telecommunications - startseite | berenberg · european telecommunications telecommunications 3...

BERENBERG EQUITY RESEARCH

Telecommunications

Mobile consolidation: a reality check

Volume 1: Thematic thinking

18 June 2013

Telecommunications

Stuart Gordon Analyst + 44 20 3207 7858 [email protected]

Paul Marsch Analyst + 44 20 3207 7857 [email protected]

Barry Zeitoune Analyst + 44 20 3207 7859 [email protected]

Usman Ghazi Analyst + 44 20 3207 7824 [email protected]

Wassil El Hebil Analyst + 44 20 3207 7862 [email protected]

Laura Janssens Marketing Analyst + 44 20 3465 2639 [email protected]

Julia Thannheiser Specialist Sales + 44 20 3465 2676 [email protected]

What is Berenberg THOUGHT LEADERSHIP?

Berenberg’s analysts are recognised by investors and by corporates for their in-depth research into the industries they cover.

Our THOUGHT LEADERSHIP brand will highlight the deep-dive fundamental industry research that we feel is most important to informing our forecasts and ratings.

For our disclosures in respect of section 34b of the German Securities Trading Act (Wertpapierhandelsgesetz – WpHG) and our disclaimer please see the end of this document. Please note that the use of this research report is subject to the conditions and restrictions set forth in the disclosures and the disclaimer at the end of this document.

European Telecommunications Telecommunications

3

Table of contents

Mobile consolidation: a reality check 4

Summary and investment conclusion 7

The anxiety of the bears: what keeps us awake at night? 15

Consolidation – hype or reality? 35

Understanding the EC remit for evaluating deals 43

The EC merger control regime and telecoms case studies 49

Case study: T-Mobile/Tele.ring, Austria (2006) 52

Case study: T-Mobile/Orange, the Netherlands (2007) 56

Case study: Everything Everywhere, UK (2010) 58

Case study: H3G/Orange, Austria (2012) 60

How has the Everything Everywhere JV performed? 63

Market reviews – key conclusions 70

Danish consolidation is much needed, but looks unlikely 83

French consolidation has potential, but is politically sensitive 102

Germany: relatively well placed for remedies 111

Italy: spectrum remedies likely to undermine upside 124

Dutch mobile has little prospect of consolidation 137

UK: consolidation debate is moot 149

Spain: unlikely to see mobile consolidation 162

Swedish consolidation too complex to attain 176

Disclosures in respect of section 34b of the German Securities Trading Act (Wertpapierhandelsgesetz – WpHG) 195

Contacts: Investment Banking 205


4

Mobile consolidation: a reality check

• We remain cautious about the sector: Sector consensus OpFCF expectations for 2014 have fallen by 7% so far this year, yet the consensus view is for stable revenues and EBITDA in 2014, and falling capex. We think the downward earnings revision cycle has further to go, with headwinds in the form of pressure on revenues (roaming rate cuts, fixed-mobile convergence, quad-play, competitive intensity in mobile and FX), pressure on EBITDA (from revenues and ongoing structural margin compression), and upside pressure on capex (from increased spend on fibre and VDSL and potential spend on 2G spectrum re-farming). While downward sector earnings revisions are similar to those for the market, we think market revisions are more cyclical, while sector revisions are more structural.

• Consolidation is unlikely to save the day: In this report, we look at the eight main European four-player mobile markets and try to identify which markets might be most/least exposed to severe consolidation remedies. We conclude that Germany and France look best positioned to mitigate consolidation remedies, the Netherlands and Italy look poorly positioned to mitigate remedies, Sweden and Denmark face structural hurdles to consolidation, and the UK – in theory a very attractive market for consolidation – faces an explicit regulatory preference to retain four mobile networks. However, our overall conclusion is that mobile in-market consolidation looks very likely to attract material anti-trust remedies which will likely act as a disincentive to the kind of in-market consolidation investors want to see – the kind that leads to clear in-market repair. The recent Austrian deal precedent suggests that 260MHz of spectrum could be handed back to remedy consolidation in the eight markets analysed. That is a lot of potential new entrant opportunities! We think a change of approach by EC anti-trust authorities is unlikely given the strong “signalling” intent of the EC’s approach to recent Austrian consolidation, and also given EC telecoms commissioner Neelie Kroes’s recent comments.

• Germany and France look best positioned to mitigate consolidation remedies: These two markets already have low prices and big bundles, a well developed MVNO presence with good consumer choice. While the most obvious consolidation moves – O2 and E+ in Germany and Bouygues and Iliad (ILD) in France – would be unlikely to escape remedies, we think the scope for remedies would be narrower than for peer group markets.

• The Netherlands, Italy, Sweden and Denmark look poorly positioned to mitigate consolidation remedies, or to consolidate at all: Consolidation in Italy would significantly increase market and spectrum consolidation, exposing consolidators to the risk of significant spectrum divestment, potentially to a new entrant, improved terms to stimulate the MVNO sector and also potentially improved roaming or network access rights for new entrants. The Netherlands remains among the highest priced and most concentrated markets and likewise appears exposed to remedies to encourage new entrant access (roaming, network access, spectrum divestment). Existing network-sharing joint ventures (JVs) in Sweden and Denmark present structural challenges to potential consolidation.

18 June 2013

Research Team Stuart Gordon Analyst +44 20 3207 7858 [email protected]

Paul Marsch Analyst +44 20 3207 7857 [email protected]

Barry Zeitoune

Analyst +44 20 3207 7859 [email protected]

Usman Ghazi Analyst +44 20 3207 7824 [email protected]

Wassil El Hebil Analyst +44 20 3207 7862 [email protected]

Laura Janssens Marketing Analyst +44 20 3465 2639 +44 7879 848 611 [email protected]

Julia Thannheiser Specialist Sales +44 20 3465 2676 +44 7879 848 611 [email protected]

mailto:[email protected]


5

• The sector has modestly de-rated versus the market on earnings, but EBITDA multiples have increased: The sector has performed broadly in line with the market ytd, but has underperformed by 17% over the last year. Over that time, the sector P/E relative has de-rated from 1.0x to 0.90x the market’s P/E, justified in our view by the downward revisions to consensus earnings. The sector P/E relative remains a modest 8% below the five-year average sector P/E relative. The sector dividend yield relative of 1.5x has tightened significantly over the last year from over 2x, reflecting the cuts to dividends across the sector, and is now back to pre-2009 levels. The sector EBITDA multiple has expanded to 5.5x, reflecting ongoing reductions to EBITDA estimates that have not been matched by lower enterprise values. Overall, we find it difficult to say that the sector is good value.

• We have more Sells than Buys: We have seven Sell-rated stocks, of which our least preferred names are Telefónica (TEF), TDC, Telekom Austria (TKA), Belgacom (BELG), Mobistar (MOBB), TalkTalk (TALK) and Elisa (ELI1V). We struggle to find Buy ratings that present a combination of valuation upside and catalysts, so we have only two incumbent stocks on a Buy rating; BT Group (BT) and Tele2 (TEL2B), and a preference for Jazztel (JAZ) and United Internet (UTDI) among the competitive sector names. In this note, we upgrade the rating on KPN (to Hold from Sell; PT reduced to €1.45 from €1.62) and downgrade our rating on ELI1V (to Sell from Hold; PT €12.85).


6

Exhibit 1.1: Stock ratings and price targets – incumbents

Source: Berenberg

Exhibit 1.2: Stock ratings and price targets, “exotics”

Source: Berenberg

Exhibit 1.3: Stock ratings and price targets – cable

Source: Berenberg

Price Price Implied

Remaining

Div.

Yield % to

Total

Returns %

Rating 17-Jun-13 Target Upside % 13E 12E

BT Group Buy £ 3.12 £ 3.75 20.0% 4.1% 24.2%

Tele2 Buy SEK 79.45 SEK 104.00 30.9% 0.0% 30.9%

Deutsche Telekom Hold € 8.76 € 9.90 13.1% 0.0% 13.1%

Vodafone Hold £ 1.80 £ 1.85 2.7% 1.8% 4.6%

Telenor Hold NOK 121.00 NOK 137.00 13.2% 0.0% 13.2%

France Telecom Hold € 7.36 € 8.00 8.6% 6.8% 15.4%

Telecom Italia (Savs) Hold € 0.43 € 0.56 30.2% 0.0% 30.2%

Telecom Italia (Ords) Hold € 0.54 € 0.68 25.9% 0.0% 25.9%

Millicom Hold SEK 509.00 SEK 540.00 6.1% 2.6% 8.7%

KPN Hold € 1.54 € 1.45 -6.0% 0.0% -6.0%

Portugal Telecom Hold € 3.11 € 4.45 43.1% 0.0% 43.1%

Swisscom Hold CHF 403.00 CHF 408.00 1.2% 0.0% 1.2%

Teliasonera Hold SEK 42.86 SEK 42.00 -2.0% 0.0% -2.0%

TDC Sell DKK 43.70 DKK 42.00 -3.9% 3.4% -0.5%

Belgacom Sell € 17.21 € 16.00 -7.0% 2.9% -4.1%

Telefonica Sell € 10.03 € 9.10 -9.3% 0.0% -9.3%

Telekom Austria Sell € 5.20 € 4.20 -19.2% 0.0% -19.2%

Incumbents Sector Average 8.7% 1.3% 10.0%

Incumbents Weighted Average 5.3% 1.4% 6.7%

Price Price Implied

Remaining

Div.

Yield % to

Total

Returns %


Jazztel Buy € 5.85 € 7.00 19.7% 0.0% 19.7%

United Internet Buy € 22.35 € 26.00 16.3% 0.0% 16.3%

QSC Buy € 2.69 € 3.00 11.5% 0.0% 11.5%

Drillisch Hold € 12.11 € 14.00 15.6% 0.0% 15.6%

Freenet Hold € 16.76 € 18.00 7.4% 0.0% 7.4%

C&W Communications Hold £ 0.40 £ 0.40 -0.3% 8.3% 8.0%

Iliad Hold € 159.40 € 150.00 -5.9% 0.2% -5.7%

Mobistar Sell € 16.30 € 16.00 -1.8% 0.0% -1.8%

Elisa Sell € 15.17 € 12.85 -15.3% 0.0% -15.3%

TalkTalk Sell £ 223.10 £ 160.00 -28.3% 6.7% -21.6%

Other Telcos Sector Average 1.9% 1.5% 3.4%

Other Telcos Weighted Average -1.5% 1.2% -0.3%

Price Price Implied

Remaining

Div.

