Professional Documents
Culture Documents
ABN-Amro - Mobile Networks - W
ABN-Amro - Mobile Networks - W
Pan-European Telecoms
3G Tsunami: the revolution begins Jamie Mariani
+44 20 7678 0243
jamie.mariani@uk.abnamro.com
Rodney Sherrington
+44 20 7678 1610
rodney.sherrington@uk.abnamro.com
Marketing Analyst
THE AMERICAS EUROPE AND AFRICA ASIA PACIFIC Simon Carrington
New York United Kingdom Hong Kong +44 20 7678 0395
ABN AMRO Inc. ABN AMRO Equities (UK) Ltd ABN AMRO Asia Ltd simon.carrington@uk.abnamro.com
Park Avenue Plaza 250 Bishopsgate 38th Floor, Cheung Kong Center
55 East 52nd Street, London EC2M 4AA, UK 2 Queen’s Road Central, Hong Kong
6th Floor Tel: +44 20 7678 8000 Tel: +852 2700 3000
New York, New York 10055 Fax: +44 20 7678 7353 Tlx: HX 74111
Tel: +1 212 409 1000 Fax: +852 2700 5000
Netherlands
Fax: +1 212 409 5224
ABN AMRO Bank N.V. Auckland
Boston Gustav Mahlerlaan 10 ABN AMRO Equities (NZ) Ltd
ABN AMRO Inc. 1082 PP Amsterdam +64 9 358 7500
+1 617 753 6900 The Netherlands Bangkok
Chicago Tel: +31 20 628 9393 ABN AMRO Asia Securities Public
ABN AMRO Inc. Fax: +31 20 628 7740 Company Limited
+1 877 711 1781 Brussels +66 2 679 5790
Houston ABN AMRO Bank N.V. Bombay
ABN AMRO Inc. +32 2 546 0460 ABN AMRO Asia Equities (India) Ltd
+1 832 681 7100 Bucharest +91 22 285 4363
Mexico ABN AMRO Bank (Romania) S.A. Jakarta
ABN AMRO Securities (Mexico) SA +40 1 2 020 400 PT ABN AMRO Asia Securities (Indonesia)
+52555 2577800 Cairo +62 21 515 1880
Madrid Sydney
ABN AMRO Equities (Spain) S.A. SVB ABN AMRO Equities Australia Ltd
+61 2 8259 5000
58% price cuts may deliver
+34 91 423 6900/82
Milan Taipei 30% ARPU growth by 2006
ABN AMRO Bank N.V. ABN AMRO Bank NV
+39 02 722671 +886 2 25037888
maintain a neutral stance and highlight mmo2 (Buy from Add) and Rodney Sherrington
rodney.sherrington@uk.abnamro.com
Orange (Buy) as preferred plays. +44 207 678 1610
Summary of recommendations
New Recomm. Old Recomm. Price Target
The European cellular equity story has changed. While stale bulls
continue to cling to the 'blue sky' mobile data service vision, mobile voice
communication is the killer application. We believe wireless networks will
account for 30%-40% of originated traffic (from 10%-15% today),
structurally shifting ARPU by up to 30% by end 2006.
(Buy from Add). We have reduced our price targets for TIM (€4.1 from €4.4) Source: Datastream
Please refer to terms relating to the provision of this research at the end of the document.
ABN AMRO UK
250 Bishopsgate
London
EC2M 4AA
http://www.abnamroresearch.com
United Kingdom
In Brief…
What's Changed
Performance. European wireless stocks have outperformed strongly since we last
looked in detail at the subsector (European Cellular Bear Trap dated 9 July 2002).
Industry restructuring, strong Q202 results and scope for even better numbers at the
Q3s have underpinned this achievement. Our preference for beta has been partially
validated with strong relative performance from Orange (Buy), Vodafone (Add) and
mmo2 (Buy from Add). To the end of the Q302 results season, we expect continued
strong share price movements. Beyond November the key management challenge will
be to turn volume into value.
15% 10%
6% 7%
4%
5%
-5% FTSE EURO DJ Stoxx 50 FTSE Global TEM TIM mmo2 Vodafone Orange
100 Telco Services
-15% -6%
-15%
-25% -20%
Percentage change
Recommendations. We have upgraded mmo2 to Buy (from Add) and increased our
fair value estimate to 60p, reflecting recent consolidation in the German market and
the prosect of further restructuring. The Buy recommendation on Orange is
maintained, with a price target of €7. Price targets for the southern European
operators have been lowered to €4.1 and €6.0 for TIM and TEM respectively. We are
leaving our view on Vodafone unchanged (Add, 105p price target).
A C T I O N P O I N T S
Issues to Consider
The mobile data services vision is increasingly bankrupt. The core problem
facing Europe’s wireless operators is simple, identifying opportunities large enough to
drive top-line growth today’s absolute level. In our view, data services should be
viewed as an incremental bonus. Disciplined capital investment should require focus
on revenue streams, where sufficient demand exists to validate investment.
Switching between business models will cause friction. In our view (should
bucket pricing plans emerge), a margin contraction is inevitable. Dramatically falling
prices will drive structurally lower gross telephony margins. Ironically, operators with
exposure to the corporate market could be the worst affected. Notwithstanding this
uncertainty, longer-term investors should view fixed to mobile substitution as a
potentially unique opportunity. We cannot think of any other sector in the wider
economy that offers such easily identifiable growth.
I NDUST RY DY NA MI CS
MA CRO DY NA MI CS
SE CT OR DY NA MI CS
3G as the facilitator 59
In our view, European 2G capacity constraints have prohibited the introduction of
bucket pricing plans (and by association fixed-to-mobile substitution). While 3G
may facilitate the use of high-speed applications, the re...
Introduction 59
RIS K A NA LYS IS
Economic impact 73
We see between 10% and 25% upside from our existing fair value estimates for
the European wireless community as a result of (1) our pre-to-post-paid
migration thesis and (2) our central case fixed-to-mobile substitution...
Timing 73
A PPE NDIX
TE AM
Acknowledgements 83
The mobile data services vision appears increasingly bankrupt. Global SMS revenue
amounted to about US$13bn (approximately £8.7bn) in 2001, contributing a mere
4% to total global mobile revenue of nearly US$300bn (approximately £200bn). In
the E5 (France, Germany, Italy, Spain and the UK), we estimate mobile revenue of
€75bn was generated during 2001. The core problem facing Europe’s wireless
operators is simple, identifying opportunities large enough to drive top-line growth
from this absolute level. Contrary to some commentators’ expectations, we do not
believe that MMS will materially affect European operators’ top line during the next
24 months. Disciplined capital investment should require operators to focus on
identifiable revenue streams. In our view data should be viewed as an incremental
“bonus.”
Introduction
In this section of the note we present:
■ the size of the E5 voice and data market (€75bn and €6bn, respectively);
■ our view on the sustainability of the low volume/high prices of SMS; Data services look
increasingly like the ‘better
■ the problem of scale – why MMS will struggle to drive growth; mousetrap’ nobody wants
European revenue
In the E5, we estimate that mobile revenue of €75bn was generated during 2001, Wireless revenue of €75bn
was generated during 2001
accounting for about 1.1% of GDP. The problem facing Europe’s major operators is
in the E5
clear – identifying opportunities large enough to generate material growth from this
absolute number.
% Change 40% 9%
Data revenue
Relative to the absolute size of the European cellular market, data pales by
comparison. We calculate that data revenue of €6.3bn was posted in 2001,
corresponding to a small 8.3% of the total top line, or nearly 10% of service
revenue.
SMS per bit pricing multiple (x) 450 SMS cost/unit volume
Source: ABN AMRO priced at a 450x premium
youth segment (the keenest users of SMS), we understand that SMS traffic
represents less than 0.5% of total traffic. We can demonstrate this with reference
to TIM, the leading Italian operator. To June 2002, TIM generated 17.8bn voice
minutes and 3.8bn messages (see following table).
We calculate that SMS as a percentage of total traffic accounts for about 0.05% of
usage. While Italy does not “lead the curve” in terms of SMS usage, our example
underlines an important point, namely, the tiny volume contribution of SMS.
Our discussions with investors suggest to us that MMS is still perceived as the silver MMS silver bullet status
provides some hope ...
bullet for European cellular. Following the somewhat unexpected success of SMS,
the market appears to believe that MMS provides the next step on the long path to
build a credible mobile data equity story.
■ Historically, J-Phone has charged between €0.12 and €0.33 per message (about Sha-mail is often cited as
example of MMS success
£0.21, at the top end).
■ At the end of August 2002, J-Phone had 6m Sha-mail enabled handsets (at the
end of August 2001, J-Phone had approximately 1.8m Sha-mail handsets sold).
■ In early 2002 (at the time of the J-Phone presentation to analysts, 28 February
2002), Vodafone management indicated that the J-Phone picture messaging
service was being used about twice a week.
For simplicity, we have factored in a constant “two messages sent a week” scenario
to calculate the revenue generated from MMS.
MMS revenue as a percentage of total sales 1.1% ... but MMS revenue is in
Source: ABN AMRO
the noise (1.1%)
Europe (W & E) 24 31 35 35
- W Europe 24 30 34 34
- E Europe 34 39 48 40
Asia/Pacific 34 37 46 47
North America 27 31 33 32
South America 30 37 50 48
Total 28 33 39 39
A three-year handset replacement cycle suggests an annual rate of handset churn Even if 50% of all Western
European handsets sold in
of nearly 33%. Even if 50% of the replacement handsets sold during 2003 were
2003 were MMS, only 17%
MMS-enabled this corresponds to only 17% of Western Europe’s existing subscriber of Western Europe’s base
base owning an MMS-enabled handset by calendar year-end 2003, some distance would be MMS-enabled
Using the 34-month handset replacement cycle, Western Europe would need to MMS critical mass is
reached CY2005
wait until year-end 2004 to reach the point of critical mass in terms of MMS
penetration. This suggests that a material revenue boost will not occur until
CY2005. Bulls of MMS beware!
MMS pricing structures vary on a case-by-case basis. The average price appears to
be around €0.40-0.60 per message (about 25p-40p; outliers such as Telenor in
Norway have recently dramatically reduced their price from as much as €1.33 per
message to €0.65).
In our example we have factored in 11% and 26% of Vodafone’s subscriber base to
buy MMS-enabled handsets by the end of March 2003 and March 2004,
respectively, and an average price of £0.25 per message. Our basic analysis shows
that to deliver a 10% upside surprise to our existing revenue forecast, Vodafone
would need to encourage its initial MMS-enabled subscribers to pay for 31.6
messages a week to the end of March 2003 (or 13.8 messages per week to the end
of March 2004). We believe this level of volume is too challenging.
■ factoring in a generous 26% of Vodafone’s base will buy (and use) MMS-enabled
handsets (implying 25.6m MMS subscribers at the end of March 2004 from
10.4m at the end of March 2003);
Pulling these drivers together suggests potential incremental revenue of nearly £2.2
per month per subscriber, corresponding to £26 per year. In total, this offers
£469m of additional revenue to the end of March 2004 (about 1.5% of consolidated
revenue). This realistic upside is in the noise (see following table).
Voice cannibalisation
We also note the potential cannibalisation of existing SMS revenue by MMS, and the Evidence to date suggests
that messaging has
potential of MMS to substitute for voice usage. Arguably MMS can be viewed as the
cannibalised existing voice
classic “technology” push innovation, for which little or insufficient demand exists. revenue streams
Should customers not perceive they have a real "new” need for the technology,
they could either:
Arguably evidence to date suggests that data has been generated at the expense of
existing voice revenue streams. In the charts that follow we show that NTT
DoCoMo,1 the widely accepted leader in data service deployment, has not yet
registered an increase in ARPU, and data ARPU appear to have been generated
largely at the expense of voice ARPU. Similarly, Vodafone’s voice ARPU in its key
European territories has continued to decline at a faster rate than blended ARPU
(assuming constant spending patterns across time), suggesting data service
cannibalisation.
