling in the problem space.
Policy makers are confronted with complexity in the problem space
and a tremendous amount of noise in the information stream. We build
on Workman (2015) in exploring the processes by which policy makers
might navigate this noise and effectively ‘tune’ the supply of information.
In doing so, we are led to consider the aspects of an informative signal or
group of signals, and in turn, how this might privilege federal bureaucrats
in the supply of information in policy subsystems. Using data on wit-
nesses at congressional hearings on energy policy, we explore our inclina-
tions concerning bureaucratic prioritization in the process. Our simple
descriptive analysis is supportive of our expectations concerning federal

bureaucrats as suppliers of information in federal energy policy.
Finally, we implore scholars of the policy process to consider how
the dynamics of signaling and communication structure the genera-
tion of information in the policy process. The information is impor-
tant precisely because it derives from particular paradigms, while at the
same time holding the potential to restructure those same paradigms.
We have outlined a few directions for future research. We think that
consideration of signaling from this communications perspective could
have tremendous benefits, especially in the study of boundary-spanning
problems. Many of the more important policy problems are boundary-
spanning in nature. These types of problems strain established policy
paradigms and are likely key in understanding how paradigms evolve in
the face of an ever-changing agenda. In particular, how policy regimes
influence the nature of the information supply reaching policy makers
is a fruitful area of research. We argue that policy analysis can teach
us much about the nature of communication, signaling, and policy
dynamics. In particular, policy analysis is itself a signal with tremendous
consequences for information generation.
Notes
1 P.L. 88-206.
2 These distributions are characterized by leptokurtosis, meaning that the prob-
abilities of large, drastic change and near stasis are more probable than would
be expected under a normal distribution. This implies that medium policy
changes appear less frequently.
3 Under the conditions of expert search, and decision making in the solutions
space, holding information private certainly conveys some advantages.
4 There is a point in the search process where considerations of transaction
costs give way to decision costs. In the initial search for additional substantive
dimensions of choice, the transaction costs, given oversupply of information,
are low compared to the opportunity costs of missing a relevant dimension of
choice and the error this creates in the system. However, at some threshold,
complex dimensionality will outstrip the ability of individual decision mak-
ers, or even institutions, to prioritize the information and construct a problem
space. In this region of the space, decision costs rise exponentially, decreasing
the value, or at least efficacy, of search.
5 We should note that this results from the constitutional position of adminis-
trative officials and interest groups relative to elected officials in the US and
other Western democracies. These institutions must bring information to
elected officials in order to influence policy.
6 Carpenter (2004) empirically tests capture theory in the context of Food and
Drug Administration (FDA) drug approvals and proposes an alternative theory
based on regulator learning and adaptation over time. The study shed major
light on the alternative sources of large group advantage in regulatory policy.
7 Note that policy activity is instigated by the bureaucracy, even amid the pleth-
ora of groups that may lobby on a particular issue. Baumgartner et al. (2009)
find that most groups registered to lobby in Washington advocate for stasis or
for ‘nothing to happen.’ We can contrast this with Yackee (2012) who shows
that interest groups are sometimes able to ‘block’ an agenda item in the pre-
proposal stage.
8 Hearings were identified using the data found at the Policy Agendas Project
at the University of Texas at Austin. Data and codebooks may be found at
www.policyagendas.org
References
Aberbach, J. D. (1990). Keeping a Watchful Eye: The Politics of Congressional
Oversight. Washington, DC: Brookings Institution.
Arrow, K. J. (1984). The Economics of Information. The Collected Papers of Kenneth
J. Arrow (Vol. 4). Cambridge, MA: Belknap Press.
Baumgartner, F. R., Breunig, C., Green-Pedersen, C., Jones, B. D., Mortensen, P. B.,
Nuytemans, M., & Walgrave, S. (2009). Punctuated Equilibrium in Comparative
Perspective. American Journal of Political Science, 53(3), 603–20.
Baumgartner, F. R., & Jones, B. D. (1991). Agenda Dynamics and Policy Subsystems.
Baumgartner, F. R., & Jones, B. D. (1993). Agendas and Instability in American
Politics. Chicago: University of Chicago Press.
Baumgartner, F. R., & Jones, B. D. (2014). The Politics of Information: Problem
Definition and the Course of Public Policy in America. Chicago: University of
Chicago Press.
Baumgartner, F. R., Jones, B. D., & MacLeod, M. C. (2000). The Evolution of
Legislative Jurisdictions. The Journal of Politics, 62(2), 321–49.
Baumgartner, F. R., & Leech, B. L. (1998). Basic Interests: The Importance of Groups
in Politics and in Political Science. Princeton, NJ: Princeton University Press.
Burstein, P., & Hirsh, C. E. (2007). Interest Organizations, Information, and Policy
Innovation in the U.S. Congress. Sociological Forum, 22(2), 174–99.
Carlson, D. (2011). Trends and Innovations in Public Policy Analysis. Policy
Carpenter, D. P. (1996). Adaptive Signal Processing, Hierarchy, and Budgetary
Control in Federal Regulation. American Political Science Review, 90(2), 283–302.
Carpenter, D. P. (2004). Protection without Capture: Product Approval by a
Politically Responsive, Learning Regulator. American Political Science Review,
98(4), 613–31.
Dery, D. (1984). Problem Definition in Policy Analysis. Lawrence, KS: University of
Kansas Press.
Dodd, L. C., & Schott, R. L. (1979). Congress and the Administrative State. New York:
Macmillan.
Epstein, D., & O’Halloran, S. (1999). Delegating Powers: A Transaction Cost Politics
Approach to Policy Making under Separate Powers. New York: Cambridge University
Press.
Esterling, K. M. (2004). The Political Economy of Expertise: Information and Efficiency
in American National Politics. Ann Arbor: University of Michigan Press.
Feldman, M. S., & March, J. G. (1981). Information in Organizations as Signal and
Symbol. Administrative Sciences Quarterly, 26(2), 171–86.
Fiorina, M. P. (1977). Congress—Keystone of the Washington Establishment. New Haven,
CT: Yale University Press.
Gailmard, S., & Patty, J. W. (2012). Formal Models of Bureaucracy. Annual Review
of Political Science, 15, 353–77.
Gormley, W. T. (1998). Witnesses for the Revolution. American Politics Quarterly,
26(2), 174–95.
Hilgartner, S., & Bosk, C. L. (1988). The Rise and Fall of Social Problems: A Public
Arenas Model. The American Journal of Sociology, 94(1), 53–78.
Subsystems (2nd ed.). Oxford: Oxford University Press.
Huber, J. D., & Shipan, C. R. (2002). Deliberate Discretion?: The Institutional
Foundations of Bureaucratic Autonomy. New York: Cambridge University Press.
Jenkins-Smith, H. (1988). Analytical Debates and Policy Learning: Analysis and
Change in the Federal Bureaucracy. Policy Sciences, 21(1), 169–212.
Jenkins-Smith, H. (1990). Democratic Politics and Policy Analysis. Pacific Grove,
CA: Brooks/Cole.
Jochim, A. E., & May, P. J. (2010). Beyond Subsystems: Policy Regimes and
Governance. The Policy Studies Journal, 38(2), 303–27.
Jones, B. D. (1994). Re-conceiving Decision-Making in Democratic Politics. Chicago,
Jones, B. D. (2001). Politics and the Architecture of Choice. Chicago, IL: University
of Chicago Press.
Jones, B. D., & Baumgartner, F. R. (2005). The Politics of Attention. Chicago,
Jones, B. D., Baumgartner, F. R., Brouard, S., Pedersen, C. G., & Walgrave, S. (Eds.)
(2011). Special Issue on the Comparative Dynamics of Policy Change.
Comparative Political Studies, 44(8).
Jones, B. D., Baumgartner, F. R., & True, J. L. (1998). Policy Punctuations: U.S.
Budget Authority, 1947–1995. The Journal of Politics, 60(1), 1–33.
Jones, B. D. (Ed.) (2012). Special Issue on Punctuated Equilibrium. The Policy
Jones, B. D., Baumgartner, F. R., Breunig, C., Wlezien, C., Soroka, S., Foucault, M.,
François, A., Green-Pedersen, C., John, P., Koski, C., Mortensen, P. B., Varone,
F., & Walgrave, S. (2009). A General Empirical Law of Public Budgets: A
Comparative Analysis. American Journal of Political Science, 53(4), 855–73.
Jones, B. D. and Breunig, C. (2007). Noah and Joseph Effects in Government
Budgets: Analyzing Long-Term Memory. Policy Studies Journal, 35(3), 329–48.
Jones, M. D., & Jenkins-Smith, H. (2009). Trans-subsystem Dynamics: Policy Topog-
raphy, Mass Opinion, and Policy Change. The Policy Studies Journal, 37(1), 37–58.
Katzmann, R. A. (1989). The American Legislative Process as a Signal. Journal of
Kingdon, J. W. (1981). Congressmen’s Voting Decisions (2nd ed.). New York: Harper
and Row.
Koski, C. (2015). Does a Partnership Need Partners? Assessing Partnerships for

Critical Infrastructure Protection. American Review of Public Administration,
45(3), 327–42.
Krehbiel, K. (1992). Information and Legislative Organization. Ann Arbor: University
of Michigan Press.
LaPira, T. M. (2014). Lobbying after 9/11: Policy Regime Emergence and Interest
Group Mobilization. The Policy Studies Journal, 42(2), 226–51.
Lewis, D. E. (2003). Presidents and the Politics of Agency Design. Palo Alto, CA:
Stanford University Press.
Lodge, M., & Matus, K. (2014). Science, Badgers, Politics: Advocacy Coalitions
and Policy Change in Bovine Tuberculosis Policy in Britain. The Policy Studies
Journal, 42(3), 367–90.
Lowi, T. J. (1969). The End of Liberalism: The Second Republic of the United States.
New York: W.W. Norton.
May, P. J., Jochim, A. E., & Pump, B. (2013). Political Limits to the Processing of
Policy Problems. Politics and Governance, 1(2), 104–16.
May, P. J., Jochim, A. E., & Sapotichne, J. (2011). Constructing Homeland Security:
An Anemic Policy Regime. The Policy Studies Journal, 39(2), 285–307.
May, P. J., & Koski, C. (2013). Addressing Public Risks: Extreme Events and Critical
Infrastructures. Review of Policy Research, 30(2), 139–59.
May, P. J., Sapotichne, J., & Workman, S. (2009a). Widespread Policy Disruption:
Terrorism Public Risks, and Homeland Security. Policy Studies Journal, 37(2),
171–94.
May, P. J., Sapotichne, J., & Workman, S. (2009b). Widespread Policy Disruption
and Interest Mobilization. Policy Studies Journal, 37(4), 793–815.
May, P. J., & Workman, S. (2009). The Paradox of Agency Issue Attention: The
Bush Administration and Homeland Security. In C. Provost and P. Teske (Eds.),
President George W. Bush’s Influence Over Bureaucracy and Policy: Extraordinary
Times, Extraordinary Powers. New York: Palgrave MacMillan.
May, P. J., Workman, S., & Jones, B. D. (2008). Organizing Attention: Responses
of the Bureaucracy to Agenda Disruption. Journal of Public Administration
Research and Theory, 18(4), 517–41.
McCool, D. J. (1998). The Subsystem Family of Concepts: A Critique and a
Proposal. Political Research Quarterly, 51(2), 551–70.
McCubbins, M. D. (1985). The Legislative Design of Regulatory Structure.
McCubbins, M. D., Noll, R., & Weingast, B. R. (1987). Administrative Procedures
as Instruments of Control. Journal of Law, Economics and Organization, 3(2),
243–77.
Miller, L. L. (2007). Rethinking Bureaucrats in the Policy Process: Criminal Justice
Agents and the National Crime Agenda. The Policy Studies Journal, 32(4),
569–88.
Mitnick, B. M. (1975). The Theory of Agency: The Policing Paradox and Regulatory
Behavior. Public Choice, 24(1), 27–42.
Moe, T. M. (1985). Control and Feedback in Economic Regulation: The Case of
the NLRB. American Political Science Review, 79(4), 1094–116.
Newell, A., & Simon, H. A. (1972). Human Problem Solving. Englewood Cliffs, NJ:
Prentice-Hall.
Nohrstedt, D., & Weible, C. M. (2010). The Logic of Policy Change after Crisis:
Proximity and Subsystem Interaction. Risks, Hazards, and Crisis in Public Policy,
1(2), 1–32.
Patty, J. W. (2009). The Politics of Biased Information. The Journal of Politics, 71(2),
385–97.
Rosenbloom, D. H. (2001). Whose Bureaucracy Is This, Anyway? Congress’ 1946
Answer. PS: Political Science and Politics, 34(4), 773–7.
Shafran, J. S. (2014). A Whirlpool of Information: Information Processing in Policy
Subsystems, 1995–2010. Ph.D. Dissertation. University of Texas, Austin.
Shepsle, K. A. (1978). The Giant Jigsaw Puzzle. Chicago, IL: University of Chicago
Press.
Shepsle, K. A., & Weingast, B. R. (1987). The Institutional Foundations of
Committee Power. American Political Science Review, 81(1), 86–108.
Simon, H. A. (1979). Models of Thought. New Haven, CT: Yale University Press.
Simon, H. A. (1983). Reason in Human Affairs. Palo Alto, CA: Stanford University
Press.
Stigler, G. J. (1961). The Economics of Information. Journal of Political Economy,
69(3), 213–25.
Stigler, G. J. (1971). The Economic Theory of Regulation. The Bell Journal of
Economics and Management Science, 2(1), 3–21.
Thurber, J. (1996). Political Power and Policy Subsystems in American Politics. In
B. G. Peters and Bert A. Rockman (Eds.), Agenda for Excellence: Administering the
State. Chatham, NJ: Chatham House.
Weible, C. M. (2008). Expert-Based Information and Policy Subsystems: A Review
and Synthesis. Policy Studies Journal, 36(4), 615–35.
Weible, C. M., & Sabatier, P. A. (2009). Coalitions, Science, and Belief Change:
Comparing Adversarial and Collaborative Policy Subsystems. Policy Studies
Journal, 37(2), 195–212.
Whitford, A. B. (2005). The Pursuit of Political Control by Multiple Principals.
Williams, W. (1998). Honest Numbers and Democracy: Policy Analysis in the White
House, Congress, and the Federal Agencies. Washington, DC: Georgetown
University Press.
Workman, S. (2014). Congressional Bureaucracy and the Administrative
Presidency: Policy Analysis as a Result of Institutional Conflict. Paper presented
at the Annual Meeting of the Western Political Science Association. Seattle, WA.
Workman, S. (2015). The Dynamics of Bureaucracy in the U.S. Government: How
Congress and Federal Agencies Process Information and Solve Problems. New York:
Workman, S., Jones, B. D., & Jochim, A. E. (2009). Information Processing and
Policy Dynamics. Policy Studies Journal, 37(1), 75–92.
Yackee, S. W. (2006). Sweet-Talking the Fourth Branch: The Influence of Interest
Group Comments on Federal Agency Rulemaking. Journal of Public Administration
Yackee, S. W. (2012). The Politics of Ex Parte Lobbying: Pre-proposal Agenda
Building and Blocking during Agency Rulemaking. Journal of Public Administration
13
How and Why Do Policy Paradigms
Change; and Does It Matter?
The Case of UK Energy Policy
Introduction
It has long been argued that materialist explanations (i.e. focusing