Yield % to

Total

Returns %


ZON Hold € 3.60 € 4.00 11.1% 0.0% 11.1%

Kabel Deutschland Hold € 82.69 € 85.00 2.8% 3.0% 5.8%

Telenet Hold € 34.00 € 25.50 -25.0% 0.0% -25.0%

Cable Sector Average -4.1% 0.8% -3.3%

Cable Weighted Average -5.3% 1.0% -4.4%

Telecoms Sector Average 4.8% 1.3% 6.1%

Sector Weighted Average 4.2% 1.4% 5.5%


7

Summary and investment conclusion

You guessed it, we are still cautious, although we admit to increased anxiety in our caution. A month ago, our main anxiety was that we might have got it wrong on in-market consolidation. The bulk of this report settles our apprehension on that front. While it is possible to differentiate markets that appear relatively better positioned to see “remedy-lite” in-market consolidation, our overall conclusion is that the type of in-market consolidation that we want to see – the type that leads to market repair with limited remedies – appears to us to be unlikely.

Our anxiety instead has moved on to new pastures. What keeps us awake at night these days is worrying whether the dim outlook for sector earnings, the main reason for our continuing cautious stance, may be eclipsed by pressures on earnings across the rest of the market. Even with this fear, it is difficult to ignore the still long list of reasons to remain cautious about the sector. In short, we remain concerned:

• that consensus estimates continue to fall;

• that pressure on revenues, margins and capex remains;

• that we are not through the credit-driven recapitalisation of the sector yet;

• that consolidation will not save the day.

1) OpFCF consensus estimates continue to fall

Consensus 2014 OpFCF estimates have fallen by 7% ytd and 18.1% since the start of 2012 (based on Bloomberg consensus). While the pace of 2014 OpFCF estimate reduction has slowed so far in Q2 (see Exhibit 1.4), this comes after a first quarter which saw a significant reduction in consensus expectations as operators used the guidance-setting season to rebase expectations.

Companies whose 2014 OpFCF consensus has fallen by more than the average include Kabel Deutschland (KD8), MOBB, KPN, Millicom (MIC), BELG and Telecom Italia (TI). Companies whose 2014 OpFCF consensus has reduced less than the average, or even increased, include BT (increase), TKA (increase), Telenet (TNET), TDC, Deutsche Telekom (DTE) and Telenor (TEL).

We calculate that about two-thirds of the ytd 7% reduction in consensus OpFCF is due to lower EBITDA expectations, and about a third is due to higher capex expectations, with the lower EBITDA expectation explained solely by reduced consensus revenue expectations.


8

Exhibit 1.4: Consensus 2014 OpFCF estimates have reduced

Source: Berenberg

Exhibit 1.5: Shares relative performance versus the change in consensus OpFCF estimates – some clear outliers

Source: Berenberg

Earnings revisions for FY 2013 show a similar adjustment, with FY 2013 expectations now looking for a 4% decline in reported revenues and EBITDA, and a 14% decline in sector OpFCF. The ongoing reduction in consensus expectations, and the acceptance that FY 2013 will see mid-digit declines in revenues and EBITDA and a mid-teens decline in OpFCF, contrasts starkly with a consensus that assumes a return to revenue and EBITDA stability, and OpFCF growth (ie capex declines) into FY 2014, according to our analysis of Bloomberg consensus expectations.

-2.0

-3.7

-5.1-0.3

-4.6

-2.4

-20.0

-18.0

-16.0

-14.0

-12.0

-10.0

-8.0

-6.0

-4.0

-2.0

0.0

Q1/12 Q2/12 Q3/12 Q4/12 Q1/13 Q2 qtd

Pe

rce

nt

red

uct

ion

in

co

nse

nsu

s O

pF

CF

e

stim

ate

s

BCOM

BT

DT

FT

KPN

MIC PT

SCMN

TDC

TIT

TEF

TKA

TEL

TLSN

VOD

KD8

TNET

VMED

ZON

MOBB ELISA

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

-35% -30% -25% -20% -15% -10% -5% 0% 5% 10%

R

E

T

U

R

N

S

R

E

L

T

O

S

X

K

P

OPERATING FREE CASH FLOW CHANGES

Change in 2014 OpFCF estimates vs relative stock price performance ytd

Shares below the line have underperfomed relative to the reduction in OpFCFestimates.

Shares above the line have outperfomed relative to the change in OpFCFestimates.


9

2) Pressure on revenues continues

Only three out of 18 European quoted operators in Exhibit 1.7 managed to deliver organic group revenue growth in Q1 2013, and two of these, MICC and TEL, are exposed to emerging markets growth. Six of the 18 saw organic revenues fall by more than 5% yoy.

The mobile environment in particular continues to appear price deflationary, with:

• mobile operators increasing bundle sizes and shaving the price curve as they seek to tempt consumers to “tier up” to a higher volume data package;

• EC commissioner Neelie Kroes’s pursuit of a truly pan-European market implying the end of roaming revenues, which account for about 5-6% of pan-European mobile service revenues;

• quad-play “creep” presaging price deflation as incumbents defensively look to bundle mobile services with wireline triple-play services.

Wireline looks a little better, with some evidence of price increases in markets like the UK and Portugal. However, competitive discounting in some key markets (eg Germany) suggests further downside to incumbent prices, despite the move to more advanced fibre-based services.

The mobile termination rate drag should start to ease into 2014, and excluding MTRs we can see that some markets in Q1 2013 are showing signs of service revenue stability (eg Sweden, Germany, France, the UK). That is not enough for some incumbents, though (eg KPN), which will need to regain market share if they are to stabilise their service revenue trends into 2014 and beyond.

The MTR drag looks set to be superseded by increased pressure on roaming revenues as EC commissioner Neelie Kroes seeks a rapid implementation of her proposals to create a single European market for telecoms services. As Ms Kroes states in her recent Financial Times interview “…roaming does not make sense in a single market”. We think European mobile roaming accounts for about 5% of European mobile service revenues and about 10% of mobile EBITDA, with some operators more exposed than others. We think TKA and MOBB are the most exposed, with roaming accounting for 18% of TKA EBITDA and estimates 15% of MOBB EBITDA. At Vodafone (VOD), we think roaming accounts for about 3-4% of proportionate EBITDA.

Exhibit 1.6: EC roaming glide path still leaves mobile roaming rates at 2.5x average in-market mobile rates

Source: Berenberg

Retail roaming limits (EUR) Before Jul-12 Jul-13 Jul-14

Data (per Mb) none 0.70 0.45 0.20

Voice calls made (per min) 0.35 0.29 0.24 0.19

Voice calls received (per min) 0.11 0.08 0.07 0.05

SMS (per SMS) 0.11 0.09 0.08 0.06

Wholesale roaming limits (EUR) Before Jul-12 Jul-13 Jul-14

Data (per Mb) 0.50 0.25 0.15 0.05

Voice (per min) 0.18 0.14 0.10 0.05

SMS (per SMS) 0.04 0.03 0.02 0.02


10

Exhibit 1.7: Organic revenue growth in Q1, Q2, Q3, Q4 2012 and Q1 2013 by company (ordered by Q1 2013 largest to smallest)

Source: Berenberg

Ericsson recently forecast that western European 4G services would account for 35% of region subscribers by 2018, and the shift to 4G clearly holds some hope for European mobile operators to put a floor under ARPU as they seek to encourage customers to pay more for data. Evidence supporting this thesis is fairly limited so far (eg leading indicator markets like Sweden, Norway and Denmark are not charging more for 4G), and we remain sceptical given the operators’ tendency to discount headline prices to win share (as in Spain and Italy).

3) Pressure on EBITDA and margins continues

Most operators in Europe are seeing EBITDA fall in organic terms (see Exhibit 1.8), and margins gradually compress. Only three of 18 European operators in Exhibit 1.8 saw EBITDA grow in organic terms in Q1 2013, while eight of the 18 saw group organic EBITDA decline by more than 5%.

Consensus expectations are for sector margins to remain stable at 32.4% throughout 2013 and 2014. Our own expectations are for margins to continue to remain under pressure, falling by 70bp into FY 2013 and FY 2014. We think margins will remain pressured by ongoing structural margin compression as low-margin revenues like pay-TV and data grow in the mix, and high-margin revenues like voice telephony and mobile roaming continue to decline in the mix.

-8.0%

-6.0%

-4.0%

-2.0%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

Q1 12 Revs Q2 12 Revs Q3 12 Revs Q4 12 Revs Q1 13 Revs


11

Exhibit 1.8: Organic EBITDA growth in Q1, Q2, Q3, Q4 2012 and Q1 2013 by company (ordered by Q1 2013 largest to smallest)

Source: Berenberg

The shift towards smartphones and fourth-generation services will also maintain pressure in margins, especially if operators fail to translate 4G into higher customer spend. There are some exceptions to the trend of lower margins (eg TEL2B, TEL and TLSN) but the overall picture continues to challenge the consensus expectation for stable margins in the near term.

4) Upside pressure on capex and investment continues

Consensus expects aggregate capex in the sector to decline into FY 2014 by 3.4% after a 10% increase in FY 2013. We think upside pressure on capex will continue as operators pursue numerous capital projects, including the ongoing rollout of 4G mobile network coverage, investment in upgrading copper wireline networks to fibre or VDSL, and ongoing investment in mobile phone spectrum.

Regulatory developments are encouraging higher capex with the European Commission (EC) clearly offering a grand bargain of lighter touch regulation of next-generation networks if the industry will commit to deploying high-speed fibre networks.

-15.0%

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

Q1 12 EBITDA Q2 12 EBITDA Q3 12 EBITDA Q4 12 EBITDA Q1 13 EBITDA


12

While the 4G spectrum auctions are now largely behind us, we face the uncertain prospect that the industry might be required to start paying up for 2G spectrum that was awarded 15-20 years ago and is now coming up for renewal. The German 900Mhz and 1,800Mhz spectrum re-farming decision, due 24 June, will be closely watched as an important leading indicator on this issue. A total 160Mhz of spectrum is up for re-farming, and if the regulator in Germany decides it wants to re-auction this spectrum and the auction generates proceeds in line with the average cost of similar spectrum across Europe (€0.47/Mhz/pop for 900Mhz and say €0.25/Mhz/pop for 1,800Mhz), then the 2G re-farm process could end up costing German operators around €4.4bn (approximately €1.1bn per operator if spectrum share remains the same). If regulators decide that this is a valuable commodity that requires relicensing via competitive auctions, then we could see fears of a new round of European spectrum spend and potential new entrant risk that could last through the next five to 10 years.