1
Our analysis of NTT DoCoMo ignores the impact of marginal investment costs for 2G. In Japan, a
high number of subscribers for available spectrum has led to higher capital costs (increased cost of
cell splitting), hence discouraging aggressive volume expansion. Nevertheless, data’s initial failure to
drive expanding ARPUs for DoCoMo provides a negative benchmark for European investors.
MO BILE NET WO RKS - W. E UROPE 16 O CT OBE R 200 2
15
IN DUST RY D YNAMI CS
Chart 3 : NTT DoCoMo ARPU trend (yen) Chart 4 : Voice/Data ARPU trends CY Q2 02/Q2 01
14,000 CY Q2 02/Q2 01
4%
12,000
2%
10,000 0%
8,000 -2%
-4%
6,000
-6%
4,000
-8%
2,000
-10%
0 -12%
1997
1998
1999
2000
2001
Q1 02
Q2 02
Q3 02
Q4 02
Q1 03
Q2 03F
Q3 03F
Q4 03F
-14%
VOD - UK VOD - VOD - Italy DoCoMo
Germany
ARPU (12m, contract) VARPU DARPU
ARPU VARPU
Subscriber acquisition costs/subscriber retention costs (SAC/SRC) historically have SAC/SRC could increase
between 100-300bp as a
represented 7.0%-45% of an operator’s top line. The top end of this range has
percentage of sales
been provided by the least mature new entrants throughout Europe. Typically, for a
more mature operator, we see SAC/SRC running at 7.0%-20.0% of sales. A major
risk to margins is the cost of (1) acceleration in the handset replacement cycle (ie
volume) and/or (2) increasing handset subsidies (ie unit cost).
(1) Volume
Our views on the prospect of acceleration in the European handset replacement We suspect an increase in
churn is inevitable
cycle are well known. We have long argued that an increase in churn is inevitable,
if the European operator community is to achieve material data-related revenue
growth in the medium term. Our margin thesis has revolved around the concept of
technology migration (stimulating the migration of exiting 2G subscriber to next-
generation terminals).
We recognise that the speed of migration will vary on a country-by-country basis, Speed of migration will vary
on a country-by-country
reflecting:
basis
We continue to believe that Spain and France will be the most benign markets in
terms of competition, with the UK confirming its reputation as the most competitive
market in Europe. Italy, currently a benign market, is a wild card that could see
much fiercer price-based competition looking forward.
In our view, the key metric on which investors should focus to assess our volume Churn becomes a key driver
of margin
thesis is gross additions. For the purpose of simplicity, our modelling of costs
includes “internal migrations” as external churn. While this is technically incorrect,
it does enable us to capture the volume driver of costs. Should the number of the
gross adds accelerate (all other factors held equal), operating costs will increase as
a percentage of sales. We highlight the structural shift in growth that Europe has
(and will) continue to undergo away from “new” growth, to the size and share of
the disconnection pool, driven by churn and SOGA (see following chart).
45 16% 35%
40 29%
30%
5.9
35
25% 25%
23% 24% 25%
30
25 19% 20%
14% 15%
20 15%
34.4
15
12.0 10%
3.5 15.2 15.7
10 14.2
2.1
5%
5 1.4 1.4 9.3
5.7 7.3
2.0 2.5 3.7 4.7 4.0
0 0%
1996 1997 1998 1999 2000 2001 2002 2003 2004
Net Adds (m) Disconnections (m) Churn %
Source: RegTP/Group3G
Our modelling incorporates full detail on the trend in gross adds across Europe. A Tracking gross adds crucial
copy of this model is available upon request. Investors requiring further detail of
our views on technology migration should refer to Vodafone – The cost of capital
challenge, dated 8 April 2002.
Our IT Hardware team has suggested an initial unit cost range of €600-700 per 3G Our IT Hardware team
handset. To put this in some context, existing unit costs for 2G handsets in Europe suggests that 3G handsets
range between €250 and €300. (Globally, costs range between €100 and €110, will cost €600-700
reflecting the low-end 2G handsets sold into markets such as China). While we
would expect this cost to fall as scale production commences, it could result in a
sharp short-term spike to subsidies.
Average subsidies at the last reported date have been running €24-350 per Post-paid subsidies
addition. At the (all-important) post-paid level, SACs have been €156-350 currently sit at €156-300
(excluding the Italian experience; see following table). Post-paid subsidies over the
past year have diverged between (1) falling subsidisation of lower-end handsets
and (2) accelerating subsidisation of higher-end handsets (eg the Ericsson T68i).
Italy 35 35 35 3.0% na na na na na na na na
Using a unit cost of (say) €650 would result in a material spike in subsidies if the
operator’s objective is to retail the product below €300 (about £200). This could Average subsidies could
have a profound impact on margin. The implied subsidy would grow to about €350 materially increase to make
3G handset prices attractive
per handset. This is significantly ahead of existing average post-paid handset
to the end user
subsidies, although we note that it is more in line with existing high-end handset
subsidies. The actual margin impact would be driven by the number of gross adds
(volume) connecting to 3G. Our central case is for a 100-300bp reduction in
profitability (purely from an increase in SG&A as a percentage of sales).
A key risk to a positive outlook is that aggressive subsidisation does not result in
significant increases in ARPU (ie value destruction occurs). This risk is most
pronounced in markets such as the UK and Germany (in Europe) given the lower
level of industry concentration and the scope for the present equilibrium to be
destabilised by aggressive new entrants.
While the operator community claim this “success-driven” capex is already built
into its models, we are sceptical given the lack of visibility on (1) MMS subscriber ... although capex would be
targets and (2) existing network capacity utilisation in major urban areas. Hence success-driven …
Hence we encourage investors to sanity-check claims that MMS (or data) is the
silver bullet for the European wireless community. In our view, a fresh Investors should demand a
more disciplined approach
management approach is required. We suspect that operator business models (and,
to capital investment
for that matter, analyst forecasts) based on material contributions from mobile data
in the short term could be fundamentally flawed.
We are not writing off the medium- to longer-term prospects of mobile data. On the Mobile porn, sport and
gambling could still drive
contrary, we recognise the potential for data (as a longer-term driver) to surprise
‘niche’ usage
on the upside. In our view the risk look finely balanced. We can use a basic
portfolio analysis tool to consider the risks.
In our view, mobile data is a highly attractive market, but in an area of low
business strength (see chart below, top right quadrant). Bulls of mobile data could
categorise it as part of “developments,” recently developed services that may have
some future, but require greater investment to achieve that future. More bearish The biggest risk (already
commentators (like ourselves) could categorise data as part of “ego trips,” or mostly discounted in prices)
is that mobile data turns
services that have strong product champions among influential managers, but for out to be an ego trip
which there is little proven demand in the marketplace. The company, because of
the involvement of powerful managers, continues to pour resources into these new
services in the hope that they eventually come good. Hence the risks, today,
appear finely balanced between data following the “Death cycle,” down to becoming
a failure or moving across and then downwards via “Tomorrow’s breadwinners” into
“Today’s breadwinners.”
Business strength
High Low
High
Tomorrow’s Developments
breadwinners Sleepers
Ego trips
Today’s Failures
breadwinners Yesterday’s
breadwinners
Low
Given this outlook, we believe that investors should demand a more disciplined Investors should demand a
more disciplined approach
approach to capital investment, until there is greater visibility that data is not a
to capital investment
classic “better mousetraps nobody wants” but actually offers scope to create value
where none existed before.
Introduction
In this section of the note we present:
■ a sensitivity analysis on the scope for voice substitution by call type and the
impact on ARPU; and finally
Background
The average European wireless subscribers still only use their handset for about 3.0 European wireless users
still only use 3 minutes a
outgoing minutes a day, or 4.5 minutes including incoming minutes of use (MOU).
day
Across Europe, mobile calls represent 10%-15% of total outgoing call volume. In
the UK there are seven times as many minutes carried on fixed-line networks than
on the UK’s wireless networks (see following chart).
300
250
200
150
100
50
0
1999 2000 2001
Fixed local calls Fixed Internet calls Fixed long distance calls
Mobile outgoing calls Fixed international calls Fixed to mobile calls
At the top line, we believe that Europe’s wireless landscape will be characterised by
three major phases during the next four years, namely:
At the top line, the wireless
■ Phase one: prepaid to post-paid subscriber migration (2002-2006); landscape will be
characterised by three
■ Phase two: material fixed-to-mobile call volume substitution (2004-2006); and major phases
Prepaid subscribers as a percentage of the total European base appear to have In Europe, prepaid
subscribers have peaked as
already peaked. Looking forward, we expect four key drivers to cajole Europe’s pre-
a percentage of total
paid subscribers into migrating: (1) handset subsidies, (2) dealer incentives, (3) subscribers
product exclusivity and (4) bucket pricing plans (value for money).
■ In terms of current period (x+0) post-and prepaid ARPU, we have used £40 and
£9, respectively.
In the following chart we show the evolution of subscribers by post-paid plan type We have grouped each
years connections by a plan
(post-paid subscribers joining high-value, Phase One, Phase Two and Phase Three
type, to model migration
plans).
10.0
9.0
8.0
7.0
Subscribers (m)
6.0
5.0
4.0
3.0
2.0
1.0
0.0
X+0 X+1 X+2 X+3 X+4 X+5 X+6 X+7 X+8 X+9 X+10 X+11
Time (Yrs)
The chart shows the highest value post-paid subscribers (the bottom area) falling in Incrementally we expect
post-paid ARPU to fall
absolute numbers over time. Each year’s connections are assumed to connect to a
new type of plan (Phase One in x+1, Phase Two in x+2 and Phase Three in x+3).
In aggregate, total post-paid subscribers grow to nearly 60% of the total base from
31%. We have then factored in the different (declining) incremental post-paid
ARPUs for new connections (see following chart).
30.0
20.0
10.0
0.0
X+0 X+1 X+2 X+3 X+4 X+5 X+6 X+7 X+8 X+9 X+10 X+11
Time (yrs)
High Value Contract Phase 1 Contract Phase 2 Contract
Phase 3 Contract Total Contract Total Blended
Together these two charts provide an expected blended ARPU growth rate. Our Together these drivers
provide blended ARPU
analysis shows blended ARPU improving by between 4%-8% on an annual
growth of 4% on an annual
basis to year-end 2005 (see chart above). In the short term, this is the most basis
important driver of growth.
Given the sensitivity of the analysis to the rate of churn, the percentage of gross
adds connecting to post-paid tariff plans and post-paid ARPU decline, a full copy of
generic model is available to investors in the appendix of the note.
UK 80 60 1,331 111
Italy 60 58 1,040 87
Spain 42 39 1,065 89
France 77 73 1060 88
Italy 60 52 1,157 96
Netherlands 25 27 950 79
Finland 10 11 849 71
In the following table we have summarised the multiple of US minutes per line by In aggregate call volume in
the US is between 2x and
call type relative to the European average by call type per line. The evidence makes
6x the European average
stark reading. For the wireless industry we do not believe that the difference in
usage is primarily attributable to differences in penetration.
Our thesis is simple – in the event of subscribers moving from pre-to-post-paid Should pre-to-post paid
migration occur blended
plans (with large bundles), usage should increase. Evidence to date supports a
spend should increase
positive view. We highlight TEM’s Q202 results, which indicated that average usage
from migrated subscribers jumped more than 2x. Applying the US propensity to call
to each of Europe’s major markets results in a significant shift in European wireless
outgoing volume.
Readers should recognise that we have not incorporated this volume increase into Material wireless growth
may not necessarily
our bottom-up models for each operator. We merely present the data as a further
terminally damage wireline
perspective. This is important from a wireline perspective, given the scope for the volume but ...
wireless subsector to gain (in terms of volume) from pre-to-post-paid subscriber
migration, but not entirely at the revenue expense of the wireline operators.
2G vs 3G economics
Increasingly, we suspect that consensus models of revenue growth are The assumption that 2G
mobile economics will
fundamentally flawed, assuming 2G economics will underpin European 3G. We
prevail in 3G could be
reject the view that blended ARPU will gently tick along at 3-5% growth pa, coupled fundamentally flawed
with margin progression and capex/sales falling to sub-10%. In our view, European
cellular models that follow this trend ignore the potential capacity for 3G to enable
the mass substitution of fixed-line voice minutes to the wireless network.