exclusively on interests) of policy-making and institutional change are
limited and that concepts developed within the ‘new’ institutional-
ism may provide some extra explanatory depth (Blyth, 2002; Fischer,
2003; Widmaier, Blyth, & Seabrooke, 2007). The ‘new’ institutionalism,
formed in the 1980s and 1990s as a response to rational choice and
behaviouralism, sought to ‘bring the state back in’ to the explanation of
political action (Hall & Taylor, 1996; Peters, 2005). Sociological and con-
structivist variants of new institutionalism have built upon Hall’s (1993)
seminal work on policy paradigms in order to provide explanations of
the role ideas play in policy and institutional change (e.g. see Béland &
Cox, 2013; Carson, Burns, & Calvo, 2009; Daigneault, 2013; Kern, 2011;
Kuzemko, 2013; Menahem, 2008; Murphy, 2012; Niemelä & Saarinen,
2012). These analyses proceed from the observation that frameworks of
ideas colour not only how a policy problem is understood, but also policy
choices and institutional structures. Likewise, frameworks of ideas can
also impact heavily upon processes of institutional change, often under-
stood as happening in response to crises. Ideas and their expression in
the form of narratives are understood as being capable of convincing
groups within society that there is a crisis, that existing policy is failing
to solve the crisis, and that alternative solutions should be pursued.
These recent analyses are particularly adept at explaining change, but
potentially at the cost of conceptualising what exactly constitutes a pol-
icy paradigm shift or how we can measure whether or not one has taken
place (also see Daigneault, 2013). We agree with Daigneault (2013) when
he argues that the concept of policy paradigm has been very influential
269
over the last two decades, but that it is also important to further develop
the concept and its operationalisation for empirical research. Both Hall
(1993) and Oliver and Pemberton (2004) offer methods of measuring
profound change. Hall (1993) suggests that a policy paradigm shift has
taken place only once the objectives and instruments of policy have
been replaced by new ones, and Oliver and Pemberton (2004) add that
these new institutions need then to become embedded for a true shift
to have occurred.
The aim of this chapter is to build a framework for analysing paradigm
change by drawing on the above work whilst adding insights into how
and why change takes place from more constructivist variants of insti-
tutionalism. It thereby contributes to conceptualisations of crisis and
change within sociological and constructivist new institutionalism and
hopes to contribute to debates about the usefulness (or otherwise) of the
concept of policy paradigms in explaining policy change. Empirically,
the chapter contributes to recent debates about whether or not a para-
digm shift has occurred in UK energy policy.
Energy policy in the UK is an interesting field in which to study policy
paradigms for three reasons. First, policy paradigms are particularly influ-
ential in areas that are considered to be highly technical and require a
body of specialist knowledge, such as energy policy (Hall, 1993, p. 291).
Second, identifying significant changes to the UK energy policy para-
digm is important given that it has long been held up as a ‘model’ for
other countries (IEA, 2007). Lastly, there is a current (unresolved) debate
about whether or not a policy paradigm shift is taking place within
energy policy in the UK (see Kuzemko, 2013; Mitchell, 2008; Rutledge,
2010), and beyond (Goldthau, 2012; Helm, 2005). Within this debate it
is also often noted that the evolving governance system is both highly
complex and difficult to understand, not least in terms of what it might
achieve (Keay, 2012; cf. Rutledge, 2010).
The chapter proceeds in four sections. The next section develops the
analytical framework and the following section applies the framework
to UK energy policy developments from 2000 to 2011 to test its analyti-
cal purchase.1 The research draws on a systematic analysis of policy doc-
uments, including White Papers and strategy documents, and secondary
literature. This data is complemented by 28 semi-structured interviews
with key stakeholders involved in UK energy policy-making over the
last decade.2 We agree with Daigneault (2013) that triangulation across
documentary sources and directly looking at the ideas policy-makers
hold (e.g. through elite interviews) is an important part of the analysis
(instead of using a ‘revealed ideas’ strategy).
Florian Kern, Caroline Kuzemko, and Catherine Mitchell 271
The third section argues that the suggested framework proved useful
for analysing whether or not a policy paradigm change has occurred but
also identifies a limitation of the framework: it has little to say about
the impacts and outcomes of institutional changes on behaviour within
energy systems. The chapter concludes by suggesting that the policy
paradigm literature should pay more attention to the outcomes of policy
change, for example, in our case to the impacts of institutional changes
on how the energy system operates.
How and why do policy paradigms change?
Hall’s (1993) seminal work on policy paradigms offers the starting point
for our analysis. He identifies a policy paradigm as a framework of ideas
which influences the way in which policy is formulated in a given policy
area. The framework of ideas colours and constrains the ways in which
problems are perceived, decisions about appropriate policy goals and
which instruments are considered to be most acceptable in attaining
these goals (Hall, 1993, pp. 278–9). As such Hall claims that policy-
making can be understood as a process that involves an active and ongo-
ing inter-relationship between the interpretive framework of ideas and
levels of policy in the form of goals and instruments (1993, p. 278).
Hall refers to policy paradigms as sometimes ‘taken for granted’ or ‘una-
menable’ to scrutiny (1993, p. 279) and other new institutionalists have
built upon these observations when analysing ways in which interpretive
frameworks can be embedded within institutions. Governance institu-
tions, such as departments, reflect and embody the interpretive frame-
work as well as limit the impact of alternative frameworks of ideas on
policy (Hay & Wincott, 1998; Schmidt & Radaelli, 2004; Yee, 1996). As
such, the ways in which formal institutions are set up and maintained and
the mandates to which they work can be understood as important aspects
of how the interpretive framework influences policy choices (Kuzemko,
2013, pp. 48–9). Menahem suggests that it is important to distinguish
between ideas carried by actors, and ideas which have been ‘embedded
in and fortified by established institutional arrangements’ (2008, p. 501).
To take better account of these ways in which ideas become embed-
ded, policy paradigms are conceptualised here to include governance
institutions, in addition to Hall’s (1993) three levels outlined above.
Thus, a policy paradigm consists of four inter-related levels: (1) ideas
about the subject and how it should be governed (interpretive frame-
work), (2) policy objectives, (3) policy instruments and (4) governance
institutions. By disaggregating the policy paradigm in this way we can
measure a shift as having taken place if significant alterations can be

identified, which depart from existing practices, on each level of the pol-
icy paradigm between two time periods, that is, in this instance between
2000 and 2011. Such a shift would contrast with instances in which
problems arise but small, ad hoc adjustments are made to only one or
two levels (Hall, 1993; Oliver & Pemberton, 2004). This method also
requires a conceptualisation of the policy paradigm both at the start of
the period of measurement and at the end.
What is also required, however, is an analysis of processes of and
drivers for change. Hall’s (1993) original definition of policy paradigm
change suggested that it could take place through the method of social
learning, but it had less to say about conditions under which institu-
tional change takes place. Recent research within sociological variants
of new institutionalism has concentrated on explaining processes of
change, some of which emphasises the evolutionary aspects of profound
change (Mahoney & Thelen, 2010; Marsh, 1999). Others, however, have
suggested profound change as a more discontinuous event and have
emphasised the role of crisis in punctuating policy evolution (Blyth,
2002; Hay, 2001; Widmaier et al., 2007). It is at such times that it can
be more convincingly argued that the existing policy paradigm is not
providing acceptable outcomes, thereby allowing its credibility to be
challenged and an alternative policy paradigm to be accepted and insti-
tutionalised (Blyth, 2002; Hay, 1996; Oliver & Pemberton, 2004). Crises
have been interpreted, therefore, as moments of breakdown but also of
possible political agency in the form of profound policy change (Hay,
1996), or at least as creating conditions which enhance the possibility of
challenge (Wilson, 2001, p. 262).
Crises are, however, understood not as self-apparent phenomena, but
events that need to be narrated and explained as constituting a prob-
lem, which requires action, in order for attention to be focused on the
policy area in question (Blyth, 2002; Widmaier et al., 2007). Narratives
in this way take on the important role, within the processes of change,
of establishing a shared perception that a crisis exists across a range of
actors. In addition to establishing material events as ‘a crisis’, the domi-
nant crisis narrative must also offer important, credible explanations
as to why events constitute a crisis as well as directly related solutions
(Blyth, 2002; Hay, 2001; Stone, 1989). This function is highly significant
in that for paradigms to be superseded they must not only be perceived
to be obsolete, but there must also be a credible, alternative interpre-
tive framework and related policies to replace it (Hay, 2001; Oliver &
Pemberton, 2004).
All this suggests, however, a quite linear process: events occur, they are
widely perceived as constituting ‘a crisis’, existing policy is understood
as incapable of addressing these problems, and a new policy paradigm
replaces the old. This also suggests that one narrative must dominate the
process both of establishing crisis and the new policy paradigm. Oliver
and Pemberton (2004), however, suggest a messier and less linear pro-
cess of change that emphasises the battle between a range of different,
competing crisis narratives. Albeit they also imply that, ultimately, the
new policy paradigm will be established based on the arguments, ideas
and solutions embodied within one alternative crisis narrative.
These sociological analyses of institutional change have had a great
deal, therefore, to say about conditions under which change occurs
and about the central role of crisis narratives in enabling change.
Sociological analyses have, however, offered little precise definition of
what needs to have occurred in order that a policy paradigm shift can be
claimed. There are references to ‘rejection’ and ‘replacement’ of the old
paradigm, but little direct measurement to show what has changed. By
constructing a framework of analysis that both measures and explains
change this chapter hopes to provide a more rigorous assessment of the
process of change.
A case of policy paradigm change? UK energy policy

between 2000 and 2011
The empirical analysis is divided into three, time divided, subsections.

Following observations above about the need to conceptualise the
energy policy paradigm at the start of the period of analysis, the first
subsection delineates UK energy policy as heavily influenced by pro-
market ideas, in line with other characterisations of UK energy policy
at this time (Mitchell, 2008; Rutledge, 2007; Scrase, Wang, MacKerron,
McGowan, & Sorrell, 2009). It does so by providing a detailed descrip-
tion of each of the four levels of this pro-market energy policy paradigm
(PEPP) (cf. Kuzemko, 2013). This section also introduces the climate
change narrative as the principal challenger of the PEPP but shows how,
despite some policy changes, pro-market ideas managed to remain dom-
inant. The second subsection introduces a further, cumulative challenge
to the dominant paradigm in the form of a geopolitically informed nar-
rative about energy (in)security. The last subsection analyses the ways
in which climate change and geopolitical narratives came together to
allow actors to demonstrate the failure of the existing policy paradigm
and to enable deeper policy changes.
The pro-market energy policy paradigm and the

climate challenge: 2000–2003
In terms of the first level of the PEPP, the interpretive framework guiding
policy-making, energy was understood as a ‘normal’ tradable commod-
ity and the market was seen as the most efficient vehicle for energy
supply (Lawson, 1982). The role of the state was simply to create and
then maintain, through regulation, a level playing field open to compet-
itive forces. Decisions about investment, fuel sources and fuel mix were
left to market players. This interpretive framework fitted well within
the overall approach of less state involvement in the economy that had
dominated elite UK circles since the 1980s (Rutledge, 2007).
If we look at the four levels of the PEPP we can see that the principal
objective of energy policy was to establish and maintain a competitive,
freely trading energy market, and this would ensure other important
outcomes such as energy security and affordability (Kuzemko, 2013;
Rutledge, 2007). This is not to say that security and affordability objec-
tives did not exist, just that they were understood to be natural out-
comes of freely trading, competitive markets (DTI, 2000, p. 7). There
were ambitions that 10% of electricity should be supplied by renew-
able sources by 2010 but these had not been formalised as objectives of
policy (Mitchell & Connor, 2004, p. 1937).
Under the PEPP the principal instruments of achieving this objective
had been centred initially around the long process of privatising and
deregulating the sector, and later around the construction of a new
regulatory framework which would effectively ‘steer towards a defined
general direction . . . [but] leave it to the market to select the means
to reach that end’ (Mitchell, 2008, p. 1). Electricity and gas compa-
nies were allowed to vertically integrate, to set their own prices and to
build their businesses according to a formula of maximising volumes
and supply.
Lastly, in terms of the fourth level, governance institutions, energy
policy was relegated to a subdivision of the Department of Trade and
Industry (DTI) after the Department of Energy was disbanded in 1992.
The DTI should maintain the regulatory framework, but responsibil-
ity for implementation rested with the independent regulator, Ofgem
(Office of the Gas and Electricity Markets). Formal mandates for both
these organisations were centred on maintaining competitive markets
and ensuring fair treatment for consumers. For a summary of the PEEP,
see Table 13.1.
Table 13.1 The pro-market energy policy paradigm in 2000
Level Description
Interpretive framework s energy as tradable commodity

s markets as most efficient vehicle for energy
trade and supply
s government should not supply energy, nor
decide energy mix
s energy to be traded and supplied in an
economically efficient manner through
competitive, freely trading markets
s free markets understood as delivering energy
security
Objectives of policy s the provision of secure, diverse, and
sustainable supplies of energy at competitive
prices as an outcome of freely trading,
competitive markets
Instruments s regulatory framework designed to enhance
ability of markets to supply energy at lowest
cost
s electricity pricing according to British
Electricity Trading and Transmission
Arrangements (BETTA) bilateral trading and
networks according to RPI-Xa
s Renewables Obligation being developed to
support renewable energy
Governance institutions s Department of Energy disbanded in 1992
s responsibility for policy at subdivision of the
DTI
s Ofgem key player: regulator to oversee
markets to ensure low prices for consumers
a
Retail Price Index excluding mortgage interest payments (RPI-X).
As a result, private energy companies, in particular the ‘big six’ gas

and electricity companies, became central to the delivery of important
energy service to the UK public. This inferred a particular set of power
relations between the non-interfering state and the private sector in
energy whereby large market players had a high degree of influence
in policy-making circles. Energy companies were incentivised, further-
more, to be cost efficient, to utilise assets, such as oil and gas from the
North Sea, at a high rate, but not to think about or act in terms of the
sustainability of the UK’s national energy system.
However, from the late 1990s onwards the PEPP was being challenged
by a coalition of actors who argued not only that a long-term global cli-
mate change crisis existed, but that government needed to take a more
hands-on approach in energy policy-making in order to avert this crisis.
They offered a range of solutions and specific ways in which govern-
ment could become more involved in establishing an environmentally
sustainable energy system (Greenpeace, 2006; RCEP, 2000; Scrase &
MacKerron, 2009). For example, in 2002 a full Energy Review was carried
out, by the Performance and Innovation Unit (PIU), which represented
a direct challenge to the PEPP. This was not least in that it suggested that
there should be new and specific climate change and renewable energy
objectives for energy policy. It also pointed out that a system of ‘trade-
offs’ should be established whereby carbon reduction objectives would
trump others: ‘Energy policy trade-offs affecting the period to 2012
should generally give priority to carbon reduction if there is a mate-
rial risk of failing to meet internationally-agreed emissions targets’
(PIU, 2002, p. 52).
However, these challenges resulted in limited changes to the four lev-
els of the PEPP. The trade-offs recommended in the PIU were by-passed
in the new 2003 Energy White Paper. In addition, although for the first
time energy policy was set towards achieving a formal climate change
objective it was phrased in a rather non-specific way: ‘to put ourselves on
a path to cut the UK’s carbon dioxide emissions . . . by some 60% from
current levels by about 2050’ (DTI, 2003, p. 11).3 As such, this new objec-
tive appeared vague, more like an ‘aim’ than a firm commitment, and
was not taken as fixed within the energy division of the DTI and Ofgem
(Interviews 5 and 15). In terms of governance institutions, while the
Energy Review had proposed a new government department responsible
for climate and energy policy-making functions (PIU, 2002, p. 144), the
White Paper overtly rejected changes in government institutions in that
they wanted ‘to concentrate . . . energies on following through the com-
mitments we have made, not on creating new machinery’ (DTI, 2003,
p. 112). The interpretive framework showed the least amount of change
in that policy-makers continued to believe in the ability of competitive,
liberalised markets to meet policy objectives, including reducing carbon
emissions, energy affordability and energy security (cf. Interviews 1, 2,
3 and 4; DTI, 2003, pp. 11, 15). The 2003 White Paper also included an
overtly benign interpretation of the international context for energy,
despite acknowledging the UK’s imminent reversal from net exporter to
importer of oil and gas as well as sharply growing demand from India
and China.
There was some change in policy instruments, but nothing that rep-
resented a particular shift in the overall pro-market orientation. For
example, a new Renewable Obligation (RO) was introduced in 2001 that
placed an obligation on electricity suppliers to source a certain percent-
age of their electricity from renewable energy (Mitchell & Connor, 2004).
It was explicitly designed so that the government only specifies the tar-
get, but the obligation certificates are tradable and companies choose
the cheapest technologies to achieve the target. The RO was therefore
well aligned with the pro-market paradigm and a far cry from ‘risk free’
systems of support for renewables in countries like Germany and Spain.
As such, although climate change narratives had mounted a signifi-
cant challenge, by arguing that a sustainable energy crisis existed and
that the PEPP needed to alter to address this problem, little change
ensued to any of the four levels. The PEPP showed quite high degrees
of path dependency whilst also claiming to respond to climate change
narratives. As will be seen below these arguments did, however, persist
and ad hoc changes to the paradigm had already started to undermine
the intellectual coherence of the PEPP.
The security of supply crisis: 2004–2007
Various events had already started to focus some attention on energy

security, such as the truckers’ strike of 2000 in Britain and the many
major network failures around the turn of the Millennium, for example,
in New Zealand, Italy and the United States. In direct contrast to benign
perceptions of the international energy context referenced in the 2003
White Paper, a new challenge for the PEPP emerged around the mid-
2000s when a geopolitically informed narrative about security of energy
supply gained political prominence.
By the end of 2006 political perceptions about energy supplies in the
UK and across Europe started to shift significantly in the wake of grow-
ing state involvement in the Russian energy sector, the Russo-Ukrainian
gas transit dispute in 2008 and rising energy prices (Barton, Redgwell,
Ronne, & Zilman, 2004; Light, 2006). Fears about Russia’s ability to
affect energy trade and prices fuelled a great degree of political interest
and fears about being dependent on unstable foreign suppliers mounted
(Fox, 2006; House of Commons, 2007b). Newspapers were replete with
terminology evocative of the Western Cold War mentality suggesting
that Russia was fast becoming an ‘energy superpower’ and reminding
importer nations of their hydrocarbon ‘dependency’ status (see, for
example, Ostrovsky, 2006; Robinson, 2006). Such ideas and interests
were reflected in academia and a new Journal of Energy Security was estab-
lished in 2008.4
The timing could not have been worse given the UK’s shifting import–
export position (Blair in DTI, 2006a). The ensuing crisis debate was simi-
lar in tone and scale to UK oil crises debates of the 1970s and marked
the start of a significant repoliticisation of energy (DTI, 2007), akin to
the ‘highly politicised and public debate’ that is claimed to accompany
paradigm shifts (Hay, 2001, p. 200). We argue that it was the nature of
this narrative, and the publically perceived threat to UK national energy
security, that resulted in renewed political interest in and deliberation
of energy and the problems it faced (see Kuzemko, 2014). It appears
as if the near-term nature of this threat to UK energy security evoked a
higher degree of political interest than the climate narrative of long-
term global crisis.
Refocused political attention resulted in a range of new energy policy
documents (see in particular DTI, 2006a, 2006b, 2006c, 2007; JESS,
2006). In 2006 another Energy Review was conducted which now
referred to energy security as being one of two ‘immense’ challenges fac-
ing energy policy, and in the same year Tony Blair used his annual Lord
Mayor’s speech to highlight energy security concerns (DTI, 2006a, p. 4).
One interviewee within Ofgem suggested that direct political pressure
was being brought to bear at this time to ‘do something’ about energy
security (Interview 15). A narrative of UK indigenous, or ‘home grown’,
energy started to emerge.
In terms of discernible change in the four levels of the UK energy
policy paradigm, what stands out most at this stage is the genuine shift
in objectives. While creating competitive markets had been the primary
goal of energy policy up until this point, ensuring energy security started
to supersede this objective (DTI, 2006a, p. 4). Policy instruments were
also showing signs of change, particularly in that more decisions about
the energy mix were being made – albeit somewhat covertly. There was
a significant refocus on facilitating production of domestic supplies
of energy, including nuclear, coal and oil and gas, to avoid imports
and maintain a level of energy independence (DTI, 2006a; House of
Commons, 2007b).
An interesting juxtaposition emerges here. Although geopolitically
informed narratives had come to dominate energy debates, and had
resulted in greater political attention to energy, understandings of supply
security were not openly understood to challenge ideas about markets
and competition. Blame for supply insecurity was placed instead on the
unstable foreign supplies, not on the structures of the PEPP. Therefore,
at this stage an important element of the process of policy paradigm