5) Scope for more capital raisings and dividend cuts remains

Credit factors remain an important overhang in the sector, particularly for names like TI and KPN, with residual concerns for TEF and more recently for Portugal Telecom (PTC). Since our last sector report, the sector has seen two rating downgrades (TI and France Télécom (FTE)) and one upgrade (KD8), and now our coverage list now has five sub-investment grade names and two companies sitting just one notch above sub-investment grade (TI and KPN).

As we have written recently, the risk profile at TI seems to us to be binary, with potential upside if the company can close on favourable transactions relating to its wireline network and mobile operations, but with potential also for a capital raising exercise if such deals cannot bring the credit agencies back on side and put a floor under the share price.

Exhibit 1.9: Credit ratings continue to be downgraded, although CDS spreads remain much tighter than a year ago

Source: Berenberg, Bloomberg

Moody's Outlook S+P Outlook

CDS spread

@

11/06/2013

CDS spread

@

11/06/2012 Change since March 2013

OTE Caa1 Stable B+ Stable 553 2885 No change

C+W Communications Ba2 NEG BB NEG 333 626 No change

Portugal Telecom Ba2 NEG BB NEG 436 919 No change

Kabel Deutschland Ba2 Stable BB Stable 196 432 S&P upgrade

Millicom Ba1 Stable na na na na No change

Telecom Italia - Ords Baa3 NEG BBB- Stable 316 461 S&P downgrade

KPN Baa2 NEG BBB- Stable 170 151 No change

Telefonica Baa2 NEG BBB NEG 229 460 No change

BT na na BBB Stable 79 109 No change

Elisa Baa2 Stable BBB Stable 90 89 No change

TDC Baa2 Stable BBB POS 96 104 No change

Telekom Austria Baa1 NEG BBB Stable 137 148 No change

Deutsche Telekom Baa1 Stable BBB+ Stable 78 105 No change

France Telecom A3 NEG BBB+ Stable 114 153 S&P downgrade

Vodafone A3 Stable A- Stable 78 92 No change

Telenor A3 Stable A- Stable 42 72 No change

TeliaSonera A3 Stable A- Stable 42 66 No change

Swisscom A2 Stable A Stable 49 89 No change

Belgacom A1 Stable A Stable na na No change


13

KPN continues to remain exposed, in our view, although this is more related to the potential for unexpected events, such as an expensive outcome to the current 900Mhz re-farming issue in Germany (a decision is due on 24 June). As noted earlier, a negative outcome in Germany could see a total spectrum cost of €4.4bn, with KPN’s exposure at €1.1bn, an amount that could again raise concerns about KPN’s balance sheet.

Cash call fears have suddenly come back into focus at PTC due to the material decline in the Oi share price this year (down 48% ytd) and the potential for its major Brazilian subsidiary Oi to abandon its dividend policy. Such a move would raise questions about how the Oi Holding company, Telemar Participacoes, would service its BRL3.3bn of debt and associated interest. A capital raising at the Telemar Participacoes level would be one way of resolving the difficulty, and that could see PTC required to contribute in order to maintain (or maybe increase) its exposure to Oi.

6) Consolidation will not save the day

Later in this report, we detail in great depth our assessment of the consolidation bull thesis. We examine the eight major European markets that still have four or more network operators, and assess the scope for consolidation and the likely remedies in each market. We also assess each market’s need to see consolidation as a mechanism for achieving market repair and stability in prices.

We conclude that there are material differences among the eight markets with markets like Germany and France potentially being well positioned with respect to consolidation remedies (see Exhibit 1.10), and markets like the Netherlands, Sweden, Denmark and Italy potentially more exposed to remedies, or less likely to see consolidation in the first place.

These are differences that could help an investor to identify the markets where consolidation remedies might at least leave the door open to market repair. Despite that, we have to acknowledge an overwhelming feeling that in-market mobile consolidation simply faces too significant a challenge from anti-trust remedies (eg spectrum divestment, infrastructure access and MVNO support) that will significantly offset, or even negate, consolidation synergies. The recent Austrian deal precedent suggests that 260MHz of spectrum could be handed back to remedy consolidation in the eight markets analysed. That is a lot of potential new entrant opportunities!

Many sceptical eyes, our own included, are now following Italy, where TI has made public its interest in seeking a combination of TIM Italy with Hutchison’s 3 Italia – a merger of the market number one with the market number four. As we show later, a combination would result in a significant increase in Italian mobile market concentration, and in spectrum concentration in a market with an underdeveloped MVNO segment, and few network-sharing deals. The risk of material remedies is high, but jurisdiction appears to lie with the local anti-trust authority, AGCM, rather than with the EC competition commission, which may allow for a more dovish outcome on remedies. If a deal is proposed, then the regulatory/anti-trust approach will be carefully scrutinised for any signs of tacit regulatory approval of market repair.


14

Exhibit 1.10: Service revenue and spectrum HHI assuming consolidation – Austrian spectrum remedy is not encouraging

Source: Berenberg

NDL

Swe

FR

DKES

UK

IT

GE

Avg.

2200

2400

2600

2800

3000

3200

3400

3600

2400 2900 3400 3900

Se

rvic

e R

ev

en

ue

HH

I

Spectrum HHI

Adjusted spectrum concentration vs market concentration (with consol.)

AUAU incl.remedies


15

The anxiety of the bears: what keeps us awake at night?

We would hate it if readers thought we were being wantonly bearish without consideration for where we might be wrong. The debate around the telecoms desk here on what we might be missing is robust. Could we be underestimating upside from mobile payments? Are we missing the potential for broadband prices to increase? Are not the worst of MTR cuts, spectrum auctions and dividend cuts behind us now? What about the scope for further cost savings to underpin earnings? Are not consensus earnings downgrades across the rest of the market as bad as, or worse than, those we are seeing in telecoms? Will regulators really not save the day? Really?

Our point is that we do stress about the bearish view, and where we might be wrong. Among the team, these are the main concerns that keep us awake at night.

1) Psst! the sector has already underperformed (ex-VOD)

The market has underperformed, but not by much.

Of the 13.55-point increase in the SXKP since the start of the year to 11 June, VOD has contributed 12.14 points. In fact, more than 100% of the sector’s increase this year is explained by UK stocks VOD and BT, which together account for 17.82 SXKP index points out of a 13.55-point rally.

Without the deal-related rally in VOD shares, the sector would hardly have moved, while the market (SXXP), with VOD excluded, would still have increased by 11.2 points. Including dividends since the start of the year, the sector would have underperformed the index by 3.6%.

We still think that given the ongoing downward pressure on estimates, the sector remains likely to underperform the broader market.

2) Revenue momentum could improve into 2014

Maybe we are misjudging the degree to which the market is prepared to look ahead. As we progress into 2014, particularly into the second half of the year, some of the pressures that have dragged on revenues should start to recede.

Mobile termination rate reductions should start to moderate from the middle of next year with many markets already close to the EC’s target of 1c per minute. As already noted, some markets appear to have stabilised service revenues once MTR effects are excluded. As highlighted earlier, we could be wrong on the impact of 4G on mobile ARPUs, although even if we are, any beneficial impact is unlikely to occur overnight – there should be time to spot emerging beneficial trends.

Prospects for southern European markets could also start to look up as we progress through 2014, if economic forecasts are to be believed. In these markets, consumers and corporates have been through a significant belt-tightening exercise to the detriment of spend on telecoms services. While we continue to believe that secular factors (such as overcapacity, too much spectrum and reckless MVNO agreements) remain an important driver of sector revenues and margins, we accept that there should be some benefit to consumer and corporate spend in the event of an economic recovery in southern European markets.

3) Remedy-lite consolidation in Italy could boost the sector

Given the pent-up desire to see some good news in telecoms, and the clearly


16

sceptical response of the market (us included) to the prospects for consolidation-driven market repair in Italy, any remote signs that a deal might actually be approved with light remedies is likely to read across positively for the sector.

We do not think it will happen, and neither, it seems, does anyone else. Why might we be wrong on this? Maybe the local anti-trust authority will adopt a more lenient approach to remedies. Maybe existing spectrum will be offered to existing players, rather than reserved for a new entrant. If that is the case, then that should significantly reduce the scope for associated remedies aimed at giving a new entrant access to incumbent operators’ infrastructure (ie towers, enforced roaming). Finally, the anti-trust authorities may decide that an underdeveloped MVNO sector is irrelevant, and therefore not enforce more favourable terms on the network operators to stimulate this segment of the market.

4) Regulators might deliver on their promises

Following our two in-depth reports last year on fibre (Fibre: dream pipe or pipe dream, dated 5 March 2012, and Fibre: dream pipe or pipe dream? The sequel, dated 26 September 2012), we took the view that EC moves to adopt a lighter touch approach to fibre regulation would turn out to be a case of pain before uncertain gain for the telecoms sector. The pain would come up front in the form of increased capex as operators invested to upgrade their copper networks to VDSL and fibre, while the anticipated gain of increased demand for superfast broadband at a higher price point was both uncertain (if our survey of fibre demand was accurate) and distant.

To pacify the competitive wireline sector (made up of over 200 smaller operators across Europe), the EC has promised that while incumbents’ next-generation fibre and VDSL services will not be priced using an ex-ante cost-based wholesale price, there will be protections for smaller players against the risk of a margin squeeze by the incumbent, and there will be a right to claim “equivalence of inputs” so that competitors can replicate the incumbent’s service offering and still make a margin.

Given that the EU is a collection of 27 states, implementation of the EC’s new fibre rules, which have yet to be published in detail, is likely to vary from country to country, at least in the early years of implementation, and this may create temporary opportunities for incumbents to exploit. Spain looks like it could be in that camp, with access to competitors’ wholesale fibre service both expensive and restricted to 30Mbs speeds. Once the regulatory landscape becomes clear, we expect to see numerous cases being pursued by competitors at both the national and the EC level as competitors seek to enforce application of the EC’s framework.

We could of course be wrong, although so far there is little evidence to suggest so. If anything, bullish hopes that operators would be allowed to raise wholesale copper unbundling rates have proven optimistic, with reductions in Germany implemented, and in Italy proposed. Overt regulatory support for the incumbent in Poland has also been challenged by both the EC, and BEREC, the pan-European regulatory body.