Chart 7 : 2G vs 3G economics
2x-5x
High margins Lower margins
Elasticity of demand
Historically, a great deal of telco equity research has been dedicated to the
discussion of elasticity of demand, the percentage change in quantity demanded for
Elasticity of demand ranges
a percentage change in price. Analysis of elasticity of demand is obscured by
from negative 0.6 to
penetration growth. The authors of this study have seen estimates ranging from negative 0.9
negative 0.6 to negative 0.9, suggesting a high degree of elasticity at historical
price levels (implying volume growth somewhat compensates for price declines).
Cross-elasticity of demand
From a theoretical perspective, the key determinant of elasticity of demand is the
... but the key is cross
availability of substitutes, which is a key point missed by many industry experts. A elasticity of demand
discussion of elasticity of demand for one good in isolation is flawed.
6.0
5.0
4.0
3.0
2.0
1.0
-
1996 1997 1998 1999 2000 2001
Euro average
France
In determining revenue, we can either include or exclude access and connection In France, the multiple has
fallen to nearly 3x
charges. Choice of methodology has a significant impact on the implied multiple of
cellular prices to fixed line. For example, in the French market, including access
fees, we calculate a multiple of around 2x vs a relative call price of nearer 3x
excluding access fees. In terms of the relative price of fixed voice calls to a mobile
voice calls, the French market offers a decent proxy for Europe in aggregate (see
following chart, including fixed-line access).
Chart 9 : France: mobile. price per minute relative to fixed line price per
minute (inc. fixed line access)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
-
1999 2000 2001E
The implied multiple of fixed to mobile calls and mobile international calls are close If cellular becomes the
primary point of local
to parity. Should the cellular handset become the primary point of local access for
access, we see significant
the customer, we would view both of these segments as having a very high scope for displacement
propensity to migrate towards mobile.
Convenience
We believe that a 50% premium for the convenience of mobility will be accepted. In
our view, the overwhelming convenience and privacy of the mobile phone (relative We think customers will be
prepared to pay a 50%
to fixed) should not be underestimated as a driver of growth. We view this driver as
premium for the
further justification for increased migration from fixed to mobile usage. Remember convenience of mobility
a 50% difference at (say) 1p increases the spend to 1.5p, for the convenience of
mobility, so for a two-minute call 1p more is spent.
At 1Q02, UK mobile prices were around 3.6x the corresponding fixed line price (see
following table, average fixed line price per minute £0.04, average cellular price per
minute £0.15). To the end of Q102, the UK mobile market generated £1.8bn in
outgoing call revenue, corresponding to a monthly ARPU of £13.2. This includes
connection, access and call charge revenue.
To get the implied multiple of cellular to fixed line down to 1.5x (a 50% premium to
the underlying fixed line price), we calculate cellular prices must fall by about 60%
(1.5 x £0.04 = £0.06, which is a 50% premium to the fixed price, and a discount to
the current cellular price per minute of about 60%). Excluding any positive revenue
impact from increased volume, we calculate a revised ARPU of £5.6 per subscriber
(see following table).
OFTEL tells us that in Q102, total UK non-call revenue was about £1.1bn. Again we When assessing relative
prices, we must consider
have assumed none of this revenue migrates to the wireless network. However,
connection and access fees
methodologically we believe it would be incorrect to ignore the access fee for
comparison of fixed line prices with cellular prices. Hence at a theoretical level we
have to allocate this non-call cost to the fixed line, to get a feel for the price per
fixed-line minute the customer actually pays, when comparing the average fixed
line price per minute with the average cellular price per minute.
The reason we must include the access fee for price comparison is that the fee paid
by the customer is regular in nature, and is not a sunk cost for the customer.
Additionally, as the number of minutes originated on wireline networks diminishes,
the implied price per minute rises, and the value for money proposition
deteriorates. While we have ignored the subject of fixed-line disconnection, we do
recognise the risk displacement of originated minutes presents to the existing
number of fixed lines without DSL emerging.
Including access fees, we estimate that the implied multiple (of cellular price per
minute vs fixed price per minute) is about 1.5x (ie cellular at a 50% premium to
the underlying fixed line price). Excluding access fees from the relative price
comparison, the price multiple between the two platform grows from 1.5x to 2.6x.
We can calculate this from total fixed call revenue post-substitution excluding
access of £1.5bn, divided by the number of minutes equals an average price per
minute of £0.02, which is 2.6x less than the implied price per mobile minute of
£0.06 post price decline.
Including access fees, total fixed line quarterly revenue equalled £2.5bn in Q102,
divided by the number of minutes equals an average price per minute of £0.04.
This is 1.5x less than the implied price per mobile minute of £0.06 post price
decline.
Our central case for volume migration is based on the following assumptions:
■ 60% of all F2M calls substituting to wireless; Our central case assumes
mobile grabbing around
■ 50% of all national calls substituting to wireless; 35% of total originated call
volume (from 13%)
■ 30% of all international calls substituting to wireless; and
This equates to 35% of UK call volume orignated by wireless networks from 13%.
Total call volume in Q102 equalled 97.3bn minutes, of which 13% were mobile-
originated. Our central case grabs an additional 22.2bn minutes, giving total
mobile minutes post price decline of 34.5bn minutes, or 35% of the total market
minutes for the quarter.
We are not suggesting that call volume at Q102 offers a good proxy for call volume
in the absence of fixed-to-mobile substitution looking forward. But given the
transparency of Q102 numbers, we have used the data to get a feel for the scope
for an ARPU shift.
Similarly we don’t want to become too bogged down by the question, What’s the
correct percentage for displacement by call type? The future is by definition
unknown, and hence the question is clearly open to debate at a bottom-up level.
In our view it is more important to focus on the bigger picture – What percentage
of total volume will migrate?, and What price reduction is required to drive this
displacement?
Factoring in a willingness to pay a 50% premium to the underlying fixed line price
in aggregate, we estimate conservatively outgoing voice ARPU growth of 18% This corresponds to
(outgoing voice ARPU £5.6 post price decline, from £13.2, adding incremental ARPU outgoing ARPU growth of
18% (using a price cut of
uplift from volume growth of £10.0, resulting in post price decline ARPU of £15.60,
58%).
which is 18% ahead of £13.2). Outgoing growth of 18% corresponds to 11%
change in overall ARPU. This would represent a significant upside surprise to
consensus expectations. We note comment from British Telecom that to date about
4% of existing volume pa has been cannibalised by mobile.
DYNAMI CS
Call volume bn mins 18.8 3.5 13.6 2.0 47.1 85.0 12.3 97.3
Call revenue (exc. access) £ bn 0.4 0.5 0.4 0.3 0.7 2.3 1.8 4.1
Implied revenue per minute 0.02 0.14 0.03 0.14 0.01 0.03 0.15 0.04
Impact of Subsitution
Mobile price per minute reduction -58%
MO BILE
Revenue lost to Fixed Line 0.1 0.3 0.2 0.1 0.1 0.8
Revenue substituted to mobile 0.3 0.1 0.4 0.04 0.4 1.4 1.4
Mobile Revenue lost due to price cut 1.0
Incremental ARPU uplift 2.5 0.9 3.1 0.3 3.2 10.0 10.0
End Game
-
Call volume bn mins 13.2 1.4 6.8 1.4 40.0 62.8 34.5 97.3
W.
Call revenue (exc. access) £ bn 0.3 0.2 0.2 0.2 0.6 1.5 2.1 3.6
E UROPE
Implied revenue per minute 0.02 0.14 0.03 0.14 0.01 0.02 0.06 0.04
Pre-price decline Post price decline
ARPU 13.2 5.6
Incremental ARPU uplift (from substitution) 10.0
New ARPU (post price decline inc. volume uplift) 15.6
% Change 18%
Difference
Total fixed line call volume 85.0 62.8 22.2
Total mobile call volume 12.3 34.5 22.2
16
97.3 97.3 -
O CT OBE R
32
MA CRO DYNAMI CS
UK local calls
Factoring 30% volume substitution at an average cellular price of £0.06 results in 30% local call volume
substitution equates to an
revenue for the quarter of £0.3bn substituting to mobile from fixed line. This
additional 41 MOU per
corresponds to 5.6bn minutes of fixed-line local voice calls substituting to the month
mobile network, or an additional 41 monthly MOU per wireless subscriber per
month (see following table).
Using an average price per cellular minute of £0.06 implies a premium of 2.8x the
average price per fixed-line minute (fixed-line price per minute of £0.022). This
massively overstates the actual premium the wireless user is paying because it
ignores the wireline access and connection fee that the customer pays.
UK F2M calls
Using an average revenue per cellular minute of £0.06, and assuming the
displacement of 60% of fixed to mobile volume to mobile, we see about £0.1bn of
60% F2M substitution
fixed calls to mobile moving to mobile. This amounts to about 2.1bn of additional
corresponds to 15 MOU per
minutes, or 15 monthly MOU per wireless subscriber. It is important for readers to sub per month
recognise that as the wireless handset becomes the “accepted” point of access, we
believe that a significant networking effect could occur (in terms of mobile to
mobile communication). Hence we believe that our 60% migration assumption
could prove too conservative.
UK national calls
Our central case uses 50% substitution of national calls to the wireless network. We
view the national call market as a core opportunity for Europe’s wireless operators.
On this basis, we see 6.8bn minutes substituting to the UK’s wireless operators (50
monthly MOU per subscriber), or about £0.4bn in revenue terms.
UK international calls
In our central case outlook, we see a less material opportunity from the
international call segment. We acknowledge that this could be overly conservative.
Our central case assumes 30% substitution of international call volume, or about
0.6bn minutes migrating to the UK wireless network operators (4 monthly MOU per
subscriber).
UK other calls
Other calls include Internet dial-up, pay phones, calls to premium services, calls
through the operator, calls to the speaking clock, calls to paging operators and free
phone numbers. While we view this market as “fair game” for the wireless operator
community, we have included only 15% of this traffic displacing to the wireless
operators, recognising that Internet traffic is a key driver. This corresponds to
about 7.1bn minutes (52 monthly MOU per subscriber).
UK market in aggregate
In aggregate, our central case estimates show mobile traffic growing to 35% of
total telco volume from 13%, or to 46% of total revenue (from 35%). This equates
to 162 monthly MOU per subscriber substituting from the UK wireline operators to
the UK wireless community. This corresponds to an additional 5.3 MOU per day
Together we see up to 18%
being substituted, or about 8.3 minutes a day in outgoing cellular usage in total growth in outgoing voice
(5.3 MOU plus the original 3.0 MOU). In revenue terms this equals £1.4bn of ARPU
additional revenue, or a monthly ARPU uplift of £10 per subscriber. This is 18%
growth on the “pre” price decline monthly ARPU scenario of £13.2 per
subscriber.
We can also think of this substitution in terms of ARPU contribution by call type
(see following table). Our analysis clearly shows outgoing voice ARPU increasing
to £15.6 from £13.2 (+18%).
Industry impact
We can also think about the potential of fixed to mobile substitution in terms of the
total industry, recognising the importance of the 50% premium to the underlying
fixed-line price per minute. Our central case for mobile (holding all other factors
equal) suggests that UK mobile revenue grows to £2.1bn (for Q102 factoring in the
indicated substitution), and UK wireline revenue declines by 24% to £2.5bn from
£3.3bn. For the wireline operators this does not include any compensating benefit
from DSL. In aggregate, this results in total outgoing telecoms revenue in the UK
declining by about 9% (from about £5.1bn to £4.7bn).
Table 26 : UK fixed and mobile market post 58% cellular price decline
Fixed: post-price decline End-Q102
% change 3%
% change 18%
% change 180%
We can present this impact more clearly using the following pie chart.
Mobile revenue
(pre-decline) Mobile revenue
% total 35% (post-decline)
% total 46%
£5.1bn £4.7bn
Source: ABN AMRO
A word of warning
Intuitively, some readers of this study may ask why total revenue is shrinking. If Without DSL growth,
telecoms shrinks
UK wireline loses more than 22bn minutes to wireless, and UK wireless customers
are prepared to pay a 50% premium to the underlying fixed line price, then
(surely) the industry should grow (not contract). This is (of course) correct,
excluding the original UK wireless revenue (see following chart). Total fixed falls to
£2.5bn from £3.3bn, mobile adds £1.4bn, resulting in total revenue of £3.8bn
(from £3.3bn). But mobile also loses £1bn from its historical revenue stream.