shift identified above, a challenge of the interpretive framework through
mounting evidence of policy failure, was missing. There were also only
very limited changes in governance institutions. Some procedures were put
in place that committed DTI and Ofgem to regular reports to Parliament
on energy security (HM Government, 2004) and the 2007 White Paper
announced for the first time ‘an integrated international energy strategy
which describes the action we are taking to help deliver secure energy
supplies and tackle climate change’ (DTI, 2007, p. 8).
During this period geopolitically informed energy security narratives
managed to establish a widespread perception that an energy crisis
existed, using evocative language to focus public and political attention.
This narrative had both repoliticised energy as well as generated a high
degree of political response outside of energy governance institutions.
Given the limited changes to the PEPP, however, it cannot be claimed
that this narrative had, on its own, resulted in a policy paradigm change.
Accumulation of changes: 2008–2011
This section observes that climate narratives, given the ongoing repoliti-
cisation of energy, managed to provide the evidence of failure of exist-
ing policy missing in geopolitical security narratives. A newly emerging
energy security–climate narrative, combining elements of geopolitical
and climate narratives, became effective in providing impetus for a
policy paradigm change in a way that neither of the two narratives had
managed alone.
By the late 2000s, results in terms of reducing carbon dioxide emis-
sions were deteriorating, particularly as the ‘easy gains’ from the switch
in electricity production from coal to gas in the 1990s were past (Carbon
Trust, 2006; Greenpeace, 2006). Climate analysts and advocacy groups
were providing empirical evidence that energy policy was not delivering
on carbon dioxide reductions and renewable energy deployment (House
of Commons, 2007a; World Wildlife Fund, 2006). These results were
problematic given overt claims made in the early 2000s, in response to
climate challenges to the PEPP that markets would deliver. Several high
profile reports from credible institutions started to suggest that the UK
government should play a more active role in developing and deploy-
ing low carbon technologies (IEA, 2007; Stern, 2006). Academics also
increasingly argued for more government leadership and investment
in research, development, demonstration and deployment of new
energy technologies (e.g. Foxon, Pearson, Makuch, & Mata, 2005;
Sauter & Watson, 2007). Importantly, pressure increased after the adop-
tion of the EU 20-20-20 targets in 2009 (Directive, 2009/28/EC) which
meant that the UK was now committed to sourcing 15% of all energy
from renewables by 2020.
Climate groups, think tanks, and some academics started to actively
utilise fears about dependency on ‘unstable’ foreign suppliers and
renewed interest in energy independence to argue for a refocus on
increasing UK domestic energy production (Bird, 2007; Giddens, 2009;
Interview 18). One example is a report for Greenpeace entitled ‘Oil and
Peace Don’t Mix’ which overtly used geopolitical ideas about unstable
suppliers, conflict and the need to increase independence to argue for
greater state commitment to domestic renewable energy (Greenpeace,
2006). The argument here is that some climate groups strategically
changed their narrative because they understood aspects of the geopolit-
ical narrative to be capable of evoking political reaction (Interview 18).
Given mounting evidence of policy failure, claims repeatedly made
by the DTI and Ofgem that markets and competition would deliver on
renewable energy and energy efficiency were increasingly looking less
credible. Tendencies to rely on market instruments could be more cred-
ibly identified as part of the problem (Scrase et al., 2009, p. 6). As such
the PEPP became increasingly vulnerable to challenge and political con-
testation. Actors brought together climate change and geopolitical nar-
ratives to argue that a nationally relevant crisis does exist, that UK policy
needs to change, and provided potential solutions in terms of a more
active role for state intervention.
By the late 2000s government was starting to more actively look for
ways to address these mounting pressures. The Climate Change Act of
2008 was one of the first outcomes of this rethink of energy policy (HM
Government, 2008a). This Act was held up as being the first of its kind
in the world in that it set legally binding carbon dioxide reduction tar-
gets up until 2050, of at least 80% (Lockwood, 2013; Watson, 2009). It
was understood that ‘[t]urning to renewables will help the UK recover
some of its energy self-sufficiency’ (DECC, 2009b, p. 10), and as such
domestically produced renewables became an answer to both climate
change and energy security goals. This marked a clear departure from
the PEPP objective of 2000.
The UK government had also started to make some substantial
changes to its governance institutions, not least the establishment of the
Department for Energy and Climate Change (DECC) in 2008. The setup
of DECC reflects the understanding ‘that climate change and energy
policies are inextricably linked’ (DECC, 2011a). This change is also
significant in that it both promoted energy backup to Cabinet Level

as well as mandating DECC to achieve the specific climate and energy
objectives. Whilst Ofgem’s duty to contribute to the achievement of
sustainable development had already been introduced in 2004, the
2008 Energy Act promoted this duty placing it on an equal footing with
other objectives (HM Government, 2008b). A new ‘Office for Renewable
Energy Deployment’ (ORED) was established within DECC to ensure
that the UK would deliver on its new renewable targets (DECC, 2009b,
p. 9). The independent Committee on Climate Change (CCC) was also
established to advise on and monitor progress towards achieving a lower
carbon economy (House of Commons, 2007a, p. 3).
The challenge to the PEPP from the energy security–climate narra-
tive also led to changes in the interpretive framework of policy-makers.
The 2009 report on UK energy security regularly referenced the need for
‘home grown’, renewable energy in order to provide for both energy and
climate security (Wicks, 2009). This not only shows the energy security–
climate narrative’s infiltration into government documents, but it like-
wise emphasises the role that renewables were now to play. Another
clear change in the interpretive framework was the emerging stance on
state–market roles in energy (cf. Miliband, 2008; Wicks, 2009). The new
understanding of energy policy did not wholesale reject the previous
paradigm in that it still recognised the many abilities of ‘the market’
to supply energy. But, it did suggest a ‘strategic role for government’
in response to market failures and that a national policy was needed to
provide incentives for the public good (Miliband, 2008, p. 4). Malcolm
Wicks’ report on energy security suggested that ‘the era of heavy reli-
ance on companies, competition and liberalisation must be re-assessed’
(2009, p. 1). What also emerged at this stage was the recognition of
energy’s fundamental socio-economic role for society, as opposed to
notions of energy as replaceable commodity popular under the PEPP
(DECC, 2009c).
These changes manifested themselves in a number of policy docu-
ments and new policy instruments. There emerged over the course of a
few years a considerable upsurge in policy documents – White Papers –
and legislation (BIS and DECC, 2009; DECC, 2009a, 2009b, 2010; HM
Government, 2008a, 2008b). Many of these documents, including ‘The
UK Low Carbon Transition Plan’ and the ‘Renewable Energy Strategy’,
overtly recognised the need for policy change (DECC, 2009a, 2009b),
albeit in the latter case, encouraged by Europe. The Renewable Energy
Strategy ‘put in place mechanisms to provide financial support for
renewable electricity and heat worth around £30bn between now and
2020’ (DECC, 2009b, p. 8). DECC also announced their intention to

directly fund four large-scale carbon capture and storage demonstra-
tion plants, to channel about £3.2bn to help households become more
energy efficient, to roll out smart meters in every home by the end of
2020 and to provide further state investment in offshore wind (DECC,
2009a, p. 4). The RO was amended such that it was ‘banded’ (provid-
ing differing levels of support for different technologies) and could,
therefore, no longer be understood as a technology-neutral instrument.
Another significant change in terms of instruments was the introduc-
tion of a ‘feed-in-tariff’ (FiT) in April 2010 aimed at incentivising small-
scale renewable energy production (DECC, 2009b, p. 8).
On top of these new instruments, the government also later embarked
on a major revision of the energy market through the electricity mar-
ket reform package. The draft Energy Bill of May 2012 included sugges-
tions for a number of mechanisms, including contracts for differences,
capacity payments, emission performance standard, carbon floor price,
to incentivise investment in low carbon electricity generation. Much of
this new legislation represented a shift away from the previous reliance
on market-based instruments for achieving energy policy objectives
(DECC, 2012; Foxon, 2012; Pearson & Watson, 2012).
Given the accumulation of changes witnessed in this last period it
can be observed, using the framework applied here, that each level of
the PEPP had shifted from its position in 2000. These changes are sum-
marised in Table 13.2 below and together suggest that a policy paradigm
shift in UK energy policy has taken place.
Discussion
It is worth at this point briefly outlining the key insights gained from the
application of this framework, as well as any gaps remaining. There are
two principle points to be made here. Firstly, our observations about the
combined influence of two different narratives, climate change and geo-
political energy security, within the process of paradigm change mark
a contradiction with previous conceptualisations of paradigm change
being facilitated by one alternative set of ideas. It has also allowed us to
identify the new energy governance system not as one coherent, alter-
native policy paradigm but as based upon multiple perspectives. This
is most evident in the new objectives of energy policy and in the new
governance institutions established. Although the interpretive framework
has altered to include alternative narratives, elements of belief in mar-
ket ideas continue to persist. Examples of this are the continued support
Table 13.2 The energy policy paradigm in 2000 and 2011
Level 2000 2011
Interpretive s energy as tradable commodity s energy is understood to have a central socio-economic role to
framework s markets as most efficient vehicle for energy play, rather than understood as a normal commodity
trade and supply s markets to supply energy but within tighter government
s government should not supply energy, nor decide specification
energy mix s market failures in energy and climate change require a relative
s energy to be traded and supplied in an economi- change in the role of the state
cally efficient manner through competitive, freely s more energy should be ‘home grown’ and should come from
trading markets clean sources
s free markets understood as delivering energy security
Objectives of s the provision of secure, diverse and sustainable s energy security, including affordability, one of two primary
policy supplies of energy at competitive prices as an objectives
outcome of freely trading, competitive markets s climate change goals now legally binding through Climate Change
Act (and specific to include precise levels of required reductions)
s increasing share of renewables now formal objective of policy
s affordability somewhat sidelined
Instruments s regulatory framework designed to enhance ability s variety of instruments put in place to facilitate more domestic
of markets to supply energy at lowest cost energy production
s electricity pricing according to BETTA bilateral s FIT introduced for small-scale production of renewable electricity
trading and networks according to RPI-X s introduction and banding of Renewables Obligation
s Renewables Obligation being developed to support s Electricity Market Reform undertaken
renewable energy
Governance s Department of Energy disbanded in 1992 s Creation of DECC with specific energy and climate change
institutions s responsibility for policy at subdivision of mandates
Department of Trade and Industry (DTI) s other new institutions include the CCC and the Office for
s Ofgem key player: regulator to oversee markets to Renewable Energy Deployment

ensure low prices for consumers s new energy and climate division within FCO
s Ofgem mandate introduced to include sustainability
283
for the EU ETS scheme to help drive emission reductions in the power
sector and the central role for private actors in delivering UK energy
supplies and services. Pro-market ideas about fiscal austerity and limited
state intervention are also arguably influential over the Treasury’s more
recent attempts to limit further funding for renewable energy.
By identifying that the new system has ‘picked and mixed’ between
narratives on energy, and how it should be governed, this chapter has
been able to explain why the current governance system is so complex
and difficult to understand (Keay, 2012). However, this finding also
questions the very definition of the policy paradigm concept as internal
coherence is suggested as a necessary condition for the existence of a
paradigm (e.g. Daigneault, 2013). Further conceptual work and empiri-
cal research is necessary to establish what ‘a certain level of internal
coherence’ means and how this can be operationalised.
The mixed nature of the new paradigm also means that assumptions
are being made about the compatibility of policy goals, for example,
that energy security, affordability and climate objectives will be met
simultaneously using a mix of policy instruments. This assumption does
not recognise, nor overtly address, the question of potential trade-offs
between objectives nor does it take account of the different perspec-
tives to which they relate. The relative importance of different policy
objectives appears to be contested between the Coalition government
partners and across different departments (esp. DECC and Treasury). For
example, energy poverty objectives appear to be under pressure given
decisions to pass most costs of low carbon energy on to consumers.
The impact on affordability, a subject of growing political relevance for
the next general election in 2015, is likely to be considerable – already
energy poverty numbers are escalating despite the objective of eradicat-
ing it by 2016 (DECC, 2011b).
The suggested analytical framework has, therefore, proved useful for
analysing whether or not a policy paradigm change has occurred and
for revealing potential tensions within the current policy paradigm, but
there is one key limitation of the framework. While emphasis has been
placed on the policy goals, instruments, interpretive frameworks and
governance institutions, the analysis has had less to say about whether
or not the institutional changes actually impact upon how the UK
energy system operates. This raises questions about whether the new
paradigm will be able to deliver on the ambitious policy goals, or achiev-
ing a transition towards the low carbon economy, or whether there will
have to an even more substantive shift in order to do so. For exam-
ple, while all of the changes observed here together amount to a policy
paradigm change, it can also be argued that during the period of analysis
relatively little changed in terms of the characteristics of the energy sys-
tem. Government ambitions that 10% of electricity should be supplied
by renewable sources by 2010 were missed, with only 6.5% of electricity
produced from renewable sources in that year. Results elsewhere indicate
that the UK is not on track to achieve the 15% renewable energy target
for 2020. The principal actors providing gas and electricity services have
also changed little. For example, only 3% of domestic electricity supply
in 2013 was provided from companies other than the big six utilities,
who continue to dominate gas and electricity markets. This is of con-
cern because recent socio-technical analyses of energy transitions argue
that system change, such as the one required to move to a secure and
low carbon future, is often driven by new entrants and innovators rather
than incumbent industry actors (Loorbach, 2007; Rotmans, Kemp, &
Van Asselt, 2001; Rotmans & Loorbach, 2008). Arguably the continued
ability of the big six utilities to influence policy based on their dominant
market position and their key role in implementing government policy
has important implications for achieving the objectives of UK energy
policy. As such, the framework applied in this chapter has told us rela-
tively little about the ways in which policy and institutional changes
impact important outcomes on the ground. Without including wider
impacts on the energy system in the analysis, it is difficult to assess
whether the new ideas and new policies are sufficient to meet objectives.
Policy paradigm change: Does it matter?
Based on the analysis presented above, it is evident that the PEPP has
changed on every level outlined in our theoretical framework. This is
clearly a significant finding in itself, but it is in terms of understanding
how and why change has taken place that the subtleties of the process
come to light. Although in the 2000–2003 period, climate change narra-
tives offered a range of direct challenges to the PEPP, these were success-
fully compromised away with claims that the existing paradigm could
deliver a low carbon future. Events took a different turn in the mid-2000s
when geopolitically informed arguments about security of supply prob-
lems managed to open up energy policy to a crisis debate across pub-
lic, media, academic and political circles. Ultimately, however, it was in
combination that climate change and geopolitical narratives were most
powerful in influencing policy paradigm change. Rather than just one
brief crisis moment, therefore, the analysis shows a weaving together
of narratives about several problematic features of the UK energy policy
paradigm over a period of ten years. As such, change took place in response
to arguments about multiple energy-associated crises that accumulated
over time to become credible, but not in the moment of one crisis. This
is of some relevance for the conceptual debate about whether paradigm
shifts occur only in moments of crises or whether they can also be more
gradual, emerging processes (see Daigneault, this volume).
This finding also runs contrary to assumptions within sociological
institutionalism whereby one (coherent) narrative comes to dominate
interpretations of crisis and then eventually replaces the incumbent
policy paradigm. The identification of more than one narrative as influ-
ential in change provides explanatory detail not only of a messier and
more contingent process of change (cf. Oliver & Pemberton, 2004), but
also about the structure of the new paradigm. Given that two sets of
ideas drove the process of change, but that some market ideas and struc-
tures persisted, the new paradigm contains a variety of internal tensions
which pose questions about the concept of paradigms as a set of coher-
ent ideas (see Wilder, this volume).
The chapter also identified a key shortcoming of the analytical frame-
work for scholars who are not just interested in processes of policy and
institutional change, but also in policy outcomes and practice change.
Whilst being adept at explaining policy paradigm change, the suggested
framework sheds little light on whether or not changes have achieved
much ‘on the ground’. In terms of taking this analysis forward we
propose that one way of remedying this situation would be to extend
the analytical framework to include a fifth level which would include
changes in system actors and their behaviours. The emerging literature
on socio-technical transitions might serve as a useful complement to
this analysis of policy paradigm change. This literature identifies policy
and institutions as important within wider systemic change processes
(Kern, 2011, 2012), but pays more attention to the characteristics of the
socio-technical system itself: the actors involved, the technologies they
use, the user demands and the physical infrastructures (Geels, 2002).
In combining insights from socio-technical literatures we would seek
to maintain the in-depth insights into political complexities gleaned
from institutional analysis whilst also including other characteristics
of the socio-technical system as influential within processes of change.
Without such an extension, the policy paradigm approach has little to
offer policy analysts interested in the outcomes of policy.
In summary, the chapter makes three contributions to the discus-
sion about policy paradigms: First, it developed a framework for both
measuring and explaining paradigm change building on historical,
sociological and constructivist variants of new institutionalism. Second,

it challenged the assumption within sociological institutionalism that
one narrative comes to dominate interpretations of crisis and eventually
replaces the incumbent paradigm. Third, it points to limitations of the
paradigm framework in terms of neglecting ‘changes on the ground’ and
suggests that the literature on socio-technical transitions might help to
address this issue.
Appendix: Interviews Conducted