5) The rest of the market could be worse than telcos

As already stated, consensus OpFCF estimates for telcos’ (SXKP) 2014 OpFCF estimates have already reduced by 7% so far this year, and by 18% since the start of 2012. Yet the sector has performed broadly in line with the market. Is the market seeing a similar level of earnings downgrades?

http://www.berenberg.de/cgi-bin/content/content.cgi?rm=show_doc&session_id=351ea7f863035f69dbb9a65c152c6000&sb_userid=0&sb_eventid=0

http://www.berenberg.de/cgi-bin/content/content.cgi?rm=show_doc&session_id=133cdedc4ed8700078f8f62ffa40dc33&sb_userid=0&sb_eventid=0


17

Exhibit 1.11: Telecoms earnings revisions are in line with the broader market, but revisions in other sectors have been focused on cyclical sectors

Source: Berenberg, Bloomberg

We cannot derive market OpFCF consensus so the comparison is better made at the adjusted EPS level. Exhibit 1.11 shows the change in consensus estimates for the SXXP sectors for the current year (Y), and two prospective years (Y+1 and Y+2).

At first glance, it seems that the telecoms sector (SXKP) has seen earnings downgrades in line with those for the market, both in the current year, and over the next three years (ie telecoms earnings have been revised down 6.3% in the current year and by an average 6% for the next three years against an average 5.7% downgrade across the broader market. The sector is the sixth-worst for earnings revisions out of the 19 sectors that make up the SXXP.

The five sectors that have seen greater downgrades than telecoms are basic resources, technology, oil and gas, utilities and banks, which collectively account for 31% of the market (SXXP) by market cap, but which explain 61% of the market earnings downgrades for the next three years – ie when we look at the 5.7% average revision to market earnings, these have been highly concentrated.

In addition, four of these five sectors are still forecast to generate higher growth than the telecoms sector, perhaps with more sensitivity to the economic cycle.

So where does that leave us? Our feeling is that telecoms earnings remain exposed to further downgrade given the headwinds that we describe above. While we are not experts on the rest of the market, the fact that downgrades elsewhere have mostly been driven by cyclical sectors leaves us more comfortable with a cautious stance on telecoms – after all, the headwinds that concern us in telecoms are more structural or secular, rather than cyclical. For telecoms earnings revisions to stabilise probably requires a change to market structure via in-market consolidation. In contrast, stabilisation of earnings revisions across the rest of the market would seem more related to the outlook for the broader economy.

6) The case to be made for near-term catalysts

• Easier “comp” into H2: Easter falling in Q1 and less working days affected Q1 performance, particularly with regards to corporate revenues and coverage.

Index Cum. Avg EPS % of SXXP

Wgt. Code Sector Y Y+1 Y+2 change revision EPS revision Y+1 Y+2 Y+1 Y+2

3.2% SXPP Basic resources -17.1% -17.0% -25.2% -59.2% -19.7% 12.2% 13.4% 14.8% 9.3 8.1

3.3% SX8P Technology -6.0% -2.8% -36.0% -44.7% -14.9% 9.4% 19.7% 21.1% 16.3 13.5

7.8% SXEP Oil & gas -9.1% -8.8% -13.5% -31.4% -10.5% 15.7% 6.7% 7.0% 8.7 8.1

4.2% SX6P Utilities -5.1% -8.0% -9.7% -22.9% -7.6% 6.1% 2.5% 4.4% 11.0 10.5

12.8% SX7P Banks -10.4% -8.4% -2.4% -21.2% -7.1% 17.4% 19.2% 18.1% 9.2 7.8

4.4% SXKP Telecommunications services -6.4% -5.7% -6.0% -18.1% -6.0% 5.1% 5.3% 5.4% 10.2 9.7

11.0% SXNP Industrial goods and services -6.7% -3.5% -3.9% -14.1% -4.7% 10.0% 12.6% 11.9% 12.8 11.4

2.2% SXOP Construction and materials -5.6% -4.1% -4.1% -13.8% -4.6% 2.0% 14.6% 15.2% 12.0 10.4

11.8% SXDP Healthcare -4.8% -4.4% -4.0% -13.2% -4.4% 10.0% 8.7% 8.5% 13.5 12.4

5.0% SX4P Chemicals -6.1% -4.9% -1.8% -12.8% -4.3% 4.1% 11.2% 9.9% 13.0 11.9

8.8% SX3P Food and beverages -5.7% -2.7% -2.6% -11.0% -3.7% 6.2% 8.0% 9.9% 16.1 14.7

2.5% SXMP Media -5.0% -3.5% -1.4% -9.9% -3.3% 1.6% 10.7% 9.0% 12.5 11.4

3.2% SXRP Retail -4.2% -1.7% -1.5% -7.4% -2.5% 1.5% 10.7% 9.0% 13.0 11.9

2.8% SXAP Autos and parts -9.3% -6.6% 12.0% -3.9% -1.3% 0.7% 15.9% 16.8% 8.1 6.9

1.4% SX86P Real Estate -1.1% -0.1% -1.3% -2.5% -0.8% 0.2% 5.4% 5.3% 17.7 16.9

6.1% SXIP Insurance -0.2% -0.1% -1.6% -1.9% -0.6% 0.7% 6.4% 5.4% 8.6 8.2

6.4% SXQP Personal and household -0.2% 0.0% 0.6% 0.5% 0.2% -0.2% 8.7% 9.7% 14.8 13.5

1.5% SXTP Transport and leisure -1.0% 2.9% 3.7% 5.5% 1.8% -0.5% -2.1% 13.8% 13.3 11.7

1.5% SXFP Financial services 13.7% -3.6% 12.6% 22.7% 7.6% -2.2% 6.2% 12.6% 12.9 11.4

100.0% SXXP Market -6.0% -5.3% -5.8% -17.1% -5.7% 100.0% 10.4% 10.8% 11.4 10.3

EPS consensus change ytd EPS growth PE ratio


18

In Q2, this should unwind, with the comp aided by Easter falling in Q2 last year. The comp should become easier as we move through the year, allowing improving trends to aid stock performance.

• Market reaction to poor trends has been less severe: In Q1, we saw stocks react more positively to earnings despite continued poor underlying trends, particularly incumbent stocks (shown in Exhibit 1.12). On average, incumbent estimates were broadly in line with consensus at EBITDA in both Q4 and Q1. However, on results days we saw 2.3ppt of underperformance against the SXXP in Q4, while in Q1, this has swung to 1.1ppt of outperformance. While admittedly much of the difference may be due to disappointing 2013 guidance with Q4 numbers, Q1 share price responses did suggest the sector was relatively well underpinned. This contrasts with the ongoing reduction in consensus 2013 and 2014 earnings estimates in the wake of both Q4 and Q1 results.

• M&A speculation: Although mobile consolidation may be a hard ask, speculation and news flow on the topic could continue to support the sector near-term. Recent comments by 3 CEO Peder Ramel in Sweden and Denmark, as well as more news flow regarding TI and 3 in Italy, could keep the flame of hope burning. Increasing newsflow regarding acquisition of cable assets is also supporting the sector.

• 4G upsell opportunity: We have seen many different approaches to 4G pricing, with some markets trying to promote upsell (the UK, Germany) and others offering it for free (Sweden, Denmark). The jury is still out on whether this will succeed, but there are signs of 4G upsell working and this could be taken positively.

• EC newsflow: In the nearer term, EC newsflow is likely to be positive, with a fibre plan that should aid incumbent sentiment.

Despite nearer-term support, we remain of the view that the fundamental headwinds will outweigh potential short-term tailwinds.

Exhibit 1.12: Incumbents have seen a more positive share price response on results day in Q1, than they did in Q4 – suggesting lower investor expectations versus analysts?

Source: Berenberg ** note for an IMS with no EBITDA, we use revenue beat or miss instead

Incumbents

Q4 EBITDA vs.

consensus**

Q4 price move

vs. SXXP

Q1 EBITDA vs.

consensus**

Q1 price move

vs. SXXP

Belgacom -1.6% -5.3% -2.3% 1.4%

BT 1.5% 6.2% 3.4% 11.8%

Deutsche Telekom -2.5% -1.0% 1.4% 4.0%

France Telecom 0.0% -1.8% 0.0% 2.0%

KPN -0.7% -16.4% 0.7% -3.1%

Millicom 2.6% -7.6% -4.0% -1.1%

Mobistar 3.0% -1.0% 1.0% 0.9%

Portugal Telecom 1.1% -1.5% 1.8% 4.1%

Swisscom 1.0% -0.8% -4.0% -2.1%

TDC -0.9% -2.0% -0.2% -0.1%

Tele2 -5.0% -10.6% 2.7% 2.2%

Telecom Italia 1.0% -2.5% -1.5% 2.3%

Telefonica 2.6% 1.1% -3.9% -1.6%

Telekom Austria 3.8% 3.2% 3.0% -1.1%

Telenor -1.0% 0.2% -1.0% -0.1%

TeliaSonera -1.0% -1.2% -1.0% -2.7%

Vodafone -1.0% 1.1% 1.0% 1.1%

Average 0.2% -2.3% -0.2% 1.1%


19

Valuation and sector performance recap

The sector has appreciated by just shy of 10% ytd in absolute terms, bang in line with the market. Around a third of this performance is the result of a 24% rally in VOD, without which the sector would have modestly underperformed. Telcos have strongly underperformed utilities (by 6% ytd) but lagged healthcare (16%). For once, the European telcos sector has broadly kept pace with the performance of the US sector. The performance of the Latin American telcos index (DWLATL) has been particularly weak, declining by 10% in absolute terms since the beginning of the year. América Móvil is down 13% ytd. PTC’s Oi has almost halved.

As we show in Exhibit 1.15 below, the sector continues to trade a P/E discount to the broader market. We estimate that the telcos P/E of 10.0x represents a discount to the market of around 8%, slightly larger than the historical discount of around 4%, with the sector de-rating over the last 12 months from a P/E relative of 1.0x to the current 0.90x. As we show in Exhibit 1.16, the dividend attractiveness of the sector has deteriorated dramatically since the numerous cuts of the last few quarters. Having offered more than double the yield of the market in mid-2012, the sector yield is now around 50% higher than the broader market. Neither appeals, given the continued pressures the sector faces.