3
2.5
2.5 2.2
1.4
£.bn
2 1.8 1
1.5
1 0.8
0.5
0
Total F.L..R Total F.L R Total F.L.R Mobile Mobile (Lost) Mobile Mobile Rev. Added Mobile Rev
(Pre-Decline) (Lost) (Post-Decline) (Pre-Decline) (Post Decline) (Post-Decline)
Sensitivity analysis
In the following table we have presented a range of scenarios for outgoing voice We have also run a
sensitivity analysis on the
ARPU based on our analysis of the UK market at Q102. This enables investors to
price change required to
take their own view of the scope for traffic substitution, for a given price reduction drive migration
and identify potential outgoing voice ARPU. For example, assuming a (say) 55%
share of traffic and 58% price reduction suggests outgoing voice ARPU shifts to
£24.4 (+84%).
Volume share 15% 4.1 5.5 6.8 8.1 9.4 10.8 12.1
Volume share 15% -69% -59% -49% -39% -29% -19% -9%
Using the same example,
25% -36% -26% -16% -6% 4% 14% 24%
we can observe 84%
35% -2% 8% 18% 28% 38% 48% 58% growth on the pre-decline
45% 31% 41% 51% 61% 71% 81% 91% ARPU
In our view, an analysis of existing call volume does not fully capture the potential
of the volume opportunity for the European wireless operator community. Why?
■ Mobiles are not generally shared and are most usually carried on the person, Should the wireless handset
become the primary access
which provides an opportunity for a networking effect of increasing usage.
point, large scale growth in
the size of the pie could
■ Our core thesis is that high prepaid charges have stunted European cellular
occur
usage.
Hence our earlier assessment of the percentage of fixed-line call volume that could
displace to mobile could fundamentally underestimate the overall size of the
wireline telephony opportunity. In other words, the total size of the voice
telephony/broadband data market could explode (see following chart).
20,000
15,000
A classic bull market chart,
10,000 or maybe realistic upside?
5,000
-
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Total Fixed Narrowband Data Total Fixed Narrowband Voice Total Fixed Narrowband
Total Broadband Data Total Broadband Voice Total Broadband
Total Mobile
■ Phase Two: an 18% structural shift in outgoing voice ARPU by 2006 (holding all
other factors equal 11% overall ARPU growth)
30
25
20
15
European termination rates
10
remain at a signifiacnt
5 premium to other major
markets
0
US
Korea
Luxembourge
France
Denmark
Greece
Germany
Norway
Sweden
UK
Ireland
Austria
Netherlands
Portugal
Spain
Italy
Finland
Switzerland
Belgium
In Regulation – Friend or Foe? dated 16 August 2002 we ran a detailed sensitivity Lower quality (primarily)
new entrant businesses are
of various operators’ revenue to reductions in termination rates. Simply put, all
most exposed
operators have material exposure – but new entrant operators have
disproportionately higher exposure. Ironically, as many regulators aim to promote
competition, they cannot cut rates too aggressively. We can assess the impact of
cuts to termination rates with reference to the UK.
A UK example
In the UK, OFTEL has suggested that RPI negative 12% should be applied to all UK The UK (again) offers an
excellent benchmark for
operators’ termination fees. Historically, UK operators have been subject to RPI
understanding the
negative 9% for fixed line calls to wireless only. The extension of the mandate will economic impact of cuts
encompass about 48% of total incoming volume from 21% under the previous
regulation (see following tables).
We can work through the impact of RPI-12% using OFTEL data. First, we know Using OFTEL data we can
work through the proposed
from OFTEL that mobile outgoing calls amounted to 12.3bn minutes in Q102 (for
RPI-12% cut
the quarter, see following table).
Roaming 233 2%
Source: OFTEL
Of these minutes we know the percentage split between fixed calls, and on-and off- OFTEL tells us the split
between fixed, on- and off-
net mobile minutes (see following table).
net mobile minutes
Other 9% 1,110
100% 12,330
We know that total interconnection volume amounted to 6.1bn minutes. Using the We know that total
interconnection minutes
data above we can imply about 1.3bn minutes attributable from the fixed line (see
amounted to 6.1bn minutes
following table). (Q102), implying 1.3bn
minutes attributable from
the fixed line
Table 31 : UK interconnection volume
Jan - Mar 2002 % split
We also know that quarterly interconnect revenue of £672m was generated during Using an inflation rate of
2%, and RPI-12% suggests
1Q02, and that £361m was attributable to off-net mobile. Hence factoring in a
a 2.4% reduction in
2.0% RPI rate (ie a 10% price cut), suggests a 2.4% reduction in total quarterly quarterly revenue, which is
revenue (24% of revenue attributable to interconnect, multiplied by the percentage in the noise
Given much of the doom and gloom surrounding the issue of termination rates, it is We put 60% probability on
the RPI-12% ruling being
important not to lose sight of the minor impact at the top line of RPI-12%. We put
confirmed
a 60% probability on the RPI-12% cut being endorsed by the UK’s competition
commission. If this were confirmed, we would view it as an upside surprise to the
market’s current expectation.
Increasingly we think investors should question this analysis that excludes volume … but remind investors of
the potential for increasing
uplift. Focusing on the impact to the profitability of a reduction in termination rates
volume to at least partially
without any compensating volume increase is appealing (due to simplicity), but it offset losses
ignores the possibility that growth in volume could occur, potentially compensating
operators for the loss in price.
Impediment to substitution
We recognise that the one of biggest impediments to potential substitution is the Ironically, cuts in
termination rates could r-
high rate of mobile termination fees throughout Europe. We expect termination
focus management on
rates will continue to decline at a double-digit rate for the foreseeable future. fixed-to-mobile substitution
Ironically, cuts in termination rates could refocus management on the fixed-to-
mobile substitution opportunity. Hence we believe falling termination rates are key
to enabling growth in outgoing voice ARPU.
The US experience
While there is a weight of evidence showing that cuts to terminations rates do
materially affect short-term profitability in markets with relatively high levels of The US model suggests
termination (see Portugal as an example), we believe there could be a silver lining there could be a silver
to the cloud. In our view, the US cellular market provides a benchmark that could lining to the cloud ...
provide an indicator of Europe’s future. US blended ARPUs are more than 50%
higher than UK ARPUs (see following chart).
90% 40.0
78% 81%
80% 35.0
67%
70%
30.0
Penetration (%)
60%
50% 25.0
ARPU (£)
50% 45%
40% 39% 20.0
40%
31%
25% 15.0
30% 23%
20%
20% 14% 10.0
10% 5.0
0% 0.0
1997 1998 1999 2000 2001 Q1 02
Year
We estimate that approximately 60% of all US traffic is outgoing. New entrant US incoming volumes are
materially higher than in
players, such as VoiceStream, generally have a greater reliance on incoming
Europe, as a result of lower
volumes (about55% of traffic is outgoing). On a per minute basis, US termination prices
rates average about €0.005. The key driver of US incoming ARPUs is higher
volumes (see following table). Our thesis is that European incoming ARPUs
(and usage) have been stunted by excessive charges.
The problem Europe’s operators face is to manage the decline in rates without
The key challenge in Europe
significantly damaging returns. The danger is that in the event of relatively small
is to manage declining
cuts to termination rates, demand fails to materialise. This is akin to the firm that termination rates without
gets caught in the middle of the road (between excessively high termination rates damaging returns
and rates similar to the fixed line operator community) – getting run over!
While declining termination rates will create a “tough” top-line growth environment, Using the US as a
benchmark suggests upside
given the US experience, it is plausible to argue that a significant decline could be
to bearish market
proportionately offset by an increase in volume. perception
Wireline spend ranges between 1.2x and 2.3x wireless spend as a percentage of Wireline spend ranges
between 1.2x and 2.3x
GDP. Hence in aggregate, telecoms spend accounts for 2.2%- 3.3% of GDP. In the
wireless spend
UK, telco spend accounted for around 2.4% at year-end 2001 (see following table).
This can be compared with the higher share of GDP that is observable in the US. US wireless share of GDP is
particularly impressive
Wireless share of GDP is particularly impressive given the materially lower rate of
given the lower rate of
penetration (c.50% vs Europe c.81%). penetration
Despite intense competition, US operators have maintained high ARPU levels by not
targeting lower-value prepay subscribers, and through the introduction of “bucket
plans.” US operators have increased the number of free minutes of service, rather
than reduce the monthly access fee. In Europe, “tariff plans” have resulted in
significant ARPU dilution, and higher margins but lower pre-capex cash flow per
subscriber. In our view, the bucket pricing plan offers superior long-term returns to
Europe’s tariff plan model. As a result, we believe that Europe will move towards
flat rate access fees. We highlight that success can breed failure as the historical
success model becomes the major obstacle to a firm’s adoption of the new reality.
The introduction of the “your plan” flat rate fee by Orange in the UK provides early-
stage validation of our view. In addition we flag that an analysis of VoiceStream,
the aggressive US new entrant, provides a benchmark for understanding the
potential impact of Hutchison in Europe, and the domino effect we forecast
throughout the European wireless landscape.
Introduction
In this section of the note we present:
The US experience
US operators, while reporting falling historic ARPU, have consistently reported US operators have
consistently reported
higher and more stable ARPUs than their European counterparts. We attribute this
higher and more stable
to two key drivers, post-paid focus and bucket plans. ARPUs than their European
peers
Post-paid focus
US operators have not aggressively targeted lower-value prepay subscribers. This
is easily identifiable from a comparative analysis of subscriber mix by geography
(see following chart). In the US, this product is unattractive because prepaid
customers would also have to pay for incoming calls.
89%
15% 79%
75%
69%
28%
60%
35%
33%
53%
46%
48%
85%
72%
65% 67% 97% 91%
52%
9%
3%
Total 50 48 29 58 37 62 130
Subs (m)
Pre-Pay Post-Pay
US vs UK blended ARPUs
Blended ARPUs in the US are materially higher than in Europe. To December 2001, US monthly industry ARPU
over 50% higher than the
in the US monthly ARPU stood at £29.7, compared with £19.6 in the UK, ie over
UK industry average
50% higher (see following chart). At 2Q02, Vodafone was the highest ARPU
operator in the UK (£23 per month) vs monthly ARPU at Nextel [NXTL-US$9.20,
N/R] of £48 and AT&T Wireless [AWE-US$4.94, N/R] of £43 per month. This is a
direct function of the higher-volume/lower-margin business model that has been
adopted (in marked contrast to the established European model).
90% 40.0
78% 81%
80% 35.0
67%
70%
30.0
Penetration (%)
60%
50% 25.0
ARPU (£)
50% 45%
40% 39% 20.0
40%
31%
25% 15.0
30% 23%
20%
20% 14% 10.0
10% 5.0
0% 0.0
1997 1998 1999 2000 2001 Q1 02
Year
Our time series analysis clearly shows that the US market increased penetration by Differences in ARPU
between the US and the UK
over 30% between 1997 and 2Q02, without material dilution to ARPU. This can be
are not primarily
contrasted to the UK’s performance. The difference in ARPU between the two attributable to penetration
markets is not entirely attributable to differences in penetration. This can be quickly
demonstrated by looking at US monthly ARPU at year-end 2000, relative to UK
monthly ARPU at year-end 1999, with similar levels of penetration. Hence our
thesis is that average European cellular MOUs/ARPUs have been stunted by tariff
plan pricing. Looking forward, we believe the introduction of higher flat rate access
fees, with a higher level of “free” minutes (bucket pricing plans) should have a
profound impact on ARPU.
700
600
500
400
Minutes
300
200
100
0
Q1 00A Q2 00A Q3 00A Q4 00A Q1 01A Q2 01A Q3 01A Q4 01A Q1 02A
Period
VoiceStream US Industry Average UK Industry Average
Following AT&T Wireless, VoiceStream introduced more aggressive bucket pricing Two main types of bucket
plans
plans into the US market, creating a domino effect. US national operators offer two
main types of bucket plans, which generally provide more minutes than most
subscribers can consume. This is a marketing ploy used particularly well by
VoiceStream to make the subscribers feel they are receiving greater value for
money.