Interview 1: Department for Business, Enterprise and Regulatory Reform
(BERR), Energy Strategy and International Unit, January 2008
Interview 2: BERR, Strategic Analysis Unit, December 2008
Interview 3: Foreign and Commonwealth Office (FCO), Analyst, January
2008
Interview 4: FCO, Analyst, August 2010
Interview 5: Department of Energy and Climate Change (DECC),
International Energy team, September 2010
Interview 6: FCO Moscow, Second Secretary, September 2008
Interview 7: Former Energy Advisor to President Putin, September 2008
Interview 8: Wintershall, Moscow, Head of Representation, September 2008
Interview 9: Deloitte, Moscow, Managing Partner, September 2008
Interview 10: Standard Chartered, Moscow, Managing Partner, September
2008
Interview 11: CERA, Founder and Consultant, December 2007
Interview 12: OXERA, Principal, August 2010
Interview 13: Member of 2002 PIU Energy Review team, September 2010
Interview 14: Member of 2002 PIU Energy Review team, February 2011
Interview 15: Ofgem, January 2011
Interview 16: DECC, International Energy Security Review team, January 2011
Interview 17: Qatar National Oil and Gas, Head of International Marketing,
December 2009
Interview 18: Worldwatch, Director, Energy and Climate Program, May 2011
Interview 19: FCO, former analyst, August 2011
Interview 20: Former Head of Policy Planning in 10 Downing Street and
Senior Policy Adviser
Interview 21: Senior civil servant at BERR, February 2008
Interview 22: Senior civil servant at DEFRA, February 2008
Interview 23: Former member of Advisory Committee on Business and the
Environment (ACBE), January 2008
Interview 24: Lecturer in renewable energy policy, February 2008
Interview 25: Former senior manager Carbon Trust, March 2008
Interview 26: Former senior civil servant at DETR/DEFRA, March 2008
Interview 27: Senior civil servant at BERR, February 2008
Interview 28: Senior manager Carbon Trust, February 2008
Notes
1 The analysis starts with the year 2000 and not in 1997, the year that New
Labour came to power since the Utilities Act of 2000 marked the first legal
change in energy policies.
2 For a full list of those interviewed see Appendix.
3 Italics authors’ own.
4 See http://www.ensec.org/
References
Barton, B., Redgwell, C., Ronne, A., & Zilman, D. N. (2004). Energy Security:
Managing Risk in a Dynamic Legal and Regulatory Environment. Oxford: Oxford
University Press.
Béland, D., & Cox, R. H. (2013). The Politics of Policy Paradigms. Governance,
26(2), 193–5.
Bird, J. (2007). Energy Security in the UK. London: Institute for Public Policy
Research.
BIS and DECC. (2009). The UK Low Carbon Industrial Strategy. Department for
Business Innovation and Skills and Department of Energy and Climate Change.
the Twentieth Century. New York: Cambridge University Press.
Carbon Trust. (2006). Policy Frameworks for Renewables: An Analysis of Policy
Frameworks to Drive Future Investment in Near or Long Term Power in the UK.
London: The Carbon Trust.
Carson, M., Burns, T. R., & Calvo, D. (Eds.) (2009). Paradigms in Public Policy:
Theory and Practice of Paradigm Shifts in the EU. Frankfurt: Peter Lang.
21(3), 453–69.
DECC. (2009a). The Low Carbon Transition Plan: National Strategy for Climate and
Energy. London: TSO.
DECC. (2009b). The UK Renewable Energy Strategy. London: HMSO.
DECC. (2009c). Government Response to Malcolm Wicks’s Review of International
Energy Security, ‘Energy Security: A National Challenge in a Changing World’.
London: Department of Energy and Climate Change.
DECC. (2010). The Green Deal. A Summary of the Government’s Proposals. London:
Department of Energy and Climate Change.
DECC. (2011a). About Us. DECC. Retrieved from https://www.gov.uk/
government/organisations/department-of-energy-climate-change/about.
DECC. (2011b). Annual Report on Fuel Poverty Statistics 2011. London: TSO.
DECC. (2012). Draft Energy Bill. London: TSO.
DTI. (2000). The Energy Report (Vol. 4). London: Department of Trade and Industry.
DTI. (2003). Energy White Paper: Our Energy Future – Creating a Low Carbon Economy.
London: HMSO.
DTI. (2006a). The Energy Challenge Energy Review Report 2006. London: TSO.
DTI. (2006b). Secretary of State’s Second Report to Parliament on Security of Gas and
Electricity Supply in Great Britain. London: TSO.
DTI. (2006c). The Effectiveness of Current Gas Security of Supply Arrangements:

A Consultation. London: The National Archives.
DTI. (2007). Meeting the Energy Challenge: An Energy White Paper. London: TSO.
Fischer, F. (2003). Reframing Public Policy. Discursive Politics and Deliberative
Practices. New York: Oxford University Press.
Fox, L. (2006). Over a Barrel: The Challenge of Defense and Energy Security. The
Westminster Energy Forum. Retrieved from http://www.westminsterenergy.
org/Over_a_Barrel_-_the_Challenge_of_Defense_and_Energy_Security.pdf.
Foxon, T. J. (2012). Transition Pathways for a UK Low Carbon Electricity Future.
Energy Policy, 52(1), 10–24.
Foxon, T., Pearson, P., Makuch, Z., & Mata, M. (2005). Transforming Policy Processes
to Promote Sustainable Innovation: Some Guiding Principles. A Report for Policy
Makers. London: Imperial College.
Geels, F. W. (2002). Technological Transitions as Evolutionary Reconfiguration
Processes: A Multi-level Perspective and a Case-Study. Research Policy, 31(8–9),
1257–74.
Giddens, A. (2009). The Politics of Climate Change. Cambridge: Polity Press.
Goldthau, A. (2012). From the State to the Market and Back: Policy Implications
of Changing Energy Paradigms. Global Policy, 3(2), 198–210.
Greenpeace. (2006). Oil and Peace Don’t Mix. London: GreenPeace.
Hall, P. A., & Taylor, R. C. R. (1996). Political Science and the Three New
Institutionalisms. Political Studies, 44(5), 936–57.
Hay, C. (1996). Narrating Crisis: The Discursive Construction of the Winter of
Discontent. Sociology, 30(2), 253–77.
Britain: An Ideational Institutionalist Approach. In J. L. Campbell &
O. K. Pedersen (Eds.), The Rise of Neoliberalism and Institutional Analysis
(pp. 193–218). Princeton: Princeton University Press.
Helm, D. (2005). The Assessment: The New Energy Paradigm. Oxford Review of
Economic Policy, 21(1), 1–18.
HM Government. (2004). Energy Act 2004. London: The National Archives.
HM Government. (2008a). The 2008 Climate Change Act. London: HMSO.
HM Government. (2008b). The Energy Act. London: HMSO.
House of Commons. (2007a). Beyond Stern: From the Climate Change Programme
Review to the Draft Climate Change Bill, a Report by the House of Commons
Environmental Audit Committee. London: TSO.
House of Commons. (2007b). Energy Security: Research Paper 07/42. London:
House of Commons Library.
IEA. (2007). Energy Policies of IEA Countries. The United Kingdom 2006 Review. Paris:
International Energy Agency.
JESS (Joint Energy Security of Supply Working Group). (2006). Long-Term Security
of Energy Supply: A DTI Report. London: DTI.
Keay, M. (2012). Death by a Thousand Regulations: The New Energy Bill. Paper pre-
sented at the Oxford Energy Comment, The Oxford Institute for Energy Studies,
Oxford, UK.
Kern, F. (2011). Ideas, Institutions, and Interests: Explaining Policy Divergence in

Fostering ‘System Innovations’ towards Sustainability. Environment and Planning
C: Government and Policy, 29(6), 1117–34.
Kern, F. (2012). Using the Multi-level Perspective on Socio-technical Transitions
to Assess Innovation Policy. Technological Forecasting and Social Change, 79(2),
298–310.
Kuzemko, C. (2013). The Energy Security-Climate Nexus: Institutional Change in the
UK and Beyond. Basingstoke and New York: Palgrave Macmillan.
Kuzemko, C. (2014). Politicising UK energy: What Speaking Energy Security Can
Do. Policy & Politics, 42(2), 259–74.
Lawson, N. (1982). Speech delivered at the Fourth Annual International Conference,
International Association of Energy Economists, Churchill College, Cambridge,
June 28.
Light, M. (2006). Russia’s Political Engagement with the EU. In R. Allison,
M. Light, & A. White (Eds.), Putin’s Russia and the Enlarged Europe (pp. 77–88).
Oxford: Blackwell.
Lockwood, M. (2013). The Political Sustainability of Climate Policy: The Case of
the UK Climate Change Act. Global Environmental Change, 23(5), 1339–48.
Loorbach, D. (2007). Transition Management. New Mode of Governance for
Sustainable Development. Unpublished doctoral dissertation, Erasmus
University Rotterdam, The Netherlands.
Mahoney, J., & Thelen, K. (2010). Explaining Institutional Change: Ambiguity,
Agency, and Power. Cambridge: Cambridge University Press.
Marsh, D. (1999). Introduction: Explaining Change in the Postwar Period. In
D. Marsh, J. Butler, C. Hay, J. Johnston, P. Kerr, S. McAnulla, & M. Watson
(Eds.), Postwar British Politics in Perspective. Cambridge: Policy Press.
Menahem, G. (2008). The Transformation of Higher Education in Israel since
the 1990s: The Role of Ideas and Policy Paradigms. Governance, 21(4),
499–526.
Miliband, E. (2008). The Rise and Fall and Rise Again of a Department of Energy.
Speech given at Imperial College, London, UK, December 9.
Mitchell, C. (2008). The Political Economy of Sustainable Energy. Basingstoke:
Palgrave Macmillan.
Mitchell, C., & Connor, P. (2004). Renewable Energy Policy in the UK 1990–2003.
Energy Policy, 32(17), 1935–47.
Murphy, M. (2012). Interests, Institutions and Ideas: Explaining Irish Social
Security Policy. Policy & Politics, 40(3), 347–65.
Niemelä, M., & Saarinen, A. (2012). The Role of Ideas and Institutional Change in
Finnish Public Sector Reform. Policy and Politics, 40(2), 171–91.
Ostrovsky, A. (2006, April 21). Oil and Gas Bring Influence. Financial Times, p. 5.
Pearson, P., & Watson, J. (2012). UK Energy Policy 1980–2010. A History and Lessons
to Be Learnt. London: The Institution of Engineering and Technology and
Parliamentary Group for Energy Studies.
Peters, B. G. (2005). Institutional Theory in Political Science: The ‘New
Institutionalism’. London and New York: Continuum.
PIU (Performance and Innovation Unit). (2002). The Energy Review. A Performance
and Innovation Unit Report. London: Cabinet Office.
RCEP. (2000). Energy – The Changing Climate. 22nd Report. London: Royal
Commission on Environmental Pollution.
Robinson, A. (2006, July 22). Russia’s Pipeline Politics. Prospect.
Rotmans, J., Kemp, R., & Van Asselt, M. (2001). More Evolution Than Revolution:
Transition Management in Public Policy. Foresight, 3(1), 1–17.
Rotmans, J., & Loorbach, D. (2008). Transition Management: Reflexive
Governance of Societal Complexity through Searching, Learning and
Experimenting. In J. Van den Bergh & F. R. Bruinsma (Eds.), The Transition to
Renewable Energy: Theory and Practice (pp. 15–46). Cheltenham: Edward Elgar.
Rutledge, I. (2010). UK Energy Policy and Market Fundamentalism: A Historical
Overview. In I. Rutledge & P. Wright (Eds.), UK Energy Policy and the End of
Market Fundamentalism (pp. 1–38). Oxford: Oxford Institute for Energy Studies.
Rutledge, I. (2007). New Labour, Energy Policy and ‘Competitive Markets’.
Cambridge Journal of Economics, 31(6), 901–25.
Sauter, R., & Watson, J. (2007). Strategies for the Deployment of Micro-generation:
Implications for Social Acceptance. Energy Policy, 35(5), 2770–9.
Schmidt, V. A., & Radaelli, C. M. (2004). Policy Change and Discourse in Europe:
Conceptual and Methodological Issues. West European Politics, 27(2), 183–210.
Scrase, I., & MacKerron, G. (2009). Energy for the Future. A New Agenda. Basingstoke
and New York: Palgrave Macmillan.
Scrase, I., Wang, T., MacKerron, G., McGowan, F., & Sorrell, S. (2009). Introduction:
Climate Policy. In I. Scrase & G. MacKerron (Eds.), Energy for the Future:
A New Agenda (pp. 3–19). Basingstoke: Palgrave Macmillan.
Stern, N. (2006). The Economics of Climate Change. London: Cambridge University
Press.
Stone, D. A. (1989). Causal Stories and the Formation of Policy Agendas. Political
Watson, J. (2009). Technology Assessment and Innovation Policy. In I. Scrase &
G. MacKerron (Eds.), Energy for the Future: A New Agenda (pp. 123–46).
Basingstoke: Palgrave Macmillan.
Wicks, M. (2009). Energy Security: A National Challenge in a Changing World.
London: Department of Energy and Climate Change.
Widmaier, W., Blyth, M., & Seabrooke, L. (2007). Exogenous Shocks or Endogenous
Constructions? The Meanings of Wars and Crises. International Studies Quarterly,
51(4), 747–59.
Wilson, C. A. (2001). Policy Regimes and Policy Change. Journal of Public Policy,
20(03), 247–74.
World Wildlife Fund. (2006). WWF-UK Response to the Energy Review. Retreived from
http://www.wwf.org.uk/research_centre/research_centre_results.cfm?
uNewsID=2445
Yee, A. S. (1996). The Causal Effects of Ideas on Policy. International Organisation,
50(1), 69–108.
Conclusion
14
Bringing Ideational Power into
the Paradigm Approach: Critical
Perspectives on Policy Paradigms
in Theory and Practice
Martin B. Carstensen
Introduction
The work of Peter A. Hall (1993) on the role of policy paradigms in

public policy is remarkable. Unrivalled by any other publication in idea-
tional scholarship in terms of citations and impact, it – together with
the work of scholars like Sabatier and Jenkins-Smith (1993), Kingdon
(2003), Baumgartner and Jones (2009), and Stone (1988) – drew on the
social learning-tradition (Heclo, 1974) to help bring ideas on the agenda
of mainstream political science and policy studies (Béland & Cox, 2013;
Blyth, 2013; Daigneault, 2014). With the subsequent important work
of what may be termed ‘second generation ideational scholarship’ (e.g.
Béland, 2007; Berman, 1998; Blyth, 2002; Campbell, 1997; Campbell
& Pedersen, 2001; Cox, 2001; Lieberman, 2002; Schmidt, 2002), idea-
tional frameworks for analyzing policymaking grew into a more or less
coherent approach in its own right, what Schmidt (2008) later dubbed
‘discursive institutionalism.’ Hall’s (1993) theory of policy paradigms
was essential for the coming of age of the ideational approach because
while it placed institutions and interests centrally in its explanatory
framework, ideas were the primary factor in accounting for processes of
stability and change.
So successful was Hall’s (1993) paper that for years it stood largely
unchallenged regarding its ontological basis and the theory of change
it posited. More recently, however, a number of key characteristics of
Hall’s (1993) approach have been subject to critical appraisal. Although
fundamentally sympathetic with Hall’s (1993) aim of placing ideas cen-
trally, ideational scholars have reacted to the structuralist predilection
of a Kuhnian approach to policy paradigms. Hall (1993) has thus been
criticized for overemphasizing punctuated equilibrium-style change
295
(Béland, 2007; Carstensen, 2011a; Cashore & Howlett, 2009; Kay, 2007;
Lieberman, 2002; Orenstein, 2013; Seabrooke, 2009; Wilder & Howlett,
2014), conceptualizing ideas as static and monolithic, and downplay-
ing agency in processes of ideational change (Berman, 2013; Campbell,
2004; Widmaier, Blyth, & Seabrooke, 2007; Wood, 2015; see also Parsons,
2007) by claiming that actors effectively internalize the policy paradigm
they adhere to (Carstensen, 2011b; Jacobs, 2009; Schmidt, 2008).
Taken together, these authors have shown that by basing his theory of
policy paradigms on a crude version of Kuhn’s (1970) notion of science
paradigms, Hall (1993) ends up with an overly systemic understanding
of the role of ideas in policymaking (Schmidt, 2011). This is a some-
what paradoxical outcome, since ideas were in the first place brought
into new institutionalist theory to balance the scale toward a greater
emphasis on explaining processes of change to supplement the char-
acteristically strong focus on stability (Blyth, 1997). Just as he had so
successfully framed the scholarship on ideas and politics, Peter A. Hall
moved on to other debates, so ideational scholars were left with the task
of figuring out how to develop the concept of policy paradigms and
its central message of the importance of ideas in policymaking with-
out succumbing to its overly systemic implications. More than two dec-
ades later it seems safe to claim that the tradition has to a large extent
been successful in this endeavor: ideational approaches to policy studies
have enthusiastically been taken up by a new generation of scholars,
while important advances have been made to bring more agency into
the analysis of the role of ideas in policymaking. Thus, the chapters in
this volume reflect well the current status of the use of paradigm theory
in public policy research. Although varied in theoretical and methodo-
logical approaches, the contributions are joined in their commitment to
placing ideas and policy paradigms centrally in explanations of political
change and stability. Moreover, and importantly, the chapters seek to
push the use of paradigms as an analytical concept toward greater room
for agency, in turn allowing for more gradual forms of change.
This chapter has two overall aims. First, based on the contributions
to this volume, it takes stock of the main debates concerning the role
of paradigms in public policy research. Specifically, it argues that three
dimensions of Hall’s (1993) policy paradigm framework are increasingly
recognized as problematic for understanding dynamics of stability and
change: the incommensurability thesis, the role of policy anomalies in
setting off processes of change, and the focus on punctuated equilib-
rium as the only dynamic of significant policy change. With the increas-
ing recognition of the analytical limitations of these key parts of the
Martin B. Carstensen 297
paradigm approach, it begs the question whether policy studies should