Exhibit 1.13: Sector is trading 16% below one-year average on price relative

Source: Berenberg

Exhibit 1.14: Sector is trading 13% below five-year average on price relative

Source: Berenberg

0.75

0.80

0.85

0.90

0.95

1.00

Jun 12 Jul 12 Aug 12 Sep 12 Oct 12 Nov 12 Dec 12 Jan 13 Feb 13 Mar 13 Apr 13 May 13 Jun 13

SXKP (telcos) vs SXXP (market) Price Chart – 1 year

SXKP Index / SXXP Index

0.70

0.80

0.90

1.00

1.10

1.20

1.30

1.40

Jul 08 Oct 08 Jan 09 Apr 09 Jul 09 Oct 09 Jan 10 Apr 10 Jul 10 Oct 10 Jan 11 Apr 11 Jul 11 Oct 11 Jan 12 Apr 12 Jul 12 Oct 12 Jan 13 Apr 13

SXKP (telcos) vs SXXP (market) Price Chart – 5 year

SXKP Index / SXXP Index


20

Exhibit 1.15: But sector P/E relative is only modestly (8%) below the five-year average

Source: Berenberg

Exhibit 1.16: Sector yield relative has moved back closer to its long-term average, reflecting dividend cuts across the sector

Source: Berenberg

Stocks that worked, stocks that did not

As we discuss in the previous section, telcos share prices are strongly driven by earnings momentum, by CDS spreads (especially in the case of the southern European names), with balance sheet vulnerability a further key driver of performance in recent months. Currency moves have also been a differentiator in recent months – for example, since the beginning of the year the pound has been 3-4% weaker versus the euro and the dollar (a further boost for VOD), and last month there was a significant weakening of several emerging-market currencies versus the euro (notably the Brazilian real). Lastly, M+A speculation (KD8, Verizon Wireless) and actual M+A (Virgin Media (VMED)) has had a considerable impact on performance.

0.70

0.80

0.90

1.00

1.10

1.20

1.30

European Telecommunications Sector 5-Year PE Relative (SXKP/SXXP)

SXKP Index / SXXP Index 5 yr average

Average = 0.9566

0.40

0.90

1.40

1.90

2.40

European Telecommuncations Sector Dividend Yield Relative - 10 years (SXXP/SXKP)

SXKP Index/ SXXP Index

1.5205


21

Exhibit 1.17: Outperformers ytd are characterised by exposure to cable, wireline, broadband, the UK, northern Europe and M&A

Source: Berenberg

Exhibit 1.18: Performers are much the same qtd, with underperformers exposed to southern Europe and weak mobile

Source: Berenberg

42.5%

39.2%

33.1%

31.2%

29.1%

22.8%

22.3%

20.1%

17.9%

17.5%

17.0%

14.9%

12.1%

9.8%

9.2%

8.8%

5.5%

(0.8%)

(2.7%)

(3.6%)

(7.0%)

(8.3%)

(8.8%)

(9.7%)

(14.8%)

(17.9%)

(30.7%)

(40.0%) (30.0%) (20.0%) (10.0%) 0.0% 10.0% 20.0% 30.0% 40.0% 50.0%

Kabel Deutschland

UTDI

BT

QSC

Freenet

Iliad

ZON

VOD

CWC

Drillisch

TNET

TDC

Telenor

DT

Jazztel

Swisscom

Teliasonera

TEF

Elisa

Tele2

Mobistar

FT

TKA

PT

Belgacom

TI

KPN

Total return to shareholders Year To Date (14/06/2013)

19.7%

14.5%

12.0%

10.7%

10.2%

8.1%

5.7%

3.2%

2.9%

2.3%

1.8%

(0.1%)

(0.4%)

(0.6%)

(0.7%)

(1.5%)

(1.8%)

(2.1%)

(2.5%)

(2.5%)

(3.1%)

(3.4%)

(3.6%)

(3.7%)

(4.6%)

(8.1%)

(12.3%)

(15.0%) (10.0%) (5.0%) 0.0% 5.0% 10.0% 15.0% 20.0% 25.0%

UTDI

DT

Kabel Deutschland

BT

ZON

TNET

QSC

Elisa

Mobistar

TKA

TI

Teliasonera

Tele2

VOD

CWC

Telenor

KPN

TDC

Swisscom

Belgacom

FT

Jazztel

TEF

Iliad

Freenet

Drillisch

PT

Total return to shareholders Quarter to Date (14/06/2012)


22

In summary, the best performing sub-sector once again has been the cable and challengers, with UTDI, KD8 and of course VMED after the announcement of the takeover by Liberty Media. The next best performing sub-sector was the UK (with BT +35% ytd performing even more strongly than VOD’s +25%). The Scandinavians performed broadly in line with the broader sector, with the continental Europeans lagging. Three of the bottom five performers (KPN, TI and TKA) have either raised straight equity or diluted equityholders through hybrid issuance (or both in the case of KPN).

Most-preferred telco holdings

We are currently buyers of only five out of the 30 stocks we cover in the sector. The following five are our top picks going into the end of the year – and these five, we feel, offer the best potential for capital appreciation and total return in the sector.

BT (Buy; PT £3.75): BT remains one of our top picks, with the cost-cutting programme set to remain solid through the next 12 months and the move into content looking like a winner. With no credit worries, the potential of double-digit dividend growth for the next three years and an improving operational dynamic in its core market, we maintain a Buy recommendation and a 375p price target.

TEL2B (Buy; PT SEK104.00): The company can return 50% of its market cap in the next four years. It is targeting a 10-12% 2012-15E EBITDA CAGR, and while we are less convinced by the SEK8.3bn EBITDA target (we are 6% below), we do expect a combination of EBITDA growth, flat leverage (c1.6x) and FCFE growth to support a high shareholder payout. TEL2B has gone from being a high-risk Russian growth play to a structural northern European growth story, which also happens to be cheap. Buy.

JAZ (Buy; PT €7.00): Following the Q1 2013 results, JAZ’s share price underperformed the SXKP and the IBEX 35 as investors became worried about JAZ’s ability to gain broadband market share after the weak Q1 performance. We believe that the poor broadband net additions figures were the result of exceptionally high disconnection levels inherent to the market’s recent move to integrated fixed-mobile offers. Given that around two-thirds of the broadband base do not have a lock-up contract, JAZ’s subscriber base is more vulnerable to the material shift in the market structure. However, JAZ’s capacity to attract mobile customers will improve the company’s churn profile as around 40-60% of mobile customers subscribe to a handset financing scheme with at least an 18-month lock-up period. With strong gross additions, coupled with lower churn, these dynamics will start playing in the company’s favour in Q3 2013, while Q2 2013 should exhibit some soft improvement in net additions momentum. We remain attracted to JAZ’s growth profile and feel confident on the execution of 2013 guidance.

UTDI (Buy; PT €26.00): The shares have already done well but UTDI’s valuation is still cheap. Are there are any catalysts? We see four clear positive catalysts over the next 6-12 months (see below) and while we acknowledge that a VOD/KD8 deal would be negative for sentiment, its financial impact over the next 24 months is likely to be limited. Positive catalysts include: i) the potential for earnings upgrades driven by revenue growth acceleration in the second half of 2013; ii) a special dividend for 2013 which we estimate at around €1 per share; iii) significant debt capacity in a range of €350m to €1.4bn to enhance an already strong competitive position in the applications segment through M&A; and iv) two web-hosting companies are scheduled for a sale/floatation in H2 2013 (Host


23

Europe/Endurance). These assets tend to go for 10-12x EBITDA and if one disaggregates the UTDI telecoms business at a sector multiple (which is harsh), the web-hosting business trades on 7x EBITDA. Negative catalysts include the proliferation of deeply discounted convergent offers in the German market after a VOD/KD8 deal.

Least-preferred telco holdings

We have Sell ratings on seven of the 30 stocks that we cover. Our least-preferred holdings are as follows.

TEF (Sell; PT €9.10 from €9.40): We maintain our cautious stance on TEF shares in the belief that consensus expectations remain about 3-5% too high at both the revenue and OIBDA level, that the company will continue to face headwinds both domestically and in Latin America, and that the premium valuation leaves little scope for disappointment. Adverse FX movements have come into view again and explain our 1% reductions to OIBDA estimates, while further down the P&L, we factor in higher interest costs and D&A resulting in a 5-10% reduction to our net income estimates. We reduce our price target by 30c to €9.1 per share. We do not foresee a major single negative catalyst, rather a drip-feed of ongoing pressures across the portfolio of assets, with a particular focus on domestic revenue and OIBDA pressures, and Latin American margin trends. German 900Mhz spectrum re-farm costs could have an impact in late June, while any more news of a cooperation between O2 Germany and KPN’s E+, or Spanish mobile consolidation developments could act as positive catalysts.

TALK (Sell; PT £1.60): Is BT’s offer of free sport and high-end football a risk to TALK? Listening to the company, it would appear it sees little to no risk from BT offering free premium sport content to broadband subscribers. Our work has focused on the demographics of UK football followers. This shows that there could be as many as 9m homes in the UK with an interest in free football which do not currently subscribe to Sky Sports. If we take TALK’s broadband share (excluding Sky Sports customers), it suggests that as much as half of its base could be at risk from free football. The reality is that its Essentials package is cheap enough to defend against that risk, but in our view the Plus package is exposed. Our analysis suggests that as many as 400,000 Plus customers could be at risk, with half of these out of their contract period.

ELI1V (Sell from Hold; PT €12.85): Despite the Q1 disaster with a 7% EBITDA miss, ELI1V had outperformed by 2% just before the results and 9% following the sell-off on results. This is impossible to justify, especially with new promotions available in mobile suggesting competitive intensity has not gone away. The stock is now trading on less than a 7% 2013E dividend yield, and with the company paying out close to 100% of its cash as a dividend and with estimated risks skewed to the downside, we think this looks expensive.

TKA (Sell; PT €4.20 from €4.50): TKA is about to enter a period of heavy spectrum investment which could see it spending €565m in Austria and CEE over the next few years. With leverage already at 2.9x net debt/EBITDA by end-2013, the group needs to stabilise domestic revenues and hence EBITDA to stand any chance of maintaining its prized credit rating of BBB (stable) from S&P. With material MTR/roaming cuts from July 2013 in CEE markets which come at an incremental EBITDA drop through of 70-80%, and competition from Hutchison in Austria still intense, we see downside risk to consensus expectations for EBITDA stabilisation in 2014. In addition, while a spectrum auction in September 2013 is unlikely to draw a new entrant (positive for sentiment), Hutchison remains


24

on course to reach a 30% subscriber market share by early 2015 and is relying on upselling its subscriber base to higher speeds to increase ARPU as opposed to putting through price increases. The launch of a discount no-frills MVNO headed by the ex-Orange CEO, Mr Krammer, has also been speculated on recently and will be taken negatively.