■ National one-rate plans eliminate long distance charges. These plans consist
of a monthly fee that includes an allocation of “anytime minutes,” off-peak
minutes and one standard overage rate. It is difficult to compare plans across
operators because some, like VoiceStream, charge extra for off-network
roaming, while others include it in the price.
■ Regional plans offer users free minutes in their region (multi-state, not
national) but typically charge for roaming out of one’s region and for long
distance minutes. This is the plan that has given VoiceStream the most publicity
– ie 3,000 “anytime” minutes for $59.99 – but accounts for less than 20% of
new customers.
Recent pricing plans announced by the big six operators are very similar as each VoiceStream the most
fights for market share. VoiceStream has been the most aggressive US operator, aggressive US operator
setting industry-low price points for both national one-rate and regional plans.
The evidence available from the US market model to date suggests that new 80% of new subscribers are
choosing national plans
subscribers are choosing national instead of regional plans. Customers are using
these national plans to replace their landline phones for long distance calls, but still
using wireline for untimed local calls.
Included in any
AT&T Wireless $0.10 $199.99 2,000 $0.25 NA time minutes
Most of the national operators have a national rate plan “anytime” yield per minute
of USD $0.07 (off-network roaming extra), following VoiceStream’s aggressive
pricing. Similarly, the established operators have increased their number of free
US national rate plans have
off-peak minutes (following VoiceStream), but none have matched VoiceStream’s an ‘anytime’ yield of
unlimited weekend minutes. As the US operators have experienced increasing US$0.07
capacity constraints (a function of rising usage and penetration), operators have
introduced a slight shift in off-peak times for new customers ie pushing back the
start time for off-peak to 9pm from 8pm, to manage network congestion.
Many investors flag the low yield per minute ($0.02) generated by VoiceStream’s Points to consider about the
3,000 ‘anytime’ minute
controversial 3,000 “anytime” minute regional plan as being an unsustainable
regional plan
business model; however, the following points must be considered.
■ This plan was introduced in 3Q01 primarily as a marketing tool by VoiceStream Little impact on margins –
only 20% of customers
to reinforce its image as being the best value wireless provider in the US. The
choosing this plan
actual effect on VoiceStream’s network capacity and margins is limited as less
than 20% of new net adds choose this plan.
■ On average, only 1,000 (16.6 hours per month) of the 3,000 minutes (50 hours Only one-third of minutes
used, generating same
per month) are actually used, implying a real yield of $0.06 per minute, which is
actual yield as national call
almost in line with the competition and not dissimilar to the $0.07 yield plans
generated by national call plans.
Spare network capacity in each market is a function of spectrum and the number VoiceStream is the smallest
national operator, but has a
of subscribers on the network. In the US, VoiceStream is by far the smallest
similar amount of spectrum
national operator in terms of subscribers, but has a similar amount of spectrum in in the top 50 markets
the top 50 markets as its competitors, and therefore has greater spare capacity to
carry additional minutes on its network (see following chart).
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
Verizon Wireless Cingular AT&T Wireless Sprint PCS Nextel Voicestream
Wireless Communications
Additionally, all of VoiceStream’s network is based on GSM technology, while some VoiceStream’s network
technology is more
operators such as AWE and Cingular are still partially using less capacity-efficient
capacity-efficient
TDMA. Cingular’s chief technology officer has stated that upgrading to GSM/GPRS
by 2004 will increase voice capacity on Cingular’s existing network by 120%, while
at the same time requiring 18% less spectrum.
0.30
0.25
0.20
Yield (USD)
0.15
0.10
0.05
0.00
Q1 00A Q2 00A Q3 00A Q4 00A Q1 01A Q2 01A Q3 01A Q4 01A Q1 02A
Period
observed), US ARPUs are substantially higher than in the UK. Hence lower yield in
the US does not result in lower ARPU. This is particularly pronounced at the
VoiceStream level. The key difference is volume.
US invested capital
The reason VoiceStream has been able to price so aggressively relative to its A more capital-light
network enables a lower
competition is its lower cumulative capex spend per covered pop (see following
unit operating cost
chart). This is a very important point. Intuitively, a more “capital-light” network proposition
requires a lower absolute level of net operating profit after tax to generate a cost of
capital return.
120
100
80
US$ m
60
40
20
0
Voicestream T-Mobile Nextel Com. Orange Vodafone MMo2 AT&T Sprint PCS
Wireless
Our analysis also highlights the lower cost proposition of T-Mobile and Orange in In terms of cumulative
capex per covered pop,
the UK relative to (say) both Vodafone UK and mmo2 UK. This is intuitive, given
there is no discernible
that the UK’s established operators (Vodafone and mmo2) have deployed their difference between the US
networks over 20 years, relative to about eight years for the new entrants. Hence and European model. The
key driver is length of
the new entrants have benefited from falling kit costs. The same relationship is
service
observable between AT&T Wireless (established operator) and VoiceStream (new
entrant) in the US.
80.0
60.0
40.0
20.0
(Euro)
-
-20.0
-40.0
-60.0
-80.0
Vodafone - Germany
Orange - UK
MMo2
Vodafone - UK
Viag
Sprint PCS
Eplus
Cingular
AT&T Wireless
VoiceStream/Powertel
T-Mobile - UK
T-Mobile Germany
Verizon Wireless
ABN AMRO does not cover Sprint PCS.
Source: ABN AMRO & Company reports
Higher pre-capex cash flow per subscriber supports our thesis that the US model
could be superior to the existing European model. Unfortunately, it is difficult to Higher pre-capex cash flow
per sub supports our
compare absolute cash flow per subscriber between the US and Europe. This is a positive thesis
function of three key factors.
If our thesis is correct, we must address why US cellular assets trade at such a
material discount to their European peers. We attribute this to four easily
identifiable factors.
■ A structurally more competitive market. We note nearly 50% of the US Weak US cellular valuations
do not affect our positive on
population has access to over six wireless operators (over 75% have access to
‘bucket pricing’ thesis
over five). The US Herfindahl index at year-end 2001 sat at 1700, suggesting
fierce competition, much lower than its European peers (eg Italy at 3729,
suggesting a high degree of pricing power).
Spare network capacity. Our analysis of Mhz per subscriber in the US market High capacity (MHz per sub)
provides scope for
offers an insight into potential pricing in Europe. Similar to VoiceStream, we expect
aggressively priced flat rate
H3G to enjoy an abundance of capacity relative to its competitors. For example in access plans
the UK, investors can replace Verizon Wireless, Cingular and VoiceStream with
(say) mmo2, Vodafone and H3G (see chart below).
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
Vodafone
One2one
Orange
BTCellnet
Technology
H3G will enter Europe using W-CDMA, relative to its competition currently
employing GSM. Similar to VoiceStream, H3G should enjoy a more capacity- W-CDMA provides a more
capacity efficient
efficient technology. We very conservatively estimate capacity increases of 2x-3x
technology
(for a more detailed discussion of capacity we refer investors to the “3G as the
facilitator” section of this report).
European yield
As part of the preparation for this note, we analysed the yield observable in the UK, Other European markets
still have very high yield
France and Italy (see following chart). The markets with the highest level of
levels
concentration ie less competition have enjoyed the highest yield (France and Italy).
This is in line with our core belief that industry structure determines longer-term
sustainable returns.
0.30
0.25
Yield (USD)
0.20
0.15
1999 Q100 Q200 Q300 Q400 Q101 Q201 Q301 Q401
Period
France Italy UK
10%
5%
0%
-5%
-10%
-15%
-20%
Q2 00A Q3 00A Q4 00A Q1 01A Q2 01A Q3 01A Q4 01A Q1 02A
Over the past two years, US industry yields have fallen by a CAGR of 20%, relative In the US, industry yields
have fallen by a CAGR of
to VoiceStream posting an annual decline of about 15%. However, we note that US
20% over the past two
yields have stabilised in recent quarters, reflecting capacity constraints. years
Cumulative capex
We understand that HWL’s original business plan budgeted for about US$10bn of
cumulative capex to year-end 2005 for its European wireless operations. We
believe that approximately 55% of this spend will be dedicated to the UK operation H3G could enjoy a similar
lowest-cost operator
(US$5.5bn), 35% to Italy (US$3.5bn) and what remains to HWL’s more minor
position
operations (eg Denmark, Sweden, Ireland, Austria etc). More recently, HWL has
indicated that the operations in Italy and the UK will benefit from infrastructure
sharing, reducing capital expenditure by around 20%, implying new cumulative
capex of €4.4bn for the UK and €2.8bn for Italy.
■ Using Companies House data, we can identify cumulative capex for each of the
UK’s major operators to year-end 2001 (with the exception of Vodafone and T-
Mobile, for which we have used an estimate of 10% of their top lines for the
period to December 2001).
■ We then assume a maintenance level of capex at 6% of each operator’s forecast We have presented a
potential cumulative capex
top line per annum to year-end 2005.
per covered pop analysis for
the UK operators
■ Finally, we take a common view on the cost of deploying a 3G network in the UK
for the established operators, conservatively assuming 50% of existing 2G sites
can be re-used (see following table).
As a % of total 65%
While this analysis is “quick and dirty,” it enables us to identify the cumulative While this is ‘quick and
dirty,’ it enables us to
capex per covered pop by operator (see following chart). Reflecting its newer
better understand the
operator status, H3G enjoys a significantly lower cumulative capex per covered potential pricing
pop. environment
180
160
140
120
100
80
60
40
20
-
mmo2 Vodafone Orange T-Mobile H3 G
As demonstrated by the pricing of VoiceStream in the US market, an operator’s Over the longer term we
would expect a movement
returns are a function of its balance sheet invested capital (of which a large
towards convergence
proportion is capex); historical cost plays an important role in framing reported
returns. To 2005, we would expect H3G to have a “lighter” balance sheet than its
established peers. Over the longer term, as the UK’s established operators replace
the capital in their balance sheet, we would expect a movement towards
convergence.
3G as the facilitator
Introduction
In this section of this note we present:
■ the potential capacity uplift from W-CDMA and additional spectrum; and
■ Covered POPS per base station. This calculation provides a rough proxy to We have presented a simple
relative capacity matrix for
compare the amount of capacity each operator has built into its footprint.
each major European
Operators with high covered POPs per base station can spend additional capex operator
on base stations to increase capacity, while operators with low covered POPs per
base station are closer to reaching maximum capacity.
■ Subscribers per base station. This calculation provides a rough proxy to the
current capacity each operator has available to service its existing subscriber
base.
Whilst our matrix does not gives us absolute capacity, in our view it provides a
good feel for relative capacity constraints across Europe (see chart below).
1,200
1,000
800 Bouygues
E-Plus
600
mmo2 (Germany)
400
Quadrant One: Quadrant Four:
High capacity built into footprint Low capacity built into footprint
200 High capacity per subscriber High capacity per subscriber
-
2,500 3,500 4,500 5,500 6,500 7,500 8,500 9,500
1H02 CPOP/Cell
■ Quadrant One. Operators in this quadrant have the most flexibility with regard Quadrant One. Excess
capacity enables operators
to medium-term capex spend because they have the greatest available capacity
to generate high returns as
per subscriber. They offer the best network access because they have already they add new subscribers
built a lot of capacity into their footprint and have low market share. As these
operators obtain greater subscribers and market share, they will move towards
Quadrant Two, and therefore report significantly higher-than-average returns on
new invested capital.
■ Quadrant Two. Operators in this quadrant are normally in highly penetrated Quadrant Two. Operators
are generating high returns
markets and are relatively capacity-constrained, requiring additional spectrum
but are relatively capacity-
or improved technology. They have already built a lot of capacity into their constrained
footprint (fewer covered POPs per cell site) but because of their high market
share have less available capacity for each subscriber (greater number of
subscribers per cell site). Subscribers will experience network congestion during
peak periods. TIM Spa fits into this category.