at all invoke the concept of a paradigm in explanations of policy devel-
opments. Starting from a positive response to this question, the second
part of the chapter suggests a potential way forward in dealing with
these issues, namely to develop a clearer understanding of how varying
forms of power play into processes of paradigm change and stability.
In short, it argues that instead of conceptualizing paradigms as enjoy-
ing monopoly over the minds of political actors, seeing paradigm shifts
as originating in diminishing problem-solving abilities, and in the end
understanding paradigmatic contests as decided through electoral poli-
tics, more should be done to understand the varying roles of power in
explaining processes of stability and change in paradigms. Specifically, it
suggests that ideational power can be conceptualized as a form of power
in its own right, which should be connected with other forms of power
like institutional and structural power (Carstensen & Schmidt, 2014).
Given limits on space, this argument necessarily remains a sketch, but
the chapter nonetheless defends the claim that developing a clearer
notion of ideational power offers a promising path for furthering our
theoretical and empirical understanding of paradigms in policymaking.
Taking stock of Hall’s paradigm approach
The starting point for Peter A. Hall’s paradigm approach to understand-

ing dynamics of policy stability and change was the question: what
motivates state action? Hall’s (1993) seminal paper was thus published
in the slipstream of a broader effort in American political science to
bring the concept of the state back in. Although the work of this group
of scholars was more diverse than the common allegiance to ‘bringing
the state back in’ indicates, Hall’s (1993) approach fit well with broader
neo-statist efforts to balance between statist and pluralist approaches, as
well as combining ideas and institutions (see Adcock, Bevir, & Stimson,
2007, p. 274). This context seems to have mattered for how ideas were
conceived by Hall (1993). As noted by Parsons (2007), institutional and
ideational types of explanations are both characterized by starting from
ambiguity in the objective environment:
Both then see people themselves creating the constraints that resolve
the ambiguity and channel their action along a certain path. One big
difference is that institutionalists only see ambiguity in environmen-
tal conditions at the starting point of their arguments, while a-rational
ideational claims see ambiguity throughout. For institutionalists,
their situation external to the individual eventually clarifies thanks

to the unintended consequences of earlier institution-building . . .
In a-rational ideational claims it is the actor’s interpretation of the
situation, not the situation itself, which ultimately indicates a way
forward. (pp. 98–9)
Writing at a time when the rational choice offensive was at its height
(Blyth, 2013, p. 212), Hall (1993) found a way to straddle institutional
and ideational explanatory logics, although it might be claimed that he
placed too much emphasis on the resolution of ambiguity rather than
the continual need for interpretation. The solution was to transpose
Thomas Kuhn’s (1970) theory of scientific revolutions into a theory of
policymaking, which proved useful for the effort to develop an insti-
tutional understanding of the role of ideas in politics. Despite Kuhn’s
(1970) recognition that his theory was ill suited for understanding social
science – and thus even less useful for studying society (see Schmidt,
2011) – Hall (1993) enthusiastically applied key parts of Kuhn’s frame-
work in his analysis of the role of paradigms in policy development,
namely the incommensurability thesis, the importance of anomalies,
and the division between ‘normal science’ and paradigm shifts. As dis-
cussed below, this analytical framework – although useful for building a
strong argument for the role of ideas in policymaking – has increasingly
been subject to criticism in ideational scholarship.
The incommensurability thesis

With the argument that policy paradigms are incommensurable –
‘Because each paradigm contains its own account of how the world facing
policymakers operates’ (Hall, 1993, p. 280) – Hall took a strong position
on both the internal purity of a paradigm and the mutual exclusivity
of paradigms. That is, paradigms were believed to be internally coherent,
composed of logically connected elements that originated in fundamen-
tally different accounts of how an economy works. On both counts, the
approach has been subject to criticism. One of the first critiques came
from Surel (2000, p. 499), who suggested that instead of seeing a para-
digm as a unifying and homogenizing social sphere, a policy paradigm
is best understood as a bounded space for conflict, which ‘marks out the
terrain for social exchange and disagreements, rather than simply sup-
porting an unlikely consensus.’ An important implication is that a dom-
inant paradigm ‘by no means is an exclusive one’ (Surel, 2000, p. 502).
This broadly aligns with Oliver and Pemberton’s (2004) re-evaluation
of Hall’s (1993) case of British economic policy. Here they show that
rather than exhibiting the incommensurability posited by Hall, the

paradigm shifts that occurred were less than wholesale with new ideas
being incorporated into the prevailing policy paradigm. That is, para-
digm replacement need not be absolute (see also the contributions from
Wilder and Wilder & Howlett, this volume). This happened during the
1930s with the Treasury’s partial integration of Keynesian instruments
into the prevailing neoclassical framework, as well as in the 1960s and
late 1970s where no paradigmatic shift occurred but a ‘subset of the new
ideas was incorporated into the prevailing framework of policy, which
thereby underwent a marked evolution’ (p. 436). In sum, Oliver and
Pemberton’s (2004) comprehensive analysis of the British case showed
clear signs that the strong version of the incommensurability thesis
did not hold even in a specialized and theoretically heavy policy area
like economic policymaking, and that paradigms may often be joined
in synthesis (see also Wilder & Howlett, this volume). Schmidt (2011,
p. 42; 2002) voices a similar argument, when she points out that in the
social sciences and society there rarely is only one predominant para-
digm ‘since there are ordinarily other minority (opposition) programs
waiting in the wing,’ perhaps made up of once dormant ideas that may
be resurrected when new events call for new explanations. In addition,
Schmidt (2011) argues that policy paradigms are not as coherent as the
Kuhnian approach leads us to expect, since they are the result of con-
flicts as well as compromises among actors who bring different ideas to
the table.
Variations exist as to how forcefully the notion of ‘pure’ policy para-
digms is rejected. Thus, while Schmidt (2011) suggests we dispense with
the use of the incommensurability thesis altogether, Daigneault (2014;
this volume) instead suggests a loosening of the concept. Although
Daigneault claims that elements from one paradigm can be understood
by people who subscribe to another paradigm, and that often ideas are
simultaneously compatible with different policy paradigms and policies,
he maintains that in order to speak of a paradigm, a high degree of inter-
nal coherence between the ideas of the paradigm must be observed. By
this he means, ‘the ideational content of all dimensions is compatible
and logically consistent.’ According to Daigneault (2013), in the absence
of this condition, paradigms do not exist and are only a loose collection
of ideas (for a similar view, see Princen & t’Hart, 2014). But, how do we
determine whether a paradigm is logically consistent? And is this really
what matters in the policy process?
Here I would suggest instead taking a political and historical
approach. That is, what makes a paradigm coherent is whether policy
actors conceive them as coherent – whether they can forcefully make an

argument for its problem-solving abilities – and the historically founded
embeddedness of the paradigm in the culture of the polity, that is, its
resonance with established values (Schmidt, 2002, see also Campbell &
Pedersen, 2014). This interpretation is consonant with Wilder’s (2015;
this volume) argument that paradigmatic incommensurability is rela-
tive. That is, notwithstanding the particular logical coherence of a
paradigm, what matters is whether policy actors perceive paradigms as
incommensurable, not that they are logically coherent and thus in any
‘objective’ sense fundamentally different in their ideational structure.
Importantly, the perception of incommensurability is subject to change
through deliberation and strategic framing on part of policy actors.
What matters for Wilder (2015) are the positive correlations individual
actors make between lower and higher order components of a given
policy (this volume). According to Wilder (2015), as policy ideas that
are part of paradigm move from the agenda setting stage to formulation
on to implementation, the paradigmatic purity of policy ideas is often
compromised as incommensurability among alternatives is eroded.
This line of argument emphasizes the politics of paradigms and ideas –
something Hall (1993) has been criticized for neglecting (e.g. Campbell,
2002; Wood, 2015) – in that policy actors’ coupling of sometimes dis-
parate policy programs during agenda setting and the marginal adjust-
ments made to policies when actors confront unexpected problems is
theorized to contribute to the erosion of paradigmatic purity. In other
words, from the perspective of policy scholars, a paradigm might appear
as logically incoherent, but what matters is the perception of policy
actors that the paradigm is coherent enough for them to subscribe to
it, and whether the ideas of the paradigm are shared by an entire policy
community (Baumgartner, 2013, p. 251; see also Carstensen, 2011b;
Wilder & Howlett, 2014).
Attesting to a growing consensus about the heterogeneous nature of
policy paradigms, a number of chapters in this volume present different
versions of the argument that policy paradigms often contain conflict-
ing ideational elements, but also, importantly, that this might turn out
to be a political strength. In their chapter to this volume, Cairney and
Weible, for example, suggest that we think of paradigms as system-wide,
which means that all participants operate within a common discursive
framework, while subsystems may work with competing frames. In a
similar vein, the chapter by Kern, Kuzemko, and Mitchell argues that
in the case of energy policy, two different narratives, one on climate
change and another on geopolitical energy, were combined leading
the authors to conclude that instead of exhibiting coherent unity, the

new energy governance system was structured by multiple ideational
perspectives.
In his critique of the paradigm approach, Zittoun draws inspiration
from the work of Michel Foucault to make the argument that policy
paradigms are most often made up of heterogeneous ideational ele-
ments ordered on a macro level by an episteme. What matters for the
political strength of this assemblage is not its logical coherence, but if it
makes sense to its promoters and is able to garner the necessary support.
Although he prefers to use the concept of a policy statement to refer
to a discursive assemblage of different heterogenous components with-
out any presupposed coherence that gives specific meaning to a policy
proposal, Zittoun’s conceptualization is not that different from the bur-
geoning consensus on the composite structure of policy paradigms.
Instead of seeing a lack of logical coherence in a paradigm as a
weakness – or as indicating that we are not really witnessing a para-
digm in the first place – it is worth recognizing that the malleability of
a paradigm also has important political strengths in dealing with chal-
lenges that arise through political battles and the rise of policy anoma-
lies. As argued by Mondou, Skogstad, and Houle (2014), the resilience of
a paradigm is connected to the capacity of its policy images to respond
effectively to criticism from competing policy images, which may entail
‘rejecting or fending off attacks, but can also entail integrating external
criticism into the understanding of the policy issue’ (p. 159). Schmidt
and Thatcher (2013) similarly point to the mutability of paradigms as
a strength in encompassing evolving and sometimes conflicting ideas
leading to conceptual expansion, rather than retreat (see also Schmidt,
2006, p. 251). And Béland and Cox (2014) have recently coined the term
‘coalition magnet’ to refer to the capacity of ambiguous or polysemic
ideas to appeal to a diversity of individuals and groups, and to be used
strategically by policy entrepreneurs (i.e. individual or collective actors
who promote certain policy solutions) to frame interests, mobilize sup-
porters, and build coalitions. The lack of ideational purity that results
from both necessary political compromises, coalition building, defense
of the reigning paradigm, and the movement from the agenda-setting
stage to the more concrete policies related to the institutionalization of a
paradigm, suggests that strategic ‘ideational bricolage’ is more pervasive
in both theory and practice than is often recognized (Wilder, 2015, p. 2).
In turn, the incommensurability thesis seems increasingly regarded as
unnecessary for claiming that policy paradigms structure the policymak-
ing process.
The role of policy anomalies

Another central argument in Hall’s (1993) original framework was the
important role of experimentation and policy anomalies in spurring
processes of paradigmatic change in policymaking. Thus, as the support-
ers of a paradigm encounter developments that contradict the basic ten-
ets of the paradigm – that is, anomalies – ad hoc attempts will be made
to stretch the terms of the paradigm. In the end, Hall (1993) argued, this
is likely to undermine the authority of policymakers and experts that
adhere to the faltering paradigm, paving the way for a shift toward a
competing paradigm.
What is particularly interesting – and, in my view, problematic –
about this argument is that it juggles two competing ontologies. On one
hand, what Blyth (2013) calls Bayesian social learning – taking place in
instances of 1st and 2nd order change – that views ideational change as
occurring following information updates and experiments, and, on the
other, a constructivist approach that understands information updates
and experiments as fundamentally in need of interpretation, a process
that can never be objective or unmediated and remains fundamentally
structured by contests over authority and power. According to Blyth
(2013), this tension in Hall’s (1993) approach does not develop into
a paradox, though, because of the central position of the concept of
‘authority’ in his account. That is so, because
Bayesian learning does not ‘hollow out’ the existing paradigm, like
acid eating away at cavity. Rather, events identified as anomalies can
either add to or subtract from the ‘authority’ of those arguing for
political power. Agency, via contests of authority is the key move in
this part piece, but it is a move that the article fails to sustain consist-
ently throughout. (Blyth, 2013, p. 200)
Thus, according to Blyth (2013), the tension between the two ontologies
is not a problem for the approach, but it is something that is not dealt
with systematically in the empirical account of British economic policy-
making. Thus, the move from a traditional social learning perspective to
a Kuhnian approach supposedly leads to ‘autonomization’ of the para-
digm from the incremental processes of Bayesian learning. However, for
Hall (1993) emphasis remains on the ability of the paradigm to solve
problems, not only in the empirical part of the paper – as argued by
Blyth (2013) – but also in the central role of anomalies in accounting for
change. That is, the question of how anomalies play into the competi-
tion for the authority of a paradigm remains open. In Blyth’s rendition
of Hall’s argument, the question is not about whether the paradigm is

‘genuinely incapable of dealing with anomalous developments’ (Hall,
1993, p. 280, my emphasis), but rather a matter of the capacity of actors
to authoritatively interpret an event as anomalous. But, this begs the
question: how do agents gain such authority? Or, how is such authority
exercised?
Except for noting that ‘the policy community will engage in a con-
test for authority over the issues at hand’ (Hall, 1993, p. 280), Hall is
conspicuously silent on this matter. This, however, is not very surpris-
ing given his simultaneous adherence to a Bayesian understanding of
learning. In other words, this is a paradox that does not only play out in
the empirical section of the paper, but a confusion that permeates Hall’s
(1993) account of how paradigms develop and change, and it amounts
to an unwarranted stretching of Hall’s use of the concept of anomalies
to claim that what is really meant is that it is interpretations of devel-
opments as constituting anomalies that undermine the paradigm, not
anomalies as ‘developments that are not fully comprehensible, even as
puzzles, within the terms of the paradigm’ (Hall, 1993, p. 280). It is
anomalies – and the policy failures that ensue from stretching the terms
of the paradigms through experiments – that ‘gradually undermine
the authority of the existing paradigm and its advocates even further’
(p. 280) – not actors’ capacity to persuade the policy community and the
public that the paradigm has failed.1
What is important, then, is how and when policy actors are able to
successfully frame instances that threaten the authority of the paradigm
as anomalies. In that regard, post-crisis developments in the ideas that
structure financial regulation is interesting. As argued, for example, by
Mügge (2013), in the case of financial regulation, debates continue to
emphasize ideas that are directly borrowed from neoliberal conceptions
of financial markets. But, if the financial crisis did not attest the failure of
the neoliberal efficiency market-paradigm, what then could? Succinctly
put, given that paradigms are fundamentally political in nature, they
cannot be proved wrong. Gaining a more analytically sophisticated under-
standing of how policy problems develop into anomalies thus requires
in the first place a greater appreciation of the interpretive battles behind
such processes. To this end, Wilder and Howlett, in their contribution to
this volume (see also Wilder & Howlett, 2014), point out how
Policy anomalies are informational signals that must be interpreted

by human agents . . . while anomalies contribute to the perception
and definition of policy problems, they are not synonymous with the
problems but rather present images necessary to understand them as

such. (p. 113)
So, ‘while anomalies contribute to the perception and definition of pol-

icy problems, it is important to stress that anomalies are not synony-
mous with problems’ (Wilder & Howlett, 2014, p. 190). How then do
actors interpret policy developments as anomalies? Particularly impor-
tant in that regard is the availability of a credible alternative paradigm
(Baumgartner, 2013; Oliver & Pemberton, 2004). If there is no coher-
ent and credible alternative paradigm to help along the delegitimizing
interpretive process, policy actors are likely to use the cognitive filters
of the existing paradigm. That is, to act despite the uncertainty brought
on by crisis, agents need creativity to ‘act back’ upon their environment
in purposive ways (Blyth, 2010, p. 97), and since they have no standard
solution to turn to, actors have to make up new solutions as they go
along. With no realistic alternative available, they move outward bas-
ing their new solutions on their experience of the world before the cri-
sis hit (Carstensen, 2013). Thus, in contrast to Hall’s (1993) approach,
anomalies may lead to paradigm stretching that restores trust in existing
strategies for action, while simultaneously changing the paradigm in a
significant fashion. In the words of Wilder and Howlett (this volume),
‘anomalous observations may thus allow dominant ideational frames
to be stretched, promoting innovative bricolage which brings solu-
tions previously deemed not worthy of consideration into the ambit of
acceptable ideas.’
Viewed this way, the process of paradigm change need not be divided
between the survival and return of the paradigm (‘normal policymak-
ing’) or a complete overhaul of the paradigm. We might additionally –
and perhaps more likely – expect the construction of and reaction to
crisis to lead to a gradual change inside the paradigm. Moreover, this
process is fundamentally political rather than ‘Bayesian’ in nature, with
policy actors seeking strategic advantage and bending interpretations
of reality to suit their needs (Wood, 2015). To this end, Wilder and
Howlett (2014, p. 194; see also this volume) usefully employ the term
‘gatekeeper’ to designate the policy actors who adapt evidence, either
positively or negatively, to conform to political preferences, and amend
existing solution sets or complement cognitive schemas.
Significant gradual change in paradigms