BELG (Sell; PT €16.00 from €17.00): We expect BELG to post positive mobile contract subscriber net additions over the next few quarters. Despite this, we maintain our negative stance as we think the run rate has been permanently rebased downwards following the success of TNET and that consensus is still underestimating the backbook-repricing risk. We also think that with mobile subscription contract periods capped at six months, the incentive for competitive disruption will remain high especially from TNET. For instance, after a lull period between March and mid-May, TNET has since launched a mobile VOIP product, introduced a 50% discount on mobile roaming and has restarted subsidies. If TNET succeeds in moving closer to its ambition for a service revenue market share of 18% by 2016, there is still considerable downside for BELG shares. We therefore maintain Sell.

MOBB (Sell; PT €16.00 from €16.50): With both BELG and TNET aggressively pushing discounts and device subsidies in quad-play bundles, MOBB was forced into protecting its high-value contract subscriber base by repricing its single-play mobile tariffs to levels matching its competitors’ convergent offers in April 2013. Since mid-May, the use of handset subsidies as a tactical retention tool has also increased, with MOBB planning two to three campaigns this year versus one last year. To make matters worse, since the introduction of the telecoms law in October 2012, MOBB has lost more retail contract subscribers to its MVNO partner (TNET) than it has gained in wholesale from BELG/Base combined. Given this market dynamic, we continue to see material risk to consensus estimates which assume 2013 EBITDA at the mid-point of guidance and stability thereafter.

TDC (Sell; PT DKK42.00): TDC’s pricing points look exposed as the main TLSN and 3 brands move to lower all-inclusive pricing. In addition, TDC has benefited from the removal of handset subsidies, while TNOR is now using them to be more aggressive, especially on retention. In our view, in trying to fix the market, TDC is building up a back-book issue that could come back to bite it later in the year. For now, we expect TDC’s consumer subscriber losses to accelerate, with an eventual pricing correction likely. Consolidation is very much in focus. An analysis of spectrum holdings as well as data usage shows that while 3 holds only 14% of the spectrum on an adjusted basis, it accounted for 37% of the data traffic in Denmark in H2 2012. 3 could be running close to its capacity limits, arguably increasing its appetite for M&A. But with a deal with TDC likely to be impossible, this leaves a deal with either TLSN or TNOR as the only alternative. However, TLSN and TNOR now pool spectrum together within their new JV, which would complicate any deal with 3. In addition, a deal would likely be scrutinised by the EC, and given the multiple-brand approach, we would expect needed divestments to add another layer of complexity. In short, while there may be appetite, the hurdles may be too challenging to overcome.

Other rating and price target changes

Other rating and price target changes are shown in Exhibit 1.19.


25

Exhibit 1.19: Summary of changes to ratings, earnings estimates and price targets

Source: Berenberg

New Old New Old Change? 2013 2014 2015 2013 2014 2015

BT Group Buy Buy £ 3.75 £ 3.75 N 6,101 6,231 6,377 na na na

Tele2 Buy Buy SEK 104.00 SEK 104.00 N 11,450 12,522 13,222 na na na

Jazztel Buy Buy € 7.00 € 7.00 N 179 221 272 -3.6% 1.7% -0.3%

United Internet Buy Buy € 26.00 € 26.00 N 398 487 565 na na na

QSC Buy Buy € 3.00 € 3.00 N 80 80 83 na na na

Telenor Hold Hold NOK 137.00 NOK 137.00 N 35,767 36,815 37,666 na na na

Drillisch Hold Hold € 14.00 € 14.00 N 68 77 84 na na na

Freenet Hold Hold € 18.00 € 18.00 N 355 360 361 na na na

ZON Hold Hold € 4.00 € 3.60 Y 329 342 352 4.2% 4.9% 4.8%

Telecom Italia (Ords) Hold Hold € 0.68 € 0.72 Y 10,988 10,967 11,033 -1.3% -0.8% -1.1%

Telecom Italia (Savs) Hold Hold € 0.56 € 0.59 Y 10,988 10,967 11,033 -1.3% -0.8% -1.1%

Deutsche Telekom Hold Hold € 9.90 € 10.50 Y 17,859 17,836 18,211 -1.3% -0.8% -1.1%

Millicom Hold Hold SEK 540.00 SEK 550.00 Y 1,881 1,944 2,146 -5.1% -6.1% -5.8%

France Telecom Hold Hold € 8.00 € 8.00 N 12,497 12,754 12,465 na na na

Teliasonera Hold Hold SEK 42.00 SEK 42.00 N 34,095 35,483 36,235 na na na

Vodafone Hold Hold £ 1.85 £ 1.85 N 13,353 13,353 13,594 -1.8% -0.7% -1.8%

Swisscom Hold Hold CHF 408.00 CHF 385.00 Y 4,296 4,352 4,360 na na na

C&W Communications Hold Hold £ 0.40 £ 0.42 Y 589 575 569 12.9% 13.0% 16.0%

Iliad Hold Hold € 150.00 € 128.00 Y 1,037 1,279 1,552 2.6% 1.0% -1.1%

Portugal Telecom Hold Hold € 4.45 € 4.45 N 2,200 2,202 2,259 na na na

Kabel Deutschland Hold Hold € 85.00 € 65.00 Y 822 930 1,017 na na na

Telenet Hold Hold € 25.50 € 25.50 N 837 898 945 0.0% -0.1% -0.2%

KPN Hold Sell € 1.45 € 1.65 Y 4,031 3,960 3,954 na na na

Belgacom Sell Sell € 16.00 € 17.00 Y 1,675 1,600 1,586 0.3% -3.4% -3.8%

TDC Sell Sell DKK 42.00 DKK 42.00 N 10,120 9,972 9,830 na na na

Mobistar Sell Sell € 16.00 € 16.50 Y 380 366 349 0.1% 0.2% -4.3%

Elisa Sell Hold € 12.85 € 12.85 N 505 522 527 na na na

TalkTalk Sell Sell £ 160.00 £ 160.00 N 312 393 417 na na na

Telefonica Sell Sell € 9.10 € 9.40 Y 19,353 18,863 18,928 -0.8% -0.9% -0.7%

Telekom Austria Sell Sell € 4.20 € 4.50 Y 1,315 1,253 1,261 3.0% -3.3% -3.5%

--Rating-- --Price target-- ---------- EBITDA ---------- ---- EBITDA revision ----


26

Exhibit 1.20: Where are we most different to consensus?

Source: Berenberg Note: ‘>’ means we are above consensus by at least the threshold amount stated in the column heading. ‘<’ means we are below consensus by at least the threshold amount stated in the column heading.. ‘n/c’ means no comment.

P/target Comment

Rating? <>5%? Y Y+1 Y+2 Y Y+1 Y+2 We are different due to:

Belgacom N Y N N N N < < 2-3% lower EBITDA/higher D&A leads to lower EPS

BT Y Y N > > N > > Higher revenues lead to higher EBITDA and EPS

C&W Comms N N N < < < < < Higher minority charge, leaves us lower on EPS

Deutsche Telekom N N N N N > > > Much lower D&A versus consensus

Drillisch N Y N N > < < < Higher D&A versus consensus

Elisa N Y N N N < N N Impacted by PPO acquisition, so consensus not clean

France Telecom N Y N N > N N > Higher EBITDA leads to higher EPS in Y+2

Freenet N Y N N N < N N Higher D&A in 2013 on acquisition of Gravis

Iliad Y Y N > N < N N Trajectory leaves us ahead in Y+1

Jazztel N N < N N < N N More cautious on short term profitability

Kabel Deutscheland N Y N N N N N N n/c

KPN N Y N N N N N N n/c

Millicom N N < < < < < < More cautious margin outlook

Mobistar Y N < < < < < < Lower EBITDA from tariff repricing/subcriber losses

Portugal Telecom N Y N N N < < < D&A, interest, minorities

QSC N Y N < < < < < More conservative assumptions on revenue growth

Swisscom N N N N N N N > Higher D&A by 2014 due to capex catchup

TalkTalk Y Y N < < N N < We expect TT to return to subscriber losses

TDC Y Y N N < < N < More cautious on mobile subscriber growth

Tele2 N Y N N N N N N Consensus estimates have moved up to ours

Telecom Italia N Y N N N < N N We are higher on interest costs.

Telefonica Y Y N < < N < < We're more cautous on domestic OIBDA

Telekom Austria N Y N < < N < < Domestic EBITDA hit due to competition/roaming

Telenet N Y N N N > > > Lower cash returns = lower interest expense

Telenor Y N N N N > > N n/c

TeliaSonera N Y N N N N N N n/c

United Internet N Y N N > N > > Higher revenue growth over 2013/14

Vodafone Y Y N N N < < < US disposal CGT, higher D&A and FX

ZON N N N > > N N N Expect more cost savings than consensus

EBITDA <>3% EPS <> 5%


27

Exhibit 1.20: Valuation and price target methodologies

Source: Berenberg

Belgacom We value Belgacom using a sum of the parts valuation model which incorporates both target multiples and segment

DCF valuation

British Telecoms We value British Telecoms using a sum of the parts valuation model which incorporates both target multiples and

segment DCF valuation

Cable & Wireless Communications We value Cable & Wireless Communications using a sum of the parts valuation model which incorporates both target

multiples and segment DCF valuation

Deutsche Telekom We value Deutsche Telekom using a sum of the parts valuation model which incorporates both target multiples and


Drillisch We value Drillisch using a DCF analysis

Elisa We value Elisa using a sum of the parts DCF analysis

France Telecom We value France Telecom using a sum of the parts valuation model which incorporates both target multiples and


freenet We value freenet using a DCF analysis

Iliad We value Iliad using a sum of the parts valuation model which incorporates both target multiples and segment DCF

valuation

Jazztel We value Jazztel using an average of multiple fair value, our SOTP fair value and a DCF analysis.