■ Quadrant Three. Operators in this quadrant have not yet built a lot of capacity Quadrant Three. Operators
face imminent future capex
into their footprint and have low capacity per each subscriber. Subscribers will
spend to meet existing
experience network congestion during peak periods; however, by spending subscribers’ needs
additional capex on new cell sites, these operators can improve their network
capacity. This quadrant is an indicator of high future capex. Most of Europe’s
major operators fit in this quadrant.
■ Quadrant Four. Operators in this quadrant have not yet built a lot of capacity Quadrant Four. Operators
can delay capex if
into their footprint, but have high capacity for each subscriber. If these
experiencing slow
operators are not experiencing rapid growth, they can delay capex. Mmo2 subscriber growth
Germany fits in this category.
These quadrants show that Hutchison3G could be relatively well placed compared
with its European peers, admittedly recognising its lower intended population
coverage (around 80%). When it is established, H3G’s network/footprint should be We can clearly observe
able to handle many more subscribers without needing substantial additional capex. emerging capacity
constraints for Europe’s
It is also important to note that it is not all rosy for the remaining European
major operators
operators with regard to capex spend and, in particular, capacity when compared
with the H3G’s potential. We can clearly observe the relative capacity
constraints of Europe’s most established operators, such as TIM Spa,
Vodafone Germany, Vodafone UK and Orange UK. In our view this is major
incentive for these operators to accelerate the deployment of 3G.
To have confidence in our fixed to mobile substitution thesis, investors must believe
in the prospect of an imminent material increase in available voice capacity. The
capacity of a cellular system is dependent upon several factors, including:
TDMA Capacity advantage of 3X over analog Average sound quality, more digitised, than normal human
voice
CDMA Has a nationwide footprint in the US Newest digital technology. Less experience about what could
happen with regard to the potential technology glitches or in
Has the most significant capacity gains over analog systems, of
overcapacity situations.
approximately at least 6x. Therefore, cost per minute of service
and capital expenditures could fall to low levels Initial costs to build CDMA are significantly higher, expensive
handsets
GSM Most widely used in the world Footprint still not comparable with CDMA or TDMA in the US
Proven technology Present system not entirely suited for wireless data
IDEN Only technology that can currently support 3-1 function of Limited use and support from equipment providers, handsets
dispatch, cellular and paging
Less spectrum that other players
Packet-switched based network should work well with wireless
data
Source: Company reports and ABN AMRO
Because radio spectrum is a limited resource shared by all users, a method must be
devised to divide up the bandwidth among as many users as possible. The method
used by GSM is a combination of TDMA and frequency division multiple access
(FDMA). The FDMA part involves the division by frequency of the (maximum) 25
MHz bandwidth into 124 carrier frequencies spaced 200 kHz apart. One or more
carrier frequencies are assigned to each base station.
The introduction of 3G should lead to a structural shift in spectral efficiency (see The introduction of 3G
should lead to a structural
following discussion, “The total number of cells”), enables scalability and provides
shift in capacity
additional spectrum (see frequency allocation discussion above). 3G provides the
additional capacity to deploy bucket pricing plans and drive mass fixed-to-mobile
substitution.
■ the other 13 traffic channels can be allocated in a 1/3 re-use; and We have calculated the
capacity benefit offered by
■ in total we get five carriers per sector, W-CDMA over 2G
W-CDMA capacity
We understand that the soft capacity of W-CDMA gives flexibility between
coverage, capacity and quality. As a cell sites coverage area increases, capacity
drops dramatically. We can calculate the capacity of a simple W-CDMA network
with 5MHz of spectrum assuming a cell site radius of 3.5km and one transciever,
which implies 42 voice channels per sector, with 2% blocking giving 33
Erlangs/sector.
Chart 24 : Allocated unpaired spectrum per country - pre and post 3G auctions
250
Spectrum (Mhz)
200
150
100
50
0
United States France Spain Germany Italy United Kingdom
2G Spectrum 3G Spectrum
Existing frequency
A simplistic approach to assessing an operator’s capacity is to look at Mhz per
subscriber. This approach clearly favours operators with a smaller subscriber base
In our view, smaller
(eg VoiceStream in the US, T-Mobile in the UK). It is our view that the smaller operators have a greater
operators have greater ability to service their existing subscriber base and ability to dictate pricing
therefore dictate pricing. It also shows the magnitude of 3G spectrum waiting on-
line in Europe (total MHz per subscriber shows 2G and 3G specturm, vs. 2G only
MHz per subscriber). We can contrast Europe’s relative capacity depth to
constraints in the US.
Chart 25 : France MHz per subscriber Chart 26 : Germany MHz per subscriber
8.00 9.00
8.00
7.00
7.00
6.00 6.00
5.00
5.00
4.00
4.00 3.00
3.00 2.00
1.00
2.00
0.00
Mannesmann
E2
E-Plus
DeTe Mobil
1.00
0.00
Itineris SFR Bouyges
MHz per subscriber 2G MHz per subscriber MHz per subscriber 2G MHz per subscriber
Chart 27 : Italy MHz per subscriber Chart 28 : Spain MHz per subscriber
4.00 60.00
3.50
50.00
3.00
40.00
2.50
2.00 30.00
1.50
20.00
1.00
10.00
0.50
0.00 0.00
TIM OPI Wind Telefonica AirTel Amena
MHz per subscriber 2G MHz per subscriber MHz per subscriber 2G MHz per subscriber
4.00
4.50
4.00 3.00
3.50
3.00
2.00
2.50
2.00
1.00
1.50
0.00
1.00
Wireless
Cingular
AT&T Wireless
Sprint PCS
Voicestream
Communications
Verizon Wireless
0.50
0.00
Nextel
Vodafone
One2one
Orange
BTCellnet
Source: ABN AMRO estimates ABN AMRO does not cover Sprint PCS or Nextel Communications.
Source: ABN AMRO estimates, average top 50 markets
On average, there are 6-8 competitors in each US market compared with individual The average European has
European countries of 3-4 for 2G and 4-6 including 3G licences. While the average 40-50MHz, US operators
have 10-30 MHz
European operator has 40-50MHz, US operators make do with 10-30MHz.
Frequency outlook
Clearly this leads to spectrum and capacity issues for the larger US operators in
each market. In Europe, the 1900-2100 spectrum secured is currently
When unlocked, 3G
“locked up” and is useable only over IMT-2000 compliant infrastructure. A spectrum should provide
term of the license awards prohibits the use of this spectrum over existing 2G significant additional
infrastructure. The International Telecommunications Union (ITU) has defined capacity
384kbps as the data rate limit required for a service to fulfil the IMT-2000 standard
in a pedestrian environment. The 384 kbps data rate corresponds to 48 kbps per
time slot, assuming an eight time slot terminal. The key point is that when
unlocked, this spectrum should provide significant additional capacity (see following
section on modulation for discussion of potential capacity increase).
Capacity is increased by uy
between 1.7x-2.8x
Using a 2.7x we estimate that capacity is increased by between 2.6x and 5.0x. Capacity is increased
between 2.6x and 5.0x
Even at this conservative level this is a material increase in capacity. Looking
using more aggressive
forward, as operators add further transceivers to each cell site, an explosion in estimate of spectral
capacity could occur at a very low incremental capital expenditure (i. real estate efficiency
costs, and civil engineering costs have already been met). We also note that
eventually existing spectrum allocated to 2G will be re-farmed for use with W-
CDMA technology. This would drive a minimal 20% increase in the 2G spectrum to
which in the analysis above we apply a 1x multiple. Hence our capacity increase
estimates appear very conservative.
EDGE uses a new modulation technique and improvements in radio protocol that
allows existing operators to use existing frequencies more efficiently (800, 900,
1800 and 1900 MHz). Software upgrades in the base station system enable use of
The core network does not
the new protocol facilitating packets over the radio interface in both the base require any adaptations
station and base station controller; new transceiver units in the base station enable
use of the new modulation technique. The core network does not require any
adaptations. EDGE increases theoretical end-user data rates up to 384Kbps (3
times speed of GPRS); however, realistic data rates range between 80Kbps and
150Kbps.
Voice
Standard GSM Transceiver
Voice
EDGE Transceiver
Freed up capacity!
Source: ABN AMRO/Nokia
Should scepticism grow further among investors and operators over the potential
Should scepticism grow
for revenue from 3G wireless services in the foreseeable future (2-3 years), some among investors over 3G,
operators may question if they need to incur the substantial costs of upgrading EDGE may be explored
their networks beyond EDGE to 3G. further
US perspective
Operators see little need for higher data rates within the current suite of products
being developed. Supporting this view for US GSM operators (AWE, Cingular and Arguably US operators
provide a sanity check
VoiceStream) is that fact that timing of availability of sufficient additional spectrum
is very unclear. Cingular, the second-largest US wireless operator, has announced
that it had no plans after EDGE to roll out W-CDMA.
European options
European operators that have no such spectrum issues following the recent 3G
auctions have long been expected to bypass EDGE in their rush to deploy third-
generation WCDMA. However, there stands a greater chance now that European
operators also concerned with returns on additional capex may decide to delay
investing in 3G infrastructure. Obviously, this would require changes in
laws/regulations regarding scheduled build-out licence requirements and use of
new 3G spectrum. However, it is important that investors are aware of this issue
and its implications for free cash flow.
Fixed-line tariff rebalancing is a significant topic that merits more space than we
have in this note (hence our intention to revisit it at a later date). However, we
must recognise that fixed line tariff rebalancing provides a potential third driver of
material fixed-to-mobile call substitution. Should fixed line players be forced to
rebalance their tariffs (ie increase the price of access and drop the price of calls),
we suspect the wireless handset could become the primary telephony access point
for many customers. Political pressure on regulators to maintain universal access
provides a barrier to rebalancing. Notwithstanding this issue, we suspect operators
could overcome the obstacle via the provision of a bundled wireless alternative.
Background
Long distance fixed-line voice calls in Europe typically do not reflect underlying
network cost structure (for example, a fixed-line long distance call may cost Fixed-line telephony prices
approximately €0.04, compared with as little as €0.02 for local traffic, without any do not reflect underlying
network cost structures
underlying cost difference). The difference in cost is an incumbent telephony
legacy, in which distance was a key driver of cost. The importance of distance as a
driver of cost has diminished (reflecting lower backhaul cost, driven by technology
innovation and the removal of physical operators connecting calls).
Wireless opportunity
In our central case analysis of the fixed-to-mobile call market, we conservatively In the event of fixed-line
tariff rebalancing, our local
assumed the migration of only about 30% of all local call volume from the wireline
access substitution scenario
to the wireless network. We further assumed that no fixed-line access fees would (30%) could prove
be displaced to wireless. We suspect that in the event of tariff rebalancing, this conservative
In our view, a core opportunity for Europe’s cellular operators pursuing fixed-to-
mobile substitution is provided by the local access/local call volume portion of voice
fixed-line calls and fixed to mobile calls. As we noted earlier in this research study,
as fixed-line voice call volume falls the implied price of each incremental unit of use
increases, thus encouraging savvy customers to re-assess their need for a wireline
connection to the home/office.
Lobbying opportunity
Looking forward, this environment may help the cellular operators to lobby Reduced cellular tariffs and
wireline tariff re-balancing
regulators to rebalance local call charges to reflect the underlying network cost
could be powerful
structure. It is possible that if tariffs were rebalanced (to reflect underlying cost stimulants of growth
structure), local calls could move towards the price of long distance calls.
Dramatically reduced European cellular tariffs, coupled with rebalancing of fixed-
line voice tariffs, could be a powerful stimulant of the next phase of cellular growth.
US perspective
We note that in the US in 1993, President Clinton mandated that auctions be held
for frequencies within the PCS airwaves. The primary motive was to open up
competition in the cellular market, drive down the cost of wireless driving
individuals to use their cellular phone for local calls.
Economic impact
We see between 10% and 25% upside from our existing fair value estimates for the
European wireless community as a result of (1) our pre-to-post-paid migration
thesis and (2) our central case fixed-to-mobile substitution belief. Our published
price targets do not factor in the upside from a structural shift in fixed-to-mobile
substitution. As evidence builds supporting our thesis we will re-visit our published
fair value estimates. Full bottom-up detail by major European wireless operator is
detailed in the sister publication, Wireless Model Builder: Edition 1. Current cellular
share prices bet against the emergence of mass fixed-to-mobile substitution. We
recognise that margins could contract in the event of a change in business model.