The larger issue at stake underlying the question of the incommensurabil-
ity of paradigms and the role of policy anomalies is how to conceive the
dynamics of paradigm change. By distinguishing between two options

of change – ‘normal policymaking’ and paradigm shifts – Hall (1993)
employed a strong version of the Kuhnian (1970) take on paradigm shifts.
As mentioned above, the approach allows for only two forms of change:
incremental shifts in the instruments and settings of the paradigm – which
in Hall’s (1993) original version does not entail change in the ideas that
structure the paradigm, and thus does not amount to significant change –
and a wholesale shift in the goals and discourse that structure policymak-
ing through the introduction of a rival policy paradigm. In effect, Hall
(1993) adopted a punctuated equilibrium model of change based on an
assumption about the incommensurability of policy paradigms and the
inability of actors to incorporate elements from other paradigms.
As mentioned above, subsequent studies of Hall’s (1993) case of eco-
nomic policymaking in Britain (Hay, 2001; Oliver & Pemberton, 2004)
have showed that in fact the processes of policy change were not eas-
ily characterized by being revolutionary, but rather as a series of itera-
tive changes that develop rather through ‘punctuated evolution.’ As
noted by Daigneault (2014, p. 465; this volume), a strong conception
of incommensurability precludes gradual transformative paradigm shifts,
which makes it all the more important to develop alternative concep-
tions of paradigms and how they change. Thus, taking up the challenge
of accounting for instances of significant gradual change in paradigms,
recent scholarship has sought to develop a less systemic take on par-
adigms while analytically leaving more room for actors to reinterpret
their ideas and perhaps practice ‘inter-paradigm borrowing’ (Hay,
2011) to answer to the rise of anomalies or pressure from competing
paradigms. One approach suggests viewing paradigms as made up of
relationally structured webs of ideas (Carstensen, 2011a). Viewing para-
digms as composite and historically founded through political battles
and compromises, rather than taken over more or less directly from sci-
ence paradigms (e.g. economics), allows policy actors through processes
of policy bricolage to strategically, reflectively, and gradually adjust their
ideas, even in cases where decision-making authority does not change
hands (Wilder & Howlett, this volume). The change that ensues may
take the form of a new hierarchical ranking of elements that already
exist (Surel, 2000), perhaps with the addition of new ideational ele-
ments from another paradigm (Carstensen, 2011a). In this perspective,
seemingly small changes over time, which appear not to have visible
or immediate impact on the basic characteristics and goals of a policy
paradigm can nonetheless add up to a change in its basic philosophy
(Skogstad & Schmidt, 2011).
The increased focus on significant cumulative change over long peri-

ods of time is also matched by greater awareness among more tradi-
tional historical institutionalists of gradual transformation through
processes of institutional change like layering or conversion (Mahoney &
Thelen, 2010; Streeck & Thelen, 2005). The gradualist strand of histori-
cal institutionalism does not employ an explicitly ideational approach –
which is surprising given their frequent reference to the importance of
interpretation – but as shown by Nohrstedt in his contribution to this vol-
ume, the degree of policy actors’ compliance to the policies emanating from
the paradigm is key for understanding its development over time. Much in
line with recent arguments of historical institutionalists, Nohrstedt empha-
sizes how unforeseen consequences of paradigmatic reform have impacted
on how policy actors in the Swedish public sector have received and acted
upon ideas coming from the New Public Management paradigm.
Similarly, Wilder, in his contribution, points out how advocacy
groups – which play a central role in policy formulation and bring-
ing new ideas on the agenda – are unlikely to have a clear conception
about the appropriate instrument mixes and seldom any ideas about
the more specific settings of those instruments, leaving greater room
for epistemic communities and technocrats to more concretely develop
what the paradigm should mean in practice. This perspective is helpful
in pointing out how the ways in which normative values and technical
expertise impact the policy process may be temporally discrete. Feeney
and Horan, in their chapter, emphasize how the practice of a paradigm –
both the formulation and implementation phase – in important ways,
depends on the development of routines that help guide policy makers.
This process takes on a central role in the actual institutionalization of a
paradigm, where the establishing of routines is iterative with the result
that the paradigm may slowly, but significantly, change shape (see also
Wilder & Howlett, 2014).
Given these developments in the literature, it seems safe to say that
the study of policy paradigms has moved beyond the one-sided focus
on change through punctuated equilibriums. A factor that may further
push this agenda is the recent financial and economic crisis. The finan-
cial crisis of 2007–2009 dealt a severe blow to the ideational and institu-
tional structures of advanced financial capitalism, and its ramifications
continue to ripple through the world economy. With policymakers
scrambling to address these challenges, resulting in institutional and
ideational change in a host of policy areas, one could reasonably expect
theories that use notions of critical junctures and paradigm shifts to
reign strongly among current accounts of change in wake of the crisis.
So far, however, in areas like financial regulation and economic gov-

ernance, scholars are finding signs of significant gradual change rather
than the paradigmatic shifts hypothesized by Hall’s (1993) paradigm
model (see, for example, Crouch, 2011; Helleiner, 2014; see also the
edited volumes by Bermeo & Pontussen, 2012; Moschella & Tsingou,
2013; Schmidt & Thatcher, 2013). Even in the studies that do identify
a paradigm shift – notably Baker (2015) that argues for a paradigm shift
toward macroprudential ideas in financial regulation – it often turns
out upon closer inspection that the change amounts rather to a signifi-
cant add-on to the pre-crisis regulatory paradigm (see also Mügge, 2013).
Although further studies are necessary to reach a clearer understanding
of the dynamics of change following the crisis, studies so far seem to
corroborate the growing consensus that significant gradual change in
paradigms is not only a possibility, but also the most prevalent outcome,
not least in times of crisis (see, for example, Braun, 2015; Carstensen,
2013; Salines, Glöckler, & Truchlewski, 2012).
A way forward: Bringing in ideational power
If we dispense with the Kuhnian components that Hall (1993) first intro-
duced to the paradigm approach, it begs the question: What should we
put in its place? Considering the central position of the assumption of
incommensurability to explain stability and the important role assigned
to policy anomalies in setting off processes of paradigm change, it is
necessary to conceptualize alternative factors that might account for sta-
bility and change in paradigms. Instead of basing the drivers for stability
and change on two competing ontologies – one on Bayesian learning
and another one on battles for authority, respectively – this chapter sug-
gests that focus be shifted to the importance of different forms of power,
ranging from the direct force of compulsory power, to the more indirect –
but no less important – effects characteristic of institutional, structural,
and ideational power. Within the confines of this concluding chapter,
the framework necessarily remains a sketch and in need of further elabo-
ration and development. However, hopefully it is clear enough as to
provide a sense of a potential way forward for the analysis of stability
and change in policy paradigms.
Ideational power versus ideas matter

To some the suggestion to focus more strongly on ideational power
might come as a surprise. After all, was ideational power not brought in
as a relevant concept the moment researchers started taking ‘the power
of ideas’ seriously? Yes and no. Yes, ideational research has been prem-
ised on the ontological argument that ‘ideas matter’ in politics, mean-
ing that policy actors need intersubjective ideas to make their interests
actionable, and so ideas have been understood as ‘powerful’, since they
have effect on how actors interpret their world (Parsons, 2007). On
the other hand, this basic argument that ideas matter – that they are
important – in political processes, has been conflated with the more
specific claim that ideas are connected to relations of power. With most
of the effort going into making the case that ideas matter and should
be included in the mainstream of political science, much less has been
done to conceptualize what is meant by the notion of ideational power
and how it relates to other forms of power, for example, compulsory,
institutional, or structural forms of power (Carstensen & Schmidt, 2014;
a notable exception is Béland, 2010; see also Barnett & Duvall, 2005 on
power in international relations).
It might also be objected that Hall’s (1993) original framework already
includes power as a central analytical component. This is true, but its
focus is too one-sidedly on institutional power, and what is left for idea-
tional power to explain is underspecified. That is, on one hand, it is
evident that the paradigm is ‘powerful’ in the impact it has on policy
actors’ view of the world: it is after all the prism through which policy-
makers view the world and thus instrumental in identifying what should
be considered a problem and what kind of solutions are relevant. On
the other hand, because Hall (1993) injects a strong assumption about
the incommensurability of paradigms – that the merits of each cannot
be determined through comparison or tests, and thus it is impossible
to persuade competing fractions about the merit of one’s paradigm –
the battle over paradigms effectively takes place in the political arena.
A paradigm shift will thus not occur until the proponents of the new
paradigm have secured power to institutionalize the paradigm. In this
perspective, power relates more to gaining access to institutional posi-
tions of authority than to the ideational or discursive power of a para-
digm. That is, the triumph of a new paradigm is secured not through
one fraction persuading others about the legitimacy of its paradigm, but
by the institutional power that actors are able to wield.
How, then, are policy actors to attain these positional advantages?
Hall (1993) is not all too clear on this matter, but it is noted that the
contest over authority
may well spill beyond the boundaries of the state itself into the
broader political arena. It will end only when the supporters of a
new paradigm secure positions of authority over policymaking and

are able to rearrange the organization and standard operating proce-
dures of the policy process so as to institutionalize the new paradigm.
(pp. 280–1)
The formulation ‘may well’ is what creates the confusion. Does the con-
test end in the electoral arena or does it not? Hall (1993) is conspicu-
ously vague on this point. It seems safe to assume, however, that given
the central role of the electoral triumph of Margaret Thatcher in the
empirical part of Hall’s (1993) paper, and Hall’s (1993) statement that
‘The monetarist paradigm ultimately triumphed over its Keynesian rival
by means of partisan electoral competition’ (p. 288), what makes and
breaks paradigmatic authority is institutional power secured through
electoral contest. There is little doubt that who holds government power
can be important for which paradigm structures the policymaking pro-
cess, and that electoral success matters for a change in paradigm, but it
is also clear that it is not the only factor of importance in understanding
the dynamics of paradigms in policymaking – a point also recognized by
Hall (1993), but with little further specification.
Conceptualizing ideational power2

To develop an account of policy paradigms that more explicitly deals
with the role of ideational and non-ideational power in processes of sta-
bility and change in paradigms, it is necessary first to provide a definition
of ideational power. Thus, following Carstensen and Schmidt (2014), we
may think of ideational power as the capacity of actors (whether indi-
vidual or collective) to influence other actors’ normative and cognitive
beliefs through the use of ideational elements. This may occur directly
through persuasion or imposition or indirectly by influencing the idea-
tional context that defines the range of possibilities of others. In this
perspective, acts of ideational power – whether successful or not – only
occur in a subset of the relations relevant for understanding how ideas
matter, namely when actors seek to influence the beliefs of others by
promoting their own ideas at the expense of others.
We may further specify three forms of ideational power that are rel-
evant for understanding developments in policy paradigms. First, power
may be exerted through ideas, that is, as a capacity of actors to persuade
other actors to accept and adopt their views of what to think and do
through the use of ideational elements. Persuasion is clearly central
to this form of ideational power. Rather than viewing power as mak-
ing someone do what they would otherwise not have done based on
force, threats, institutional position, material resources, and so on, the

ideational power actors exert is based on their capacity to induce other
actors to do something through reasoning or argument. This form of
power relates both to the efforts of elites to convince each other about
the pertinence of certain programmatic ideas – effected through the use
of coordinative discourse, and perhaps enhanced by actors’ ability to
accumulate authority from combining knowledge from different kinds
of policy areas (Seabrooke, 2014) – as well as the communication that
goes on between elite and the mass population, what Schmidt (2002)
refers to as communicative discourse. In both cases, the exercise of
power is strategic. That is, although actors are dependent on the idea-
tional framework of the paradigm they seek to support, the selection of
symbols and rhetoric is a ‘deliberate activity because framers are acutely
aware that a frame that fits the prevailing public mood is important
in generating support for policy proposals’ (Campbell, 1998, p. 397).
However, as Campbell (1998) also notes, the effectiveness of power
through ideas also depends on the kinds of material resources brought
to bear by policy actors, that is, how effectively actors are able to project
their ideas to the mass public.
Second, the capacity to exert power through ideas, also in large meas-
ure hinges on the power over ideas, which Carstensen and Schmidt (2014)
refer to as the capacity of actors to control and dominate the mean-
ing of ideas. This form of power may take multiple shapes, including
when actors deploy ideational power to ensure that their ideas remain
predominant so as to guard against challenge to their exercise of coer-
cive power, or questioning of their structural and institutional powers (a
classic example being authoritarian regimes); or, at the other end of the
scale, when otherwise powerless actors successfully employ discursive
means to pressure otherwise powerful actors to act in ways they would
not otherwise have done (e.g. social movements). But, in the case of
policy paradigms, the most relevant deployment of power over ideas
comes in the form of the ability of actors – normally quite powerful also
in terms of institutional position and authority – not to listen, that is, a
capacity to resist alternative ideas. Not rarely does it characterize policy
actors clustered in closed groups of people, as part of, say, epistemic
communities, discourse coalitions, or advocacy coalition networks that
are able to harness enough legitimacy around their policy ideas to avoid
considering alternative approaches. One policy area where this form of
ideational power has been especially prevalent is financial regulation.
Many of the most important ideas in financial regulation – ideas gen-
erally consistent with private sector preferences – were thus hatched
inside transnational networks of experts held together by elite peer rec-

ognition, common and mutually reinforcing interests, and an ambition
to provide global public goods in line with values its members consider
honorable (Tsingou, 2015). Clearly institutional position matters as well,
since granting certain professional groups near-monopoly on develop-
ing regulation and policy provides a privileged position for dominating
which policy ideas and discourses are allowed onto the decision-making
agenda.
Third, and finally, ideational power shows itself as power in ideas,
which refers to the authority certain ideas enjoy in structuring thought
at the expense of other ideas (Carstensen & Schmidt, 2014). This power
is exerted through background ideational processes – constituted by sys-
tems of knowledge, discursive practices, and institutional setups – that
in important ways affect which ideas enjoy authority at the expense
of others. One way to think about this kind of authority of ideas at
the expense of others is in terms of the power of public philosophies
(Schmidt, 2008) or public sentiments (Campbell, 1998) that form the
background of policymaking processes. That is, while the other forms
of ideational power are focused more directly on the interaction going
on between ideational agents, power in ideas concerns the deeper-level
ideational and institutional structures that actors draw upon and relate
their ideas to in order for them to gain recognition among elites and in
the mass public. Power in ideas also affects how actors are able to exert
power over ideas. That is, if over time certain sets of ideas become so
firmly entrenched that they become part of the public philosophy of a
polity, it may enhance the ability of actors to maintain and defend their
ideational authority. One particularly relevant example is the transna-
tional spread of economic discourse in the post-World War II period
that has reshaped how non-economist policymakers understand a given
issue (Hirschman & Popp Berman, 2014).
Ideational power and policy paradigms

Conceptualizing ideational power in this way has important implications
for how we think about policy paradigms. Here I want to draw out four
of these. First, bringing in ideational power, points us toward the impor-
tance of authority. Where Hall (1993) creates an analytical split between
the Bayesian automatic loss of authority following the onslaught of
policy anomalies, and the constructivist contests for authority between
incommensurate policy paradigms, a perspective on ideational power
instead puts the emphasis squarely on the constructivist battle of ideas.
This aligns with Blyth’s (2013, p. 204) point that coalitions of policy
actors ‘may both “power and puzzle,” but successful ones authoritatively
dictate what a puzzle is and how power should be applied to solve it.’
The question then becomes: how are certain paradigms identified by
policy actors as being incommensurate and what kind of material and
non-material resources are policy actors able to muster in the fight for
the authority to make such a call and have it accepted? When is some-
thing considered an anomaly and when are the opinions of experts con-
sidered controversial? Why are actors able to resist the interpretation
of their paradigm as failing while keeping alternative paradigms off the
table? A focus on ideational power opens up the question of how actors
perceive paradigms and their ability to solve problems, and it connects
the question to relations of power – in as well as outside crises – rather
than to the effects of experience and information updating.
Second, the agency-orientation of this understanding of ideational
power distinguishes it from the structural theories of theoretical domi-
nance like Hall’s (1993) paradigm approach, since it emphasizes actors’
ability to ‘stand outside’ and critically engage with the ideas they hold
and promote. This follows from the distinction between ideational
power at the subjective and intersubjective level implied by the inclu-
sion of power through ideas as central for understanding processes of
paradigmatic change. In such a perspective, ideas are not thought of
as internalized or ‘contained’ in the minds of actors, but instead as a
resource – a toolkit and not a coherent system – that exists between and
not inside the minds of actors, and the use of ideas thus demands some
creativity and critical faculty of the actor (Carstensen, 2011b), at times
enabling him or her to ‘buck the system’ (Widmaier et al., 2007). That
is, actors not only have ‘background ideational abilities’ that enable
them to think beyond the (ideational) structures that constrain them
even as they (re)construct them. They also have ‘foreground discursive
abilities’ that enable them to communicate and deliberate about taking
action collectively to change their institutions (Schmidt, 2008). From
this perspective, ideas do indeed become powerful when they are taken
for granted (see also Baumgartner, 2014, p. 476), but here ‘taken for
granted’ does not mean that the ideas have become internalized, but
instead that an intersubjective consensus has arisen – a consensus sub-
ject to challenge from competing coalitions and in need of continual
ideational power wielding to remain stable.
Third, by analytically granting actors the ability to think outside
and strategically about the policy paradigm they support, bringing in
ideational power opens up the possibility for gradual, but significant,
change inside paradigms. As noted in the previous section, the Kuhnian
understanding of paradigms employed by Hall (1993) has the effect of

disposing the paradigm approach for only acknowledging punctuated
equilibriums, but if we open for the possibility that indeed actors are
able and willing to adjust a paradigm, for example, in the effort to bro-
ker between coalitions of actors, or in adjusting it to unforeseen conse-
quences, the approach is better able to detect significant gradual change
either over long periods of time, or even following a crisis. The idea-
tional power of policy actors is important in this context, since it helps
account for the cases where actors have had to defend the status quo by
acknowledging competing ideas and discourses.
Finally, employing the concept of ideational power in no way excludes
the possibility of considering the relevance of other forms of power, like
structural or institutional power. To the contrary, the approach explicitly
includes other forms of power – like institutional or structural power –
to understand why a paradigm changes, or why it survives despite
the appearance of anomalies and pressure from competing advocacy
coalitions.
Conclusion
Western economies are currently coping with tough times of sluggish

growth and austerity, but despite the hardship, for scholars interested
in how paradigms matter for policymaking, these are also quite excit-
ing times. Thus, the financial crisis and the Great Recession that ensued
offer something of a laboratory for studying the dynamics of policy sta-
bility and change. At the outset of the crisis, a fundamental shift in the
ideas that guide economic governance were considered realistic, if not
outright preordained (e.g. Fukuyama, 2008; Hobsbawm, 2009). As the
years have passed, and pre-crisis neoliberal ideas still stand strong, argu-
ments about a pending paradigm shift in the coming years is increas-
ingly viewed as ludicrous. From another perspective, though clearly not
fully borne out, the predictions were not totally off the mark. The world
has thus witnessed the enactment of policies that could not have been
imagined within the standard ‘normal policymaking’ of the pre-crisis
paradigm, including the introduction of policy ideas from competing
paradigms that received little or no hearing in the decades preceding the
current crisis. In other words, we are seeing significant gradual ideational
change rather than the major overhaul critical voices had hoped for.
Approaches to policy paradigms that only allow only for either gen-
eral stability or the drama of punctuated equilibriums have a hard time
dealing with such developments. One of such frameworks, Hall’s (1993)
policy paradigm approach, seems particularly vulnerable to this critique.