Kabel Deutschland We value Kabel Deutschland using a sum of the parts valuation model which incorporates both target multiples and


KPN We value KPN using a sum of the parts valuation model which incorporates both target multiples and segment DCF

valuation

Millicom We value Millicom using a sum of the parts DCF analysis

Mobistar We value Mobistar using a DCF analysis

Portugal Telecom We value Portugal Telecom using a sum of the parts valuation model which incorporates both target multiples and


QSC We value QSC using a DCF analysis

Swisscom We value Swisscom using a DCF analysis

Talktalk We value Talktalk using a sum of the parts DCF analysis

TDC We value TDC using a sum of the parts DCF analysis

Tele2 AB We value Tele2 AB using a sum of the parts DCF analysis

Telecom Italia We value Telecom Italia using a sum of the parts valuation model which incorporates both target multiples and


Telefonica We value Telefonica using a sum of the parts valuation model which incorporates both target multiples and segment

DCF valuation

Telekom Austria We value Telekom Austria using a DCF analysis

Telenet We value Telenet using a DCF analysis

Telenor We value Telenor using a sum of the parts DCF analysis

Teliasonera We value Teliasonera using a sum of the parts DCF analysis

United Internet We value United Internet using a DCF analysis

Vodafone We value Vodafone using a sum of the parts valuation model which incorporates both target multiples and segment

DCF valuation

Zon Multimedia We value Zon Multimedia using an average of multiple fair value, our SOTP fair value and a DCF analysis.


28

Exhibit 1.21: Incumbents valuation comparables

Source: Berenberg

BerenbergEuropean Equity Research: Telecommunications

Stuart Gordon +44 203 207 7858 Laura Janssens +44 203 465 2639 Wassil El Hebil +44 20 3207 7862

Paul Marsch +44 203 207 7857 Barry Zeitoune +44 203 207 7859 Usman Ghazi +44 20 3207 7824

Price Price Implied Mkt Cap

Remaining

Div.

Yield % to

Total

Returns %

Berenberg

EV/EBITDA EV/EBITDA P/E

Berenberg

EV/OpFCF

Berenberg

Real

FCFE Yield

Normalized

FCFE Yield

Net Debt/

EBITDA

Berenberg

Organic

EBITDA

CAGR %

Rating 17-Jun-13 Target Upside % LC mn 13E 13E 13E 13E 13E 13E 13E 13E 13E 12-15E

Belgacom Sell € 17.21 € 16.00 -7.0% 5,525 2.9% -4.1% 5.2x 4.4x 9.4x 10.3x 6.9% 7.3% 1.2x -3.9%

BT Group Buy £ 3.12 £ 3.75 20.0% 24,516 4.1% 24.2% 5.7x 5.2x 12.4x 9.7x 8.7% 8.9% 1.1x 2.3%

Deutsche Telekom Hold € 8.76 € 9.90 13.1% 38,977 0.0% 13.1% 5.1x 4.5x 11.5x 10.6x 8.0% 8.5% 2.4x 0.4%

France Telecom Hold € 7.36 € 8.00 8.6% 19,959 6.8% 15.4% 3.8x 3.9x 7.2x 7.9x 19.4% 18.5% 2.3x -3.4%

KPN Hold € 1.54 € 1.45 -6.0% 6,566 0.0% -6.0% 5.0x 4.2x 6.7x 9.8x 8.4% 12.2% 2.4x -6.3%

Millicom Hold SEK 509.00 SEK 540.00 6.1% 52,259 2.6% 8.7% 7.0x 5.5x 20.0x 14.5x 3.3% 3.2% 1.3x 0.5%

Portugal Telecom Hold € 3.11 € 4.45 43.1% 2,724 0.0% 43.1% 4.9x 4.9x 13.0x 12.5x 6.1% 11.4% 3.6x 0.3%

Swisscom Hold CHF 403.00 CHF 408.00 1.2% 20,876 0.0% 1.2% 6.8x 6.9x 10.6x 16.4x 5.7% 6.4% 1.8x -0.6%

TDC Sell DKK 43.70 DKK 42.00 -3.9% 34,929 3.4% -0.5% 6.0x 5.7x 12.2x 9.3x 9.5% 11.7% 2.2x -2.3%

Tele2 Buy SEK 79.45 SEK 104.00 30.9% 35,555 0.0% 30.9% 6.9x 7.2x 2.2x na 1.3% -2.4% 1.3x -10.7%

Telecom Italia (Agrg) nm nm nm nm 9,750 nm nm 3.9x 3.9x 3.4x 7.2x 15.1% 18.9% 2.5x -2.4%

Telecom Italia (Ords) Hold € 0.54 € 0.68 25.9% 7,159 0.0% 25.9% 3.9x 3.9x 4.8x 7.2x 14.0% 17.7% 2.5x -2.4%

Telecom Italia (Savs) Hold € 0.43 € 0.56 30.2% 2,591 0.0% 30.2% 3.9x 3.9x 3.9x 7.2x 17.8% 22.1% 2.5x -2.4%

Telefonica Sell € 10.03 € 9.10 -9.3% 45,667 0.0% -9.3% 5.9x 5.7x 9.2x 10.3x 8.3% 10.9% 2.6x -3.8%

Telekom Austria Sell € 5.20 € 4.20 -19.2% 2,299 0.0% -19.2% 5.5x 4.5x 16.5x 11.7x 11.0% 11.9% 2.3x -4.0%

Telenor Hold NOK 121.00 NOK 137.00 13.2% 194,222 0.0% 13.2% 5.6x 6.3x 10.7x 11.2x 5.7% 7.4% 1.1x 5.8%

Teliasonera Hold SEK 42.86 SEK 42.00 -2.0% 185,587 0.0% -2.0% 6.1x 7.0x 10.7x 10.7x 7.2% 8.6% 1.5x -0.3%

Vodafone Hold £ 1.80 £ 1.85 2.7% 88,783 1.8% 4.6% 5.5x 8.9x 12.0x 8.7x 10.7% 10.5% 2.0x -0.6%

Incumbents Sector Average 8.7% 1.3% 10.0% 5.4x 5.4x 9.8x 10.3x 9.3% 10.8% 2.0x -1.9%

Incumbents Weighted Average 5.3% 1.4% 6.7% 5.5x 6.4x 10.6x 9.8x 9.5% 10.3% 2.0x -0.7%

Telecoms Sector Average 5.2% 1.3% 6.5% 6.3x 6.4x 89.3x 11.4x 8.8% 8.6% 1.9x 2.0%

Sector Weighted Average 4.5% 1.4% 5.9% 5.8x 6.7x 18.2x 10.1x 9.0% 9.6% 2.0x 0.4%

Note : Calculation methods:

P&L items are calendarised (e.g. for UK March year-end operators, f igures show n are for prior December year-end). Dividend Yield is forw ard 12 months rolling to include interim dividends

Comps do not use rolling market cap, or rolling EV. Berenberg EV/EBITDA based on last reported Net Debt adjusted for certain items (i.e. contingent liabilities, pension obligations, certain provisions and guarantees)

NA - not available NM = not meaningful LC = Local currency EV/EBITDA based on as reported net debt and EBITDA

Weighted average uses market capitalisation converted to EUR. Berenberg PE includes mainly EBITDA adjustments f low ing through to earnings

*For Millicom apart from the price in SEK, all other data are in Dollar Berenberg EV/OpFCF based on adjusted EV (see above). OpFCF = EBITDA - Capex

Berenberg Real FCFE Yield:

FCFE: OpFCF adjusted for future investments, acquisitions, disposal gains, net interest, tax, minority dividend payments, associate dividend income

*Virgin Media price slightly different than closing price of Friday Adjusted Market Cap: Minorities grossed up using current market cap w ere available else current year EV/EBITDA multiple; added PV of future investment spend

Normalized FCFE Yield:

Based on Adjusted Market Cap (as above) and normalized FCFE (EBITDA - capex - net int - tax - FCF haircut related to specif ic issues such as cash f low repatriation)

Unlevered FCFF Yield: based on Berenberg adjusted OpFCF and tax payments at stated corp tax rate on reported OpFCF

Berenberg net debt / EBITDA based on same year net debt and EBITDA

Organic grow th assumes constant currencies and constant perimeter (excluding acquired businesses)


29


Source: Berenberg

Mngmt. Berenbg Berenbg

Upside to Total Mkt Cap Net Debt Net Debt EV EV Net Debt/ N Debt/ EBITDA/ Outlook

Rating Curr. Target (%) Returns % LC mn LC mn LC mn Rtn (%) LC mn SH Funds EBITDA Net Int. S&P Moody S&P/Mdy

Belgacom Sell EUR -7.0% -4.1% 5,525 1,847 2,473 7,372 8,788 0.6 1.0 13.8 A A1 S/S -0.5% -3.9%

BT Group Buy GBP 20.0% 24.2% 24,516 7,736 7,723 32,252 34,499 -35.7 1.3 9.0 BBB NA S/NA -2.1% 2.3%

Deutsche Telekom Hold EUR 13.1% 13.1% 38,977 39,203 38,866 80,676 80,340 1.3 2.2 7.1 BBB+ Baa1 S/S 0.3% 0.4%

France Telecom Hold EUR 8.6% 15.4% 19,959 29,517 29,517 49,476 48,200 1.1 2.1 6.6 A- A3 N/N -2.1% -3.4%

KPN Hold EUR -6.0% -6.0% 6,566 10,220 12,434 16,786 20,326 5.0 2.1 5.2 BBB Baa2 S/N -3.1% -6.3%

Millicom Hold SEK 6.1% 8.7% 52,259 15,164 13,653 67,423 10,048 1.1 0.8 -13.1 NR Ba1 NR/S 9.8% 0.5%

Portugal Telecom Hold EUR 43.1% 43.1% 2,724 7,711 8,384 10,435 11,873 2.7 3.5 9.7 BB Ba2 N/N -0.8% 0.3%

Sw isscom Hold CHF 1.2% 1.2% 20,876 9,144 8,929 30,020 29,805 2.2 2.0 12.3 na na / 0.4% -0.6%

TDC Sell DKK -3.9% -0.5% 34,929 22,648 22,648 57,577 57,577 1.0 2.1 11.8 BBB Baa2 Posit ive/stable -2.1% -2.3%

Tele2 Buy SEK 30.9% 30.9% 35,555 8,840 7,513 44,395 43,068 0.8 1.4 11.5 NA NA NA/NA -9.3% -10.7%