Investors will need to be nimble, timing will be crucial.
Timing
Two distinct positive drivers
As already noted we see two distinct positive drivers of ARPU looking forward:
of ARPU growth
■ Phase One (2002-2006): 4%-8% annual growth in blended ARPU from pre-
to-post-paid migration for the next four years; and
■ Phase Two (2004-2006): an 18% structural shift in outgoing voice ARPU by 1) Pre-to-post migration,
and
2006 (holding all other factors equal 11% overall ARPU growth)
2) F2M substitution
Holding incoming ARPU constant, this results in up to 30% overall ARPU growth in
by year-end 2006. This would be a material upside surprise to consensus
expectations.
Market-by-market positioning
With the help of David Wright from our wireline telecoms team, we have developed
a schema to better understand the impact of the fixed-to-mobile substitution
opportunity/threat by major European market. This schema highlights two key
drivers of a market’s attractiveness:
■ the difference in price per minute between an average fixed and cellular call.
7000
Intermodal HH, outgoing voice minutes
6500
6000
5500
5000
4500
4000
3500
3000
2500
1.2 1.4 1.6 1.8 2 2.2 2.4 2.6
mob/fix price factor
Spain Italy Portugal Germany Uk Holland France
Quadrant One (top left): Bad long term, bad short term, is the worst possible
place for an incumbent wireline operator to occupy on the schema. Operators in
this quadrant have a high level of concentration (indicated by the monopoly style
HHI score on the y-axis) suggesting negative regulatory intervention and mobile
pricing has fallen below the crucial substitution level of 1.5x (see earlier discussion
on fixed to mobile for rationale).
Quadrant Two (bottom left): Good long term, bad short term, is marginally
better in terms of the level of concentration, suggesting adverse regulatory
intervention is unlikely. Against mobile enjoys large-scale substitution in markets in
this schema, as a result of the decline in relative price. Portugal Telecom sits in
this quadrant. As we noted in European Cellular Bear Trap, Portugal has been
charecterised by the emergence of fixed call volume substitution to mobile. Minutes
of domestic calls have fallen by 8% pa, and 10% of households use mobile only,
compared with the UK at an estimated 4%.
Quadrant Three (top right): Bad long term, good short term, is a sweet spot
for existing returns for incumbent wireline operators, as they enjoy high levels of
market concentration (ie low competition) and mobile is still priced at a high
premium to fixed, suggesting little imminent impact from fixed to mobile
substitution. Spain and the Netherlands fit into this category. Longer term, the
prospect of fixed-line tariff rebalancing (increasing competition, lowering the HHI)
and aggressive 3G-driven price cuts suggest that a downward and “left” trajectory
will be established.
Quadrant Four (bottom right): Good long term, good short term, arguably is
the sweet spot for both existing and future returns for incumbent wireline
operators, as they have relatively minimal regulatory risk, and continue to display a
high premium for wireless prices over wireline (delaying the impact of substitution).
BT and DT sit within this quadrant.
In the following table we have presented a framework for identifying the risk to We have considered the
impact on margin
margins. In the left-hand column we have the potential volume increase from fixed
to mobile substitution; in the top row we detail the expected price decline per
minute. For example, using an initial price per minute decline of 30%, a volume
increase of 10% results in a margin decline of 13%. We can envisage a run-rate of
decline moving diagonally upwards towards the top right of the table. The exact
margin contraction will be determined by the volume increase that follows the price
cuts.
Table 47 : Generic example of margin impact (for given price and volume)
Price per minute fall
Year end December Units X+0 X+1 X+2 X+3 X+4 X+5 X+6 X+7 X+8 X+9 X+10 X+11
Country X Population m 65.7 66.3 67.0 67.6 68.3 69.0 69.7 70.4 71.1 71.8 72.5 73.2
Penetration rate % 75.0 76.0 77.0 78.0 79.0 80.0 81.0 82.0 83.0 84.0 85.0 85.0
All subscribers (period end) m 49.2 50.4 51.6 52.8 54.0 55.2 56.4 57.7 59.0 60.3 61.6 62.3
ANALY SIS
Country X Net Adds per period m 1.14 1.16 1.17 1.19 1.21 1.23 1.25 1.27 1.29 1.31 1.33 0.62
Market Churn % per month % 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Country X Churn per period m 11.5 11.8 12.1 12.4 12.7 13.0 13.2 13.5 13.9 14.2 14.5 14.8
MO BILE
Country X Gross Adds m 12.7 13.0 13.3 13.6 13.9 14.2 14.5 14.8 15.1 15.5 15.8 15.4
Operator Y Subscribers (period end) m 13.0 13.2 13.3 13.5 13.7 14.0 14.3 14.5 14.8 15.1 15.5 15.6
Net Adds per period m 0.05 0.11 0.16 0.20 0.23 0.25 0.27 0.28 0.30 0.31 0.32 0.14
NET WO RKS
-
Churn % per month % 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
W.
Churn per period m 3.1 3.1 3.2 3.2 3.2 3.3 3.4 3.4 3.5 3.6 3.6 3.7
Gross Adds m 3.17 3.24 3.32 3.39 3.47 3.55 3.62 3.70 3.78 3.87 3.95 3.85
Share of Gross Adds % 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0%
E UROPE
Share of Net Adds % 4.7% 9.8% 13.6% 16.5% 18.6% 20.2% 21.4% 22.3% 23.0% 23.5% 23.9% 23.2%
Overall Market Share % 26.5% 26.1% 25.8% 25.6% 25.5% 25.3% 25.2% 25.2% 25.1% 25.1% 25.1% 25.1%
Contract Subscribers (period end) m 4.17 5.11 5.88 6.50 7.02 7.46 7.85 8.19 8.49 8.77 9.04 9.18
Contract Net Adds per period m 0.27 0.95 0.76 0.63 0.52 0.44 0.38 0.34 0.31 0.28 0.26 0.14
16
Contract Churn % per month % 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
Contract Churn per period m 0.9 1.0 1.2 1.4 1.6 1.7 1.8 1.9 2.0 2.0 2.1 2.2
Contract Gross Adds m 1.2 1.9 2.0 2.0 2.1 2.1 2.2 2.2 2.3 2.3 2.4 2.3
O CT OBE R
Contract Share of overall net adds % 504% 834% 477% 318% 231% 178% 143% 120% 103% 92% 83% 100%
200 2
Contract Share of overall gross adds % 38% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60%
Contract % of Total Subs % 32% 39% 44% 48% 51% 53% 55% 56% 57% 58% 58% 59%
77
Table 49: Generic adoption model (cont’)
R IS K
Year end December Units X+0 X+1 X+2 X+3 X+4 X+5 X+6 X+7 X+8 X+9 X+10 X+11
Prepaid Subscribers (period end) m 8.87 8.04 7.44 7.01 6.72 6.52 6.41 6.35 6.34 6.37 6.42 6.42
Prepaid Net Adds per period m -0.22 -0.83 -0.60 -0.43 -0.30 -0.19 -0.12 -0.06 -0.01 0.03 0.05 0.00
Prepaid Churn % per month % 2.0 2.1 2.1 2.1 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
ANALY SIS
Prepaid Churn per period m 2.2 2.1 1.9 1.8 1.7 1.6 1.6 1.5 1.5 1.5 1.5 1.5
Prepaid Gross Adds m 1.97 1.30 1.33 1.36 1.39 1.42 1.45 1.48 1.51 1.55 1.58 1.54
Prepaid Share of overall net adds % -404% -734% -377% -218% -131% -78% -43% -20% -3% 8% 17% 0%
MO BILE
Prepaid Share of overall gross adds % 62% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40%
Prepaid % of Total Subs % 68% 61% 56% 52% 49% 47% 45% 44% 43% 42% 42% 41%
High Value Contract Subscribers (period end) m 4.17 3.75 3.25 2.77 2.42 2.16 1.97 1.83 1.73 1.66 1.62 1.23
NET WO RKS
-
High Value Net Adds per period m 0.27 -0.42 -0.50 -0.47 -0.35 -0.26 -0.19 -0.14 -0.10 -0.07 -0.04 -0.39
W.
High Value Churn % per month % 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
High Value Churn per period m 0.9 1.0 0.9 0.8 0.7 0.6 0.5 0.5 0.4 0.4 0.4 0.4
High Value Gross Adds m 1.2 0.6 0.4 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.0
E UROPE
High Value Share of overall contract net adds % 100% -44% -66% -76% -68% -59% -50% -41% -32% -24% -16% -273%
High Value Share of overall contract gross adds % 100% 30% 20% 15% 15% 15% 15% 15% 15% 15% 15% 0%
High Value % of Total Contract Subs 100% 73% 55% 43% 34% 29% 25% 22% 20% 19% 18% 13%
Phase 1 Contract Subscribers (period end) m 1.36 1.83 1.90 1.96 2.02 2.08 2.14 2.19 2.25 2.30 2.33
16
Phase 1 Net Adds per period m 1.36 0.47 0.07 0.06 0.06 0.06 0.06 0.05 0.05 0.05 0.03
Phase 1 Churn % per month % 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
Phase 1 Churn per period m 0.0 0.3 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.6
O CT OBE R
Phase 1 Gross Adds m 1.4 0.8 0.5 0.5 0.5 0.5 0.6 0.6 0.6 0.6 0.6
200 2
Phase 1 Share of overall contract net adds % 144% 61% 11% 12% 14% 15% 16% 18% 19% 20% 18%
Phase 1 Share of overall contract gross adds % 70% 40% 25% 25% 25% 25% 25% 25% 25% 25% 25%
Phase 1 % of Total Contract Subs % 27% 31% 29% 28% 27% 27% 26% 26% 26% 25% 25%
78
Table 50: Generic adoption model (cont’)
R IS K
Year end December Units X+0 X+1 X+2 X+3 X+4 X+5 X+6 X+7 X+8 X+9 X+10 X+11
Phase 2 Contract Subscribers (period end) m 0.80 1.11 1.37 1.57 1.74 1.88 1.99 2.10 2.18 2.24
Phase 2 Net Adds per period m 0.80 0.32 0.25 0.20 0.17 0.14 0.12 0.10 0.09 0.05
Phase 2 Churn % per month % 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
ANALY SIS
Phase 2 Churn per period m 0.0 0.2 0.3 0.3 0.4 0.4 0.5 0.5 0.5 0.5
Phase 2 Gross Adds m 0.8 0.5 0.5 0.5 0.5 0.6 0.6 0.6 0.6 0.6
Phase 2 Share of overall contract net adds % 104% 51% 49% 46% 43% 41% 38% 36% 34% 38%
MO BILE
Phase 2 Share of overall contract gross adds % 40% 25% 25% 25% 25% 25% 25% 25% 25% 25%
Phase 2 % of Total Contract Subs % 14% 17% 19% 21% 22% 23% 23% 24% 24% 24%
Phase 3 Contract Subscribers (period end) m 0.71 1.27 1.71 2.06 2.34 2.58 2.77 2.93 3.39
NET WO RKS
Phase 3 Net Adds per period m 0.71 0.56 0.44 0.35 0.28 0.23 0.19 0.16 0.45
-
Phase 3 Churn % per month % 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
W.
Phase 3 Churn per period m 0.0 0.2 0.3 0.4 0.5 0.6 0.6 0.7 0.7
Phase 3 Gross Adds m 0.7 0.7 0.7 0.8 0.8 0.8 0.8 0.8 1.2
E UROPE
Phase 3 Share of overall contract net adds % 114% 107% 99% 92% 84% 76% 69% 62% 318%
Phase 3 Share of overall contract gross adds % 35% 35% 35% 35% 35% 35% 35% 35% 50%
Phase 3 % of Total Contract Subs % 11% 18% 23% 26% 29% 30% 32% 32% 37%
16
High Value Period End Contract Subscribers m 4.2 3.7 3.2 2.8 2.4 2.2 2.0 1.8 1.7 1.7 1.6 1.2
Phase 1 Period End Contract Subscribers m - 1.4 1.8 1.9 2.0 2.0 2.1 2.1 2.2 2.2 2.3 2.3
Phase 2 Period End Contract Subscribers m - - 0.8 1.1 1.4 1.6 1.7 1.9 2.0 2.1 2.2 2.2
O CT OBE R
Phase 3 Period End Contract Subscribers m - - - 0.7 1.3 1.7 2.1 2.3 2.6 2.8 2.9 3.4
Total Period End Contract Subscribers m 4.2 5.1 5.9 6.5 7.0 7.5 7.8 8.2 8.5 8.8 9.0 9.2
200 2
Total Period End Prepaid Subscribers m 8.9 8.0 7.4 7.0 6.7 6.5 6.4 6.4 6.3 6.4 6.4 6.4
Total Period End Subscribers m 13.0 13.2 13.3 13.5 13.7 14.0 14.3 14.5 14.8 15.1 15.5 15.6
High Value Average Contract Subscribers m 4.0 4.0 3.5 3.0 2.6 2.3 2.1 1.9 1.8 1.7 1.6 1.4
Phase 1 Average Contract Subscribers m - 0.7 1.6 1.9 1.9 2.0 2.1 2.1 2.2 2.2 2.3 2.3
Phase 2 Average Contract Subscribers m - - 0.4 1.0 1.2 1.5 1.7 1.8 1.9 2.0 2.1 2.2
Phase 3 Average Contract Subscribers m - - - 0.4 1.0 1.5 1.9 2.2 2.5 2.7 2.9 3.2
Total Average Contract Subscribers m 4.0 4.6 5.5 6.2 6.8 7.2 7.7 8.0 8.3 8.6 8.9 9.1
Average Prepaid Subscribers m 9.0 8.5 7.7 7.2 6.9 6.6 6.5 6.4 6.3 6.4 6.4 6.4
Average Total Subscribers m 13.0 13.1 13.2 13.4 13.6 13.9 14.1 14.4 14.7 15.0 15.3 15.5
79
Table 51: Generic adoption model (cont’)
R IS K
Year end December Units X+0 X+1 X+2 X+3 X+4 X+5 X+6 X+7 X+8 X+9 X+10 X+11
High Value Contract ARPU £ 40.0 41.2 42.4 43.7 45.0 46.4 47.8 49.2 50.7 52.2 53.8 55.4
% change PoP % 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0%
ANALY SIS
Phase 1 Contract ARPU £ 32.0 33.0 33.9 35.0 36.0 37.1 38.2 39.4 40.5 41.8 43.0
% change PoP % -20.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0%
Phase 2 Contract ARPU £ 25.6 26.4 27.2 28.0 28.8 29.7 30.6 31.5 32.4 33.4
MO BILE
% change PoP % -20.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0%
Phase 3 Contract ARPU £ 20.5 21.1 21.7 22.4 23.1 23.7 24.5 25.2 25.9
% change PoP % -20.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0%
Total Blended Contract ARPU (Calculated) £ 40.0 39.8 38.5 36.8 35.4 34.7 34.6 34.7 35.1 35.7 36.4 36.7
NET WO RKS
-
% change PoP % -0.4% -3.5% -4.4% -3.8% -1.8% -0.5% 0.5% 1.2% 1.6% 2.0% 0.7%
W.
Prepaid ARPU £ 9.0 9.2 9.4 9.6 9.7 9.9 10.1 10.3 10.5 10.8 11.0 11.2
% change PoP % 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0%
Blended ARPU (subscriber revenue only) £ 18.6 20.0 21.4 22.1 22.5 22.9 23.4 23.9 24.5 25.1 25.8 26.1
E UROPE
% change PoP % 7.7% 7.0% 3.1% 1.6% 1.9% 2.1% 2.3% 2.4% 2.5% 2.6% 1.4%
Revenues:
Total Contract revenues £m 1,935 2,218 2,536 2,729 2,869 3,017 3,174 3,340 3,515 3,699 3,893 4,011
% change PoP % 14.6% 14.3% 7.6% 5.1% 5.2% 5.2% 5.2% 5.2% 5.2% 5.2% 3.0%
16
Prepaid revenues £m 970 932 870 828 802 789 786 791 803 820 841 862
% change PoP % -4.0% -6.7% -4.8% -3.1% -1.6% -0.4% 0.6% 1.5% 2.1% 2.6% 2.4%
Total service revenues 2,905 3,150 3,405 3,557 3,671 3,807 3,961 4,131 4,318 4,519 4,734 4,873
O CT OBE R
Source: ABN AMRO
200 2
80
A P P E N D I X
Italy 20 Regional capitals <30 months. Provincial capitals within next 30 months
Norway
90% of the population of Greater Oslo, Bergen, Stavanger/Sandnes, Trondheim,
12 Fredrikstad/Sarpsborg, Porsgrunn/Skien, Drammen, Kristiansand, Tromsø, Tønsberg/ Åsgårdstrand,
Sandefjord, and Bodø within five years
Denmark 15 End-04 30% population coverage, end 08-80%
Technical glossary
Erlang
An Erlang is a unit of telecommunications traffic measurement. Strictly speaking,
an Erlang represents the continuous use of one voice path. In practice, it is used to
describe the total traffic volume of one hour. For example, if a group of users made
30 calls in one hour, and each call had an average call duration of 5 minutes, then
the number of Erlangs this represents is worked out as follows:
Trunking
Erlang traffic measurements are made to help telecommunications network
designers understand traffic patterns within their voice networks. This is essential if
they are to establish the necessary trunk group sizes. The more efficient the
existing lines are used, the higher the trunking efficiency is of the network.
Erlang B
Several traffic models exist that share their name with the Erlang unit of traffic.
They are formulae that can be used to estimate the number of lines required in a
network, or to a central office (PSTN exchange lines).
Erlang B is the most commonly used traffic model, and is used to work out how
many lines are required if the traffic figure (in Erlangs) during the busiest hour and
the number of blocked calls are known. The model assumes that all unlocked calls
are immediately cleared.
Acknowledgements
We would like to thank Nimeshh Patel, from ABN AMRO’s wireline telecom equity
research team, for his invaluable insights and resource in researching global telco
volume and revenue trends by market and being a generally good bloke.
© Copyright 2002 ABN AMRO Bank N.V and affiliated companies ("ABN AMRO"). All rights reserved.
ABN AMRO Bank NV and/or one of the companies in the ABN AMRO Group have specific interests with respect to
ENEL and TIM, as defined under article 69, paragraph 1 of Consob Resolution n.11971/1999, as subsequently
amended and supplemented, and Consob Notice n.DME/102955 or 20 April 2001. In particular, ABN AMRO Bank
NV and/or its holding company has issued financial instruments related to the securities of ENEL and TIM.
This material1 was prepared by the ABN AMRO affiliate named on the cover or inside cover page. It is provided for informational
purposes only and does not constitute an offer to sell or a solicitation to buy any security or other financial instrument. While
based on information believed to be reliable, no guarantee is given that it is accurate or complete. While we endeavour to
update on a reasonable basis the information and opinions contained herein, there may be regulatory, compliance or other
reasons that prevent us from doing so. The opinions, forecasts, assumptions, estimates, derived valuations, and target price(s)
contained in this material are as of the date indicated and are subject to change at any time without prior notice. The
investments referred to may not be suitable for the specific investment objectives, financial situation or individual needs of
recipients and should not be relied upon in substitution for the exercise of independent judgment. ABN AMRO may from time to
time act as market maker, where permissible under applicable laws, buy or sell as agent or principal securities, warrants,
futures, options, derivatives or other financial instruments referred to herein. ABN AMRO, officers, directors, employee benefit
programmes or employees, including persons involved in the preparation or issuance of this material may from time to time
have long or short positions in securities, warrants, futures, options, derivatives or other financial instruments referred to in
this material. ABN AMRO may at any time solicit or provide investment banking, commercial banking, credit, advisory or other
services to the issuer of any security referred to herein. Accordingly, information may be available to ABN AMRO, which is not
reflected in this material, and ABN AMRO may have acted upon or used the information prior to, or immediately following its
publication. ABN AMRO may also within the last three years have acted as manager or co-manager for a public offering of
securities of issuers referred to herein. The stated price of any securities mentioned herein is as of the date indicated and is not
a representation that any transaction can be effected at this price. Neither ABN AMRO or other person shall be liable for any
direct, indirect, special, incidental, consequential, punitive or exemplary damages, including lost profits arising in any way from
the information contained in this material. This material is for the use of intended recipients only and the contents may not be
reproduced, redistributed, or copied in whole or in part for any purpose without ABN AMRO's prior express consent. In any
jurisdiction in which distribution to private/retail customers would require registration or licensing of the distributor which the
distributor does not have currently, this document is intended solely for distribution to professional and institutional investors.
Should you require additional information please contact your local ABN AMRO account representative, unless governing laws
dictate otherwise.
Australia: This document is distributed in Australia by ABN AMRO Equities Australia Ltd (ABN 84 002 768 701), a participating
organisation of the Australian Stock Exchange Ltd. Australian investors should note that this document was prepared for
wholesale investors only.
Canada: The securities mentioned in this material are available only in accordance with applicable securities laws and may not
be eligible for sale in all jurisdictions. Persons in Canada requiring further information should contact ABN AMRO Incorporated.
Hong Kong: This document is being distributed in Hong Kong by ABN AMRO Asia Limited to persons whose business involves
the acquisition, disposal or holding of securities.
India: Shares traded on stock exchanges within the Republic of India may only be purchased by different categories of resident
Indian investors, Foreign Institutional Investors registered with The Securities and Exchange Board of India ("SEBI") or
individuals of Indian national origin resident outside India called Non Resident Indians ("NRIs") and overseas corporate bodies
("OCBs"), predominantly owned by such persons or Persons of Indian Origin (PIO). Any recipient of this document wanting
additional information or to effect any transaction in Indian securities or financial instrument mentioned herein must do so by
contacting a representative of ABN AMRO Asia Equities (India) limited.
Italy: Persons in Italy requiring further information should contact ABN AMRO Bank N.V. Milan Branch.
New Zealand: This document is distributed in New Zealand by ABN AMRO Equities New Zealand Limited, a member firm of the
New Zealand Stock Exchange.
Russia: The Russian securities market is associated with several substantial risks: legal, economic and political ; and, high
volatility. There is a relatively high measure of legal uncertainty concerning rights, duties and legal remedies in the Russian
Federation. Russian laws and regulations governing investments in securities markets may not be sufficiently developed or may
be subject to inconsistent or arbitrary interpretation or application. Russian securities are often not issued in physical form and
registration of ownership may not be subject to a centralized system. Registration of ownership of certain types of securities
may not be subject to standardized procedures and may even be effected on an ad hoc basis. The value of investments in
Russian securities may be affected by fluctuations in available currency rates and exchange control regulations.
Singapore: This document is distributed in Singapore by ABN AMRO Asia Securities (Singapore) Private Limited to clients who
fall within the description of persons in Regulation 44(i) to (iii) of the Securities Industry Regulations. Investors should note
that this material was prepared for professional investors only.
United Kingdom: This document has been issued in the United Kingdom by ABN AMRO Equities UK Limited, which is registered
in England No 2475694 and is regulated by the Financial Services Authority. The investment and services contained herein are
not available to private customers in the UK.
United States: Distribution of this document in the United States or to U.S. persons is intended to be solely to major
institutional investors as defined in Rule 15a-6 under the U.S. Securities Act of 1934. All U.S. persons that receive this
document by their acceptance thereof represent and agree that they are a major institutional investor and understand the risks
involved in executing transactions in securities. Any U.S. recipient of this document wanting additional information or to effect
any transaction in any security or financial instrument mentioned herein, must do so by contacting a registered representative
of ABN AMRO Incorporated, Park Avenue Plaza, 55 East 52nd Street, New York, N.Y. 10055, U.S., tel + 1212 409 1000, fax
+1212 409 5222.
1
Material means all research information contained in any form including but not limited to hard copy, electronic form,
presentations, e-mail, SMS or WAP.