As argued in this chapter, despite the profound success and significance
of this seminal statement about the role of policy paradigms in poli-
cymaking, the transposition of a Kuhnian understanding of paradigms
into a context of policymaking is not without its weaknesses. Notably,
it leads to an understanding of political actors as unable to critically
scrutinize the paradigm they hold, thus understating the transformative
potential of agency, and, in turn, it accounts for significant ideational
change with reference to exogenous shocks and the rise of anomalies
that undermine the authority of the policy paradigm, allowing only for
the possibility of paradigm shift, or continuing ‘normal policymaking.’
Given the criticism leveled at Hall’s (1993) paradigm approach, would
it make best sense to altogether simply stop using the concept of a para-
digm? I do not believe that is the best way forward. The concept of
policy paradigms is intuitive, simple, and powerful and has gained wide
recognition in the policy literature and beyond, a status worth defend-
ing. Rather than suggesting that we dispense with the concept of a
policy paradigm, further work is necessary to develop a more nuanced
understanding of the dynamics of change in paradigms. This volume
suggests that important steps in that direction have already been taken.
By flagging the importance of recognizing strategic and creative agency,
as well as placing greater emphasis on the flexibility and malleability
of policy paradigms, these contributions are helpful in solidifying a
promising new direction for the study of paradigms in policymaking.
Moreover, this chapter has suggested a potential new agenda for research
on policy paradigms, namely, first, to divest the paradigm approach of
the Kuhnian inspiration brought in through Hall’s original formulation
and instead attenuate its constructivist base to open for a more dynamic
and agency-centered approach to policy paradigms, and, second, to
emphasize the role of different kinds of power relations in the work-
ings of policy paradigms. To be sure, throughout Hall’s (1993) paper are
sprinkled elements of power through, over, and in ideas – for example,
reference to the importance of media debates, or ‘organized interests
trying to influence the political discourse of the day’ (p. 290) – but these
insights are never really developed and play an underspecified role in
the paradigm framework. In the end, the institutional power accrued
from electoral success ends up in an analytically too-singular position.
It is thus the hope that by bringing in a more developed notion of idea-
tional power, we may gain a firmer grip on how different forms of power
play into processes of stability and change in policy paradigms – an issue
that is more relevant than ever in these times of crisis.
Notes
1 Blyth (2013) seems to generate a bit of a paradox himself, since despite good
arguments for the constructivist approach, and very few good arguments for
the usefulness of Bayesian logic in understanding ideational change – together
with a convincing demonstration that the constructivist logic clearly does
the better job in explaining the outcome of paradigm stability in face of the
largest financial crisis in 80 years – Blyth maintains that the tension in Hall’s
(1993) model is ‘generative’ – or at least was generative in bringing ideas into
historical institutionalism (p. 212) – and ‘demands acknowledgement rather
than resolution’ (p. 211).
2 The conceptualization of ideational power is presented at length in Carstensen
and Schmidt (2014).
References
Adcock, R., Bevir, M., & Stimson, S. C. (2007). Historicizing Historical
Institutionalism(s). In R. Adcock, M. Bevir, & S. C. Stimson (Eds.), Modern
Political Science: Anglo-American Exchanges since 1880 (pp. 259–89). Princeton,
NJ: Princeton University Press.
Baker, A. (2015, forthcoming). Varieties of Economic Crisis, Varieties of Ideational
Change: How and Why Macroeconomic Policy and Financial Regulation Differ.
New Political Economy, DOI:10.1080/13563467.2014.951431.
Barnett, M., & Duvall, R. (2005). Power in International Politics. International
Organization, 59(4), 39–75.
Baumgartner, F. R. (2014). Ideas, Paradigms and Confusion. Journal of European
Baumgartner, F. R., & Jones, B. (2009). Agendas and Instability in American Politics
(2nd ed.). Chicago, IL: Chicago University Press.
Béland, D. (2010). The Idea of Power and the Role of Ideas. Political Studies Review,
8(2), 145–54.
Béland, D., & Cox, R. H. (2013). The Politics of Policy Paradigms. Governance,
26(2), 189–328.
Béland, D., & Cox, R. H. (2014). Ideas as Coalition Magnets: Power, Coalition
Building and Policy Entrepreneurs. Paper Presented at the Workshop on Ideas, Political
Power and Public Policy, June 9–10, Snekkersten, Denmark.
Berman, S. (1998). The Social Democratic Moment: Ideas and Politics in the Making of
Interwar Europe. Cambridge, MA: Harvard University Press.
Berman, S. (2013). Ideational Theorizing in the Social Sciences ‘Since Policy
Paradigms, Social Learning, and the State’. Governance, 23(6), 217–37.
Bermeo, N., & Pontussen, J. (2012). Coping with Crisis: Government Reactions to the
Great Recession. New York: Russell Sage Foundation.
Blyth, M. (1997). Any More Bright Ideas? The Ideational Turn of Comparative
Political Economy. Comparative Politics, 29(1), 229–50.
Blyth, M. (2010). Ideas, Uncertainty, and Evolution. In D. Béland & R. H. Cox

(Eds.), Ideas and Politics in Social Science Research (pp. 83–101). New York: Oxford
University Press.
Braun, B. (2015, forthcoming). Preparedness, Crisis Management and Policy
Change: The Euro Area at the Critical Juncture of 2008–2013. British Journal of
Politics and International Relations, DOI:10.1111/1467-856X.12026.
Campbell, J. L. (1997). Institutional Analysis and the Role of Ideas in Political
Campbell, J. L. (2002). Ideas, Politics and Public Policy. Annual Review of Sociology,
28, 21–38.
Campbell, J. L., & Pedersen, O. K. (Eds.). (2001). The Rise of Neoliberalism and
Institutional Analysis. Princeton, NJ: Princeton University Press.
Campbell, J. L., & Pedersen, O. K. (2014). The National Origins of Policy Ideas:
Knowledge Regimes in the United States, France, Germany, and Denmark. Princeton,
NJ: Princeton University Press.
Carstensen, M. B. (2011a). Ideas Are Not as Stable as Political Scientists Want
Them to Be: A Theory of Incremental Ideational Change. Political Studies, 59(3),
596–615.
Carstensen, M. B. (2011b). Paradigm Man vs. the Bricoleur: An Alternative Vision
of Agency in Ideational Change. European Political Science Review, 3(1), 147–67.
Carstensen, M. B. (2013). Projecting from a Fiction: The Case of Financial Crisis
in Denmark. New Political Economy, 18(4), 555–78.
Carstensen, M. B., & Schmidt, V. A. (2014). Power through, over and in Ideas:
Conceptualizing Ideational Power in Discursive Institutionalism. Paper Presented
at the Workshop on Ideas, Political Power and Public Policy, June 9–10, Snekkersten,
Denmark.
Cashore, B., & Howlett, M. (2009). Punctuating Which Equilibrium? Understanding
Thermostatic Policy Dynamics in Pacific Northwest Forestry. American Journal of
Cox, R. H. (2001). The Social Construction of an Imperative: Why Welfare Reform
Happened in Denmark and The Netherlands but Not in Germany. World
Politics, 53(3), 463–98.
Crouch, C. (2011). The Strange Non-death of Neoliberalism. Cambridge: Polity
Press.
21(3), 453–69.
Fukuyama, F. (2008, October 4). The End of America Inc. Newsweek. Retrieved
from http://www.newsweek.com/fukuyama-end-america-inc-91715
Britain: An Ideational institutionalist Approach. In J. L. Campbell &
O. K. Pedersen (Eds.), The Rise of Neoliberalism and Institutional Analysis
(pp. 193–218). Princeton, NJ: Princeton University Press.
Hay, C. (2011). Pathology Without Crisis? The Strange Demise of the Anglo-
Liberal Growth Model. Government and Opposition, 46(1), 1–31.
Heclo, H. (1974). Modern Social Policies in Britain and Sweden: From Relief to Income
Maintenance. New Haven: Yale University Press.
Helleiner, E. (2014). The Status Quo Crisis: Global Financial Governance after the
2008 Meltdown. New York: Oxford University Press.
Hirschman, D., & Berman, E. P. (2014). State of the Art: Do Economists Make
Policies? On the Political Effects of Economics. Socio-Economic Review, 12(4),
779–811.
Hobsbawm, E. (2009, April 10). Socialism Has Failed. Now Capitalism Is Bankrupt.
So What Comes Next? The Guardian. Retrieved from http://www.theguardian.
com/commentisfree/2009/apr/10/financial-crisis-capitalism-socialism-
alternatives
Jacobs, A. M. (2009). How Do Ideas Matter? Mental Models and Attention in
German Pension Politics. Comparative Political Studies, 42(2), 252–79.
Health Insurance. Australian Journal of Political Science, 42(4), 579–91.
Kingdon, J. W. (2003). Agendas, Alternatives, and Public Policies (2nd ed.). New York:
Longman.
Kuhn, T. (1970). The Structure of Scientific Revolutions (2nd ed.). Chicago, IL:
Chicago University Press.
Lieberman, R. C. (2002). Ideas, Institutions and Political Order: Explaining
Political Change. American Political Science Review, 96(4), 697–712.
Mahoney, J., & Thelen, K. (2010). A Theory of Gradual Institutional Change. In
J. Mahoney & K. Thelen (Eds.), Explaining Institutional Change: Ambiguity,
Agency, and Power (pp. 1–37). New York: Cambridge University Press.
Mondou, M., Skogstad, G., & Houle, D. (2014). Policy Image Resilience,
Multidimensionality, and Policy Image Management: A Study of US Biofuel
Policy. Journal of Public Policy, 34(1), 155–80.
Moschella, M., & Tsingou, E. (Eds.). (2013). Great Expectations, Small Transformations:
Incremental Change in Financial Governance. Colchester: ECPR Press.
Mügge, D. (2013). Resilient Neo-liberalism in European Financial Regulation. In
V. A. Schmidt & M. Thatcher (Eds.), Resilient Liberalism in Europe’s Political
Economy (pp. 201–26). Cambridge: Cambridge University Press.
26(2), 259–81.
Parsons, C. (2007). How to Map Arguments in Political Science. New York: Oxford
University Press.
Princen, S., & t’Hart, P. (2014). Putting Policy Paradigms in Their Place. Journal of
Sabatier, P. A., & Jenkins-Smith, H. C. (1993). Policy Change and Learning: An
Salines, M., Glöckler, G., & Truchlewski, Z. (2012). Existential Crisis, Incremental
Response: The Eurozone’s Dual Institutional Evolution 2007–2011. Journal of
Schmidt, V. A. (2002). The Futures of European Capitalism. Oxford: Oxford
University Press.
Schmidt, V. A. (2006). Democracy in Europe: The EU and National Polities. Oxford:

Oxford University Press.
Schmidt, V. A. (2011). Ideas and Discourse in Transformational Political Economic
Change in Europe. In G. Skogstad (Ed.), Policy Paradigms, Transnationalism, and
Schmidt, V. A., & Thatcher, M. (Eds.). (2013). Resilient Liberalism in Europe’s
Political Economy. Cambridge: Cambridge University Press.
Seabrooke, L. (2009). Everyday Legitimacy and Institutional Change. In A. Gofas &
C. Hay (Eds.), The Role of Ideas in Political Analysis: A Portrait of Contemporary
Debates (pp. 167–86). London: Routledge.
Seabrooke, L. (2014). Epistemic Arbitrage: Transnational Professional Knowledge
in Action. Journal of Professions and Organization, 1(1), 49–64.
Skogstad, G., & Schmidt, V. A. (2011). Introduction. In Grace Skogstad (Ed.),
Policy Paradigms, Transnationalism, and Domestic Politics (pp. 3–35). Toronto:
University of Toronto Press.
Stone, D. (1988). Policy Paradox and Political Reason. Glenview, IL: Scott,
Foresman & Co.
Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond Continuity:
Institutional Change in Advanced Political Economies (pp. 1–39). New York: Oxford
University Press.
Tsingou, E. (2015). Club Governance and the Making of Global Financial Rules.
Review of International Political Economy, 22(2), 225–256.
Widmaier, W. W., Blyth, M., & Seabrooke, L. (2007). Exogenous Shocks or
Endogenous Constructions? The Meanings of Wars and Crises. International
Studies Quarterly, 51(4), 747–59.
Wilder, M. (2015, forthcoming). Ideas beyond Paradigms: Relative Commen-
surability and the Case of Canadian Trade-Industrial Policy, 1975–1995. Journal
of European Public Policy, DOI:10.1080/13501763.2014.981568
Wilder, M., & Howlett, M. (2014). The Politics of Policy Anomalies: Bricolage and
the Hermeneutics of Paradigms. Critical Policy Studies, 8(2), 183–202.
Wood, M. (2015). Puzzling and Powering in Policy Paradigm Shifts: Politicization,
Depoliticization and Social Learning. Critical Policy Studies, 9(1), 2–21.
Index
advocacy coalition framework (ACF) origin of, 197–8

agency, 94 ostensive policy formation and
hierarchical belief system, 220 performance, 193–4
individuals by belief system, 92–4 ostensive–performative theory of
information as policy dynamics, routines, 192–3
248 outcomes- and competencies-based
methodological individualism, 94 approaches, 211
paradigms and ideas, 93–6 paradigmatic and theoretical
policy change, 95 commensurability, 195–6
policy core beliefs, 220 Prague Communiqué, 202–5
policy subsystem, 93 routines theory approach, 190–2
public policies, 93, 95 Sorbonne Declaration, 198–9
ARAMIS, 137 boundary-spanning problems, 252–4
bureaucracies
Baumgartner, F. agenda control by, 259
paradigm definition by, 5 congressional decision making
paradigmatic change, 8 and, 258
paradigmatic shift, 9 information supply, 250
search in policy process, 241, 244 policy making by, 257
Bayesian social learning, 302, 307 of witnesses in energy policy, 261
belief system, 92–4
Bergen Communiqué, 206–7, 212 Cairney, Paul, advocacy coalition
Berlin Communiqué, 203, 205–6 framework, 83
Berman, S. calculus approach, 66, 73
discursive institutionalism, 69 Carstensen, Martin B., 295
ideational analysis, 67–8 cause–effect relationships, 24
Blyth, Mark Civil emergency planning, 145
Bayesian learning, 302 Climate Change Act of 2008, 280
constructivist institutionalism, 74 climate changes (2008–2011) in UK
policy making, 9 carbon dioxide emissions
Bologna Declaration, 199–201 reduction, 279
Bologna Process Climate Change Act of 2008, 280
Bergen Communiqué, 206–7, 212 domestic energy production, 280
Berlin Communiqué, 203, 205–6 energy and climate security, 281
Bologna Declaration, 199–201 new policy documents and
continuous change, 194–5 instruments, 281–2
emergence and convergence, 207–8 Colorado Oil and Gas Conservation
formation and implementation, Commission (COGCC), 222–3
208–11 Colorado, US
foundation, 207 COGCC meeting on 2011, 222
inter-organisational nature of hydraulic fracturing processes, 229
routines, 196–7 policy actors survey, 224
objective, 190 shale gas and oil development, 221
319
320 Index
communication philosophical ideas, 71