Telecom Italia (Agrg) nm EUR nm nm 9,750 28,274 28,020 38,024 42,840 1.2 2.4 6.7 BBB Baa2 N/N -2.0% -2.4%

Telecom Italia (Ords) Hold EUR 25.9% 25.9% 7,159 19,438 19,264 26,597 29,908 1.2 2.4 6.7 BBB Baa2 N/N -2.0% -2.4%

Telecom Italia (Savs) Hold EUR 30.2% 30.2% 2,591 8,836 8,756 11,427 12,932 1.2 2.4 6.7 BBB Baa2 N/N -2.0% -2.4%

Telefonica Sell EUR -9.3% -9.3% 45,667 57,131 59,043 102,798 111,554 1.9 2.8 5.9 BBB Baa2 N/N -3.6% -3.8%

Telekom Austria Sell EUR -19.2% -19.2% 2,299 3,677 4,943 5,975 7,242 4.4 2.5 7.4 na na S/S -0.5% -4.0%

Telenor Hold NOK 13.2% 13.2% 194,222 33,082 79,135 227,304 273,357 0.4 0.6 -19.5 A- A3 S/S 2.0% 5.8%

Teliasonera Hold SEK -2.0% -2.0% 185,587 60,807 69,887 246,395 255,474 0.6 1.6 -9.5 A- A3 S/S -1.0% -0.3%

Vodafone Hold GBP 2.7% 4.6% 88,783 26,958 29,988 115,741 124,313 0.4 2.0 11.2 A- A3 S/S 0.3% -0.6%

Average 8.7% 8.7% 1.7 2.0 5.6 - - - -0.4% -1.3%

Weighted average 5.3% 6.7% NA 1.9 5.5 - - - -0.7% -0.7%

2011A 2012E 2013E 2014E 2011A 2012E 2013E 2014E 2011A 2012E 2013E 2014E 2011A 2012E 2013E 2014E 2011A 2012E 2013E 2014E 2011A 2012E 2013E 2014E

Belgacom 12.2% 10.0% 6.9% 6.9% 11.2% 10.2% 7.3% 7.5% 16.5% 14.5% 10.3% 9.7% 9.5% 7.9% 5.7% 5.8% 9.7% 9.0% 7.2% 6.9% 11.6% 10.7% 8.5% 8.3%

BT Group 7.7% 6.5% 8.7% 9.7% 6.6% 7.8% 8.9% 8.8% 6.8% 8.0% 9.5% 9.9% 6.9% 6.5% 7.3% 8.0% 6.2% 7.0% 7.6% 7.8% 6.8% 7.6% 8.2% 8.5%

Deutsche Telekom 14.5% 15.3% 8.0% 8.4% 14.1% 12.5% 8.5% 8.7% 17.1% 16.6% 12.0% 12.4% 8.8% 8.9% 9.3% 5.0% 8.9% 7.7% 5.8% 5.6% 10.2% 10.0% 7.1% 6.9%

France Telecom 45.8% 36.8% 19.4% 25.1% 36.6% 29.2% 18.5% 20.8% 34.6% 14.0% 15.5% 18.7% 16.5% 14.5% 9.6% 11.1% 16.8% 9.2% 10.0% 11.1% 17.2% 9.4% 10.2% 11.4%

KPN 26.6% 11.0% 8.4% 4.3% 25.9% 20.8% 12.2% 11.8% 35.3% 19.4% 16.1% 13.9% 11.2% 7.6% 6.2% 4.0% 12.3% 9.6% 8.1% 6.8% 14.9% 11.6% 9.8% 8.3%

Millicom 10.7% 7.7% 3.3% 5.7% 7.4% 4.8% 3.2% 5.7% 13.3% 9.3% 6.0% 8.7% 9.4% 7.4% 4.2% 6.2% 7.8% 5.1% 3.3% 5.4% 11.3% 8.5% 6.2% 8.4%

Portugal Telecom 11.3% -6.1% 6.1% 11.0% 17.7% 9.7% 11.4% 13.5% 19.8% 11.9% 12.8% 16.1% 5.5% 1.8% 5.2% 6.4% 6.8% 6.1% 6.5% 7.0% 6.2% 6.1% 6.5% 6.9%

Sw isscom 7.9% 8.7% 5.7% 6.6% 9.1% 7.9% 6.4% 6.9% 7.5% 7.7% 8.6% 5.7% 6.6% 7.2% 5.2% 5.9% 4.8% 5.3% 5.6% 6.7% 6.2% 5.1% 5.4% 5.7%

TDC 10.5% 9.3% 9.5% 9.8% 12.4% 13.7% 11.7% 11.5% 10.6% 9.4% 9.6% 9.9% 7.3% 7.1% 7.2% 7.4% 9.5% 10.0% 8.6% 8.5% 7.3% 7.1% 7.3% 7.4%

Tele2 14.8% 12.6% 1.3% 4.4% 12.3% 11.8% -2.4% 3.3% 12.3% 13.4% -2.7% 3.5% 12.9% 11.6% 1.9% 4.5% 10.7% 10.8% -0.9% 3.8% 10.4% 11.8% -1.1% 3.6%

Telecom Italia (Agrg) 25.7% 20.3% 15.1% 12.8% 27.0% 22.0% 18.9% 17.3% 41.0% 33.5% 28.8% 26.3% 17.7% 11.1% 7.5% 7.3% 18.5% 11.4% 8.9% 9.0% 8.7% 21.0% 13.0% 10.1%

Telecom Italia (Ords) 24.2% 19.0% 14.0% 11.8% 25.4% 20.7% 17.7% 16.1% 37.7% 30.7% 26.3% 24.0% 17.3% 10.8% 7.2% 7.0% 18.1% 11.1% 8.6% 8.7% 20.5% 12.6% 9.8% 9.8%

Telecom Italia (Savs) 29.6% 23.6% 17.8% 15.3% 31.0% 25.5% 22.1% 20.3% 49.9% 41.2% 35.6% 32.7% 18.7% 11.9% 8.1% 7.9% 19.5% 12.2% 9.6% 9.6% 22.3% 13.9% 10.9% 11.0%

Telefonica 12.4% 9.2% 8.3% 9.8% 14.6% 12.3% 10.9% 10.5% 19.7% 15.9% 14.8% 14.6% 6.2% 5.7% 4.9% 5.4% 8.5% 7.9% 7.0% 6.8% 8.8% 8.3% 7.4% 7.1%

Telekom Austria 20.0% 14.8% 11.0% 6.4% 22.2% 15.8% 11.9% 7.0% 20.6% 15.4% 15.0% 11.4% 7.0% 5.6% 4.8% 4.0% 7.1% 5.8% 5.6% 5.1% 10.5% 8.5% 8.0% 7.1%

Telenor 4.6% 6.1% 5.7% 7.4% 5.2% 10.1% 7.4% 8.7% 4.8% 3.8% 5.8% 7.0% 3.5% 4.7% 4.4% 5.7% 4.2% 7.9% 5.8% 6.8% 4.4% 3.9% 5.5% 6.6%

Teliasonera 6.0% 9.1% 7.2% 8.7% 6.9% 8.9% 8.6% 9.3% 6.7% 7.0% 6.4% 7.5% 5.1% 7.3% 5.9% 7.0% 5.7% 7.3% 7.0% 7.5% 5.8% 6.1% 5.5% 6.3%

Vodafone 19.8% 6.7% 10.7% 11.5% 10.5% 10.3% 10.5% 10.8% 6.9% 5.7% 5.0% 5.7% 12.9% 3.3% 6.4% 6.5% 8.0% 7.7% 7.9% 8.0% 5.2% 4.4% 4.2% 4.3%

Average 16.5% 11.7% 9.4% 9.9% 16.1% 13.8% 11.1% 11.1% 19.2% 14.4% 13.1% 13.0% 9.5% 7.4% 6.4% 6.5% 9.6% 8.0% 7.1% 7.4% 10.6% 8.4% 7.5% 7.7%

Weighted average 16.2% 10.7% 9.5% 10.5% 13.2% 12.2% 10.3% 10.6% 13.9% 11.1% 10.0% 10.3% 9.8% 6.4% 6.5% 6.5% 8.6% 7.9% 7.2% 7.5% 8.0% 7.3% 6.4% 6.6%

8.00

408.00

1.851.80

121.00 137.00

1.45

4.45

540.00

104.00

5.20 4.20

42.0042.86

0.43

0.68

Price

Target

16.00

3.75

9.90

Berenberg Real FCFE Yield

Price

17-Jun-13

17.21

3.12

8.76

7.36

nm

1.54

3.11

509.00

79.45

10.03

0.54

43.70

9.10

403.00

0.56

nm

42.00

Organic EBITDA GrowthCredit Metrics

Rating3 Yr fcast CAGR % 3 Yr fcast CAGR %

Organic Rev Growth

Standard FCFE YieldNormalized FCFE Yield Standard Unlevered FCF YieldNormalized Unlevered FCF YieldBerenberg Real Unlevered FCF Yield


30


Source: Berenberg


Belgacom 4.6 4.9 5.2 5.5 3.9 4.1 4.4 4.6 7.3 7.7 9.4 10.7 7.3 7.7 9.4 10.7 29.8% 27.9% 26.4% 25.3% 11.6% 11.5% 13.3% 12.3%

BT Group 6.0 5.7 5.7 5.6 5.4 5.3 5.2 5.2 15.2 12.7 12.4 11.3 15.2 12.7 12.4 11.3 29.5% 32.8% 33.3% 33.7% 13.6% 14.0% 13.5% 13.7%

Deutsche Telekom 4.9 5.3 5.1 5.1 4.3 4.5 4.5 4.5 18.5 30.6 14.7 13.4 13.4 14.9 11.5 10.6 31.9% 30.9% 31.1% 30.4% 14.1% 14.1% 17.1% 16.9%

France Telecom 3.2 3.5 3.8 3.9 3.3 4.0 3.9 4.0 6.0 9.7 7.1 7.1 5.1 23.7 7.2 7.2 33.4% 28.7% 30.4% 30.3% 13.2% 13.8% 15.2% 14.3%

KPN 3.9 4.2 5.0 5.1 3.3 3.6 4.2 4.2 1.4 2.3 6.7 8.4 1.4 2.3 6.7 8.4 39.5% 37.9% 34.6% 34.2% 15.5% 17.4% 18.9% 19.1%

Millicom 