boundary-spanning problems, 252–4 policy ideas, 71
bureaucratic channel, 257–9 programmatic ideas, 71
complexity and noise, 251 Dublin Descriptors, 203, 204
feedbacks, 254
goal refinement, 255 The Economist, 173
information supply in policy economy policy of UK, 89, 91–2, 123
subsystems, 256–7 educational policy harmonisation. see
information supply, under Bologna Process
pluralistic competition, 249–51 egalitarianism, 219
policy analysis, 254–5 energy and climate security, 281
signaling, 252 Energy Information Administration
signaling and policy dynamics, 249 (EIA), 259
communicative discourse, 133 Energy Review – 2006, 278
Congress, United States, 239–62 enigmas in science, 119–20, 122
congressional policy making, 259–62 entropic search, 247
constructivist institutionalism environmental and tobacco policy, 92
actors and institutions, 74 epiphenomenalism, 62
Blyth, M., 72 epistemological consequences, of
calculus, 73 Kuhn’s paradigms, 21–3
crises and paradigm change, 74–5 European qualifications framework
cultural ontologies, 73 (EQF), 212
social learning and change, 73 European policy change, 70
Cooper–Whelan report, 175 institutional setting, 70
crisis testing, 169 philosophical ideas, 71
Critical Junctures Theory (CJT) policy ideas, 71
crisis testing, 169 programmatic ideas, 71
discursive institutionalism, 168 Schmidt, V., 69–72
evaluation of findings, 171 exemplar, 44
ideational change testing, 169–70 explanation of stability, 75
policy change testing, 170–1
crystallized coalitions, 110, 111 fatalism, 219
Culliton report 1991, 181 Feeney, Sharon, 189
cultural approach to Fianna Fáil, 171, 174
institutionalism, 66 Foucault’s concept
Cultural Theory, 219–20 dispositif, 128–9
episteme, 128
Daigneault, Pierre-Marc fracing/fracking. see hydraulic
ideas and paradigms, 10 fracturing regulations
ideational content, 299 French economic policy, 125
deep core beliefs, 92, 218–20 functional disruption, 150–1, 155–6
Department for Energy and Climate functionalist, 62
Change (DECC), 280–1
dependent variable policy, 28–9 Governing the economy, 123
disciplinary matrix, 44
disclosure rule, 222, 225–7 Hall, Peter
discursive institutionalism (DI), 69 institutionalism, contribution to
European policy change, 70 calculus approach, 66
institutional setting, 70 cultural approach, 66
Index 321
policy paradigm, 64–5 Cultural Theory, 219–20

state-centric approach, 65 deep core beliefs, 218–19
orthodox paradigm change disclosure and setback variables, 231
models, 103–5 empirical analysis data, 223–4
paradigm approach hierarchists views, 225
commensurability, 32–4 independent variables, 232–3
description of paradigmatic individualists views, 225
policy change, 89–91 new rules impact, 230
incommensurability thesis, online survey, 224
298–301 policy actors, on government
incremental policymaking, 23 regulation, 225, 227
model of ideational politics, 103 policy core beliefs, 218, 226, 230
paradigm change, 304–7 single model, 228, 229
policy anomalies role, 302–4 stakeholders data collection, 224
policy settings, 23 two models, 227, 228
problems in paradigm-based Howlett, Michael, 3, 101
models, 105 policy paradigms and change, 5
social science study, 24
taking stock of Hall’s paradigm ideas and paradigms, 84–6
approach, 297–8 Advocacy Coalition Framework,
paradigmatic policy changes 93–6
policy failure, 89, 90 descriptions, 84–8
political system, shift in Hall’s policy paradigms and ideas,
authority, 89, 90 85–93
punctuated evolution, 91 ideas and policy-making, 101–3
paradigms and ideas ideas and the policy process, 218–21
versus Advocacy Coalition ideas and welfare reform
Framework, 93–6 access to policy actors, 56
categorization, 85 data collection, 55
definitions, 85–6 policy ideas determination, 56
hegemonic policy paradigms, 88 policy paradigms development, 55
institutionalism, 87 policy paradigms measurement,
interpretation of ideas, 86–9 55–6
persuasion and argument, 86 ideational analysis
political system, 88 policy order changes, 67–8, 89–91
policy paradigm, 19 social learning literature, 67–8
Hay, Colin, constructivist state theories, 67
institutionalism, 73–5 ideational change testing, 169–70
Heikkila, Tanya, 217 ideational power
hierarchism, 219 conceptualizing ideational power,
historical institutionalism (HI), 63 309–11
Hogan, John, 167, 168 definition, 309
critical juncture theory, 167 versus ideas matter, 307–9
policy paradigms and change, 5 policy paradigms, 311–13
Horan, Conor, 189 power in ideas, 311
hydraulic fracturing regulations power over ideas, 310–11
case study, 221–3 identity, policy proposal by, 136
COGCC disclosure and setbacks incentive-based information supply,
rule, 224, 226, 227 244–5
322 Index
incommensurability thesis, 12, 44, inflation, 171

105–6, 298–301 media perception, 173
British economic policy and, 298–9 oil crisis (1974), 171
fundamentally different, 298 political systems, 173–4
ideational content by Daigneault, public expenditure, 173
299 unemployment, 171
internally coherent, 298 Irish industrial policy
individualism, 219 alterations to industrial policy,
industrial policy changes, 175–80 178–9
inferential sequence linking, 29 Cooper–Whelan report, 175
information critical juncture, testing for, 167–8
as a private good, 244 Culliton report 1991, 181
competition-based information ideas underlying industrial
supply, 247–9 policy, 175
congressional policy making, ideational collapse, 176–7, 183
259–62 inter-coder agreement scores, 183
economics of, 244 level of policy change, 180, 183
expert search and solution macroeconomic crisis
definition, 245–6 identification, 171–5
incentive-based information new ideational consolidation, 180
supply, 244–5 political instability/policy drift, 177
policy dynamics, 239, 243 political response to Telesis, 178
political institutions, 241 Telesis report, 175–6
problem space, 242–3 The Irish Times, 173
public policy, 241–2
signals and incentives, 246–7 Kern, Florian, 269
solution space, 242, 243 Keynesianism, and monetarism, 66
information processing perspective, Kingdon’s policy stream with ideas, 86
247–8 Kordig, C., 26
institutionalism Kuhn, Thomas
constructivist institutionalism, 72–6 gestalt, 104
discursive institutionalism, 69–72 gestalt metaphor and policy
a fourth institutionalism, 61 sciences, 21–4
ideational analysis, 67–9 natural science concept, 19, 118–21
new institutionalism, 62–4 scientific development theory
Peter Hall’s contribution to, 64–6 definition of paradigms, 22–3
programmatic fashion, 62 gestalt metaphor, 21–5
instrumentalist, 62 incremental policymaking, 23
intergovernmental systems, NPM knowledge production, 117
reform in, 143–4 policymakers, 24
inter-organisational nature of political revolutions, 22
routines, 196–7 scientific paradigms, 23–4, 103–5
intra-paradigmatic policy changes, 104 scientific paradigm to policy
Irish economy in 1980s paradigm, 121–3
economic indicators (1974–1989), Krippendorff’s alpha, 171
172 Kuzemko, Caroline, 269
economist’s policy, 173
fiscal policy (1977), 171 Lakatos, I., 33
GDP growth, 172 legitimacy, policy proposal by, 136–7
Index 323
Maastricht Treaty, 201 incommensurability thesis, 105–6

marginal policy change, 122 iterative intra-paradigmatic cycling,
Mitchell, Catherine, 269 106, 107
monetarism, Keynesianism and, 66 modifications to Hall’s model, 107
monitoring costs, 149, 154 re-evaluation of Hall’s thesis, 298–9
operationalize paradigms
natural science concept dependent variable policy, 28–9
coherent and knowledge, principles, inferential sequence linking, 29
laws, and ideas, 119 transnational advocacy and
enigma and paradigm, 119–20 learning, 30
Newtonian paradigm, 119 Orenstein, M., 25
normal period and paradigm, 118–19 Organisation for Economic
scientific revolution, 120–1 Cooperation and Development
social group, 120 (OECD), 30
neo-institutionalists, 63–4, 67 O’Rourke, Brendan, 167
new institutionalism, 62–4 discursive institutionalism, 169
new public management (NPM) reform ossification, 149–50, 155
description, 142–3 ostensive–performative theory of
intergovernmental systems, routines, 192–3
implementation in, 143–4 ostensive policy formation and
measures performance, 193–4
functional disruption, 150–1, 155–6 overcommitment, 151, 156
monitoring costs, 149, 154
ossification, 149–50, 155 paradigm
overcommitment, 151, 156 Baumgartner’s definition, 88
performance dimensions, 155 Bringing in ideational power, 307
suboptimization, 151, 156–7 gradual change in paradigms, 304–7
tunnel vision, 149 Hall’s definition, 7
motivations hard paradigms, 21–5
functional disruption, 150–1 heuristic concept, 118–21
local-level public managers, 146–7 operative paradigms, 28–32
monitoring costs, 149 softer policy paradigms, 25–8
manager perceptions, 147–8 Kuhn’s definition, 22
overcommitment, 151 unintended consequences of NPM
political institutions, 147 ideas, 143–4
professional culture, 147 paradigm concept
reformers and managers, 148 an application, 53–6
suboptimization, 152 coherent ideas, 50
tunnel vision, 149–50 conceptual framework fare in
unintended consequences, 152–3 practice, 53–5
Newtonian paradigm, 119 empirical research on, 51–3, 105–9
Nohrstedt, Daniel, paradigms and first and second order change, 46
unintended consequences, 141 ideas, consequences of, 47–8
results, 154 measuring policy paradigms, 55–6
ontological level, 50
Oliver and Pemberton theory outstanding research questions in
British monetary policy, 25 new paradigmatic theory,
evolutionary iterative framework, 110–11
107–8 paradigmatic status, 51
324 Index
paradigm concept – continued incommensurability, 12

policy actors, 56 institutionalism, 271
policy ideas, 56 normal and stable periods, 122
policymaking-instrument settings, 46 ontology and methodology, 49
policy paradigm literature, 46–9 outstanding issues, 11–12
problem definition and solution paradigms and social groups, 122
construction, 111 policy change, study of, 5–8
in public policy, 45 re-evaluation of, 9–10
roots of concept, 44 reproducible effects, 126
in scientific inquiry, 44 scientific knowledge, 122
social policy literature, 48–9 significant policy change, 122
paradigmatic change, 21–5 six-part disaggregation of policy, 29
performance dimensions, 155 as social learning, 272
Pierce, Jonathan J., 217 speed and pattern of change, 11–12
policy anomalies success and failure, 126–7
contingent strategies and frame supporters of paradigm, 126
extension, 109 technical analysis/policy bricolage, 11
crystallized coalitions, 110, 111 UK energy policy, 273–83
descriptions, 101–2 policy proposal, as cost of recruitment
empirical basis, 105–9 ideas, 137–8
ex post regulation, 108 by identity, 136
frame compliance, 108 power and legitimation, 136–7
frame defection, 108 by time, 136
gatekeeping functions, 108–9 policy regimes, 252–4
incommensurability thesis, 105–6 policy statement
low order defections, 108 career of a policy proposal, 138
paradigm change models, 103–5 communication and appropriation,
paradigmatic change and 130–1
hermeneutics, 111–13 critique and to enroll, 133–4
policy actors, 110 interaction, 130
policy entrepreneurialism, 110, 111 persuade and to convince, 133, 134
and policy paradigm theories, 102 policy proposal and value, 130
political dimensions, 102 redefinition of policy proposal, 135
subsystemic actors, 110 statement coalition, 132–5
policy bricolage, 11 policy subsystems, information supply
policy change testing, 170–1 in, 256–7
policy core beliefs, 92, 218, 226, 230 political instability/policy drift, 177
policy designs, 53 political system, shift in authority,
policy paradigm concept 89, 90
academic field, 123–4 power and legitimation, policy
changing/replacing, 123, 271–3, proposal by, 136–7
273–82 Prague Communiqué, 202–5
context, 8–9 problem of the explanandum, 53
developments of, 19–20 problem space, 242–3
enigmas in science, 122 pro-market energy policy paradigm
as framework of ideas, 271 in 2000, 275
governing the economy, 26–7, 123 coalition of actors, 276
hierarchical structure, 125 governance institutions, 274, 276
hypothesis order, 125 interpretive framework, 274
Index 325
objective, 274, 276 The Structure of Scientific Revolutions

policy instruments, 274, 277 (Kuhn), 19, 22–4, 65
punctuated equilibrium model, 91, suboptimization, 151, 156–7
194–5, 213 Swedish local government, emergency
planning in, 141–6
rational choice, 73 case background and data, 144–6
rational choice institutionalism (RI), 63
rational processes, 76 Telesis report, 175–6, 178
Rayner, Jeremy, a fourth time, policy proposal by, 136
institutionalism, 61 transnational advocacy and
reductionist, 62 learning, 30
réferéntiels, 25 tunnel vision, 149
A Review of Industrial Policy, 175
routines theory approach, 190–2 UK economic policy, 89, 91–2
Russo-Ukrainian gas transit dispute, 277 UK energy policy among 2000
and 2011
Schmidt, Vivien, 69–71 climate changes (2008–2011)
scientific development theory carbon dioxide emissions
coherent and knowledge, principles, reduction, 279
laws, and ideas, 119 Climate Change Act of 2008, 280
definition of paradigms, 22–3 domestic energy production, 280
enigma and paradigm, 119–20 energy and climate security, 281
incremental policymaking, 23 new policy documents and
Newtonian paradigm, 119 instruments, 281–2
normal period and paradigm, policy paradigm change, 285–6
118–19 pro-market energy policy paradigm
policymakers, 24 in 2000, 275
political revolutions, 22 coalition of actors, 276
scientific paradigms, 23–4 governance institutions, 274, 276
scientific revolution, 120–1 interpretive framework, 274
social group, 120 objective, 274, 276
scientific paradigm, 121–3 policy instruments, 274, 277
scientific revolution, 44 security of supply crisis (2004–2007)
secondary beliefs, 92 Energy Review – 2006, 278
security of supply crisis (2004–2007) policy instruments, 278
in UK Russo-Ukrainian gas transit
Energy Review – 2006, 278 dispute, 277
policy instruments, 278 utilitarian, 62
Russo-Ukrainian gas transit
dispute, 277 Weible, Christopher M., 83, 217
setback rule, 218, 223–7 advocacy coalition framework, 83
shale gas policymaking, 131–2, 134–5 Weltanschauungen, 71
Shafran, JoBeth S., 239 Wilder, Matt, 19, 101
Simon’s model, 122 exploring operative paradigms, 28
Single European Act, 201 six-part disaggregation of policy, 29
sociological institutionalism (SI), 63 Workman, Samuel, 239
solution space, 242, 243
Sorbonne Declaration, 198–9 Zittoun, Philippe, episteme concept,
Stigler, George, 244 301

Policy Paradigm-Theory and Pratices

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Policy Paradigm-Theory and Pratices

Uploaded by

Copyright:

Available Formats

Studies in the Political Economy of Public Policy

International Advisory Board

All rights reserved. No reproduction, copy or transmission of this

No portion of this publication may be reproduced, copied or transmitted

First published 2015 by

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,

Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

Palgrave Macmillan is the global academic imprint of the above companies

Palgrave® and Macmillan® are registered trademarks in the United States,

ISBN 978-1-349-56900-7 ISBN 978-1-137-43404-3 (eBook)

List of Tables and Figures vii

List of Abbreviations xvi

2 What Is a Policy Paradigm? Overcoming Epistemological

3 Can You Recognize a Paradigm When You See One?

4 Is There a Fourth Institutionalism? Ideas, Institutions

6 Paradigm Construction and the Politics of Policy Anomalies 101

7 From Policy Paradigm to Policy Statement: A New Way to

8 Paradigms and Unintended Consequences: New Public

Part III 165

10 The Bologna Process and the European Qualiﬁcations

11 The Role of Ideas in Evaluating and Addressing Hydraulic

12 Communications Frameworks and the Supply of

13 How and Why Do Policy Paradigms Change; and

8.1 Overview of unintended consequences,

9.1 Ireland’s main economic indicators, 1974–1989 172

9.2 Identiﬁcation of macroeconomic crisis in Ireland

9.3 Indication of ideational collapse 176

9.4 Indication of (1) new ideational consolidation and

10.1 The objectives of the Bologna Declaration, 1999 200

10.2 Artifacts that relate to Objective #1 of the Bologna

10.3 Qualiﬁcations descriptors (known as the

11.1 Explaining policy actor evaluations of problem

11.2 Explaining variance in changes in positions on

11.3 Explaining changes in support for government

11A.1 Means and factor loadings for disclosure and

11A.2 Measures, means, and factor loadings for changes

11A.3 Measures, means, and factor loadings for

12.1 Witnesses at congressional hearings on energy policy,

12.2 Bureaucratic composition of witnesses in energy policy,

13.1 The pro-market energy policy paradigm in 2000 275

13.2 The energy policy paradigm in 2000 and 2011 283

2.1 Six-part disaggregation of policy 29

2.2 Normative and technical elements of policy in a

6.1 Oliver and Pemberton’s evolutionary iterative

6.2 Contingent strategies and frame extension 109

6.3 Two processes of problem deﬁnition and

8.1 Summary of survey responses indicating unintended

10.1 The ostensive–performative theory of routines 194

10.2 Convergence from policy formation to policy

This volume ﬁnds its earliest origins in a workshop at the ECPR in

Paul Cairney is Professor of Politics and Public Policy, University of

Martin B. Carstensen is Assistant Professor in the Department of

Pierre-Marc Daigneault is Assistant Professor of Public Policy and Public

He also co-edits, with Daniel Béland, Welfare Reform in Canada: Provincial

Sharon Feeney lectures in Business Policy and Corporate Strategy in the

Tanya Heikkila is Associate Professor and Doctoral Program Director

John Hogan lectures on Irish Politics and International Political Economy

Conor Horan is Lecturer in Marketing at the Dublin Institute of

Michael Howlett is Burnaby Mountain Chair in the Department of

Florian Kern is Senior Lecturer at SPRU (Science Policy Research Unit)

Caroline Kuzemko is a research fellow working as part of an EPSRC-

She has recently published The Energy Security-Climate Nexus: Institutional

Catherine Mitchell is Professor of Energy Policy at the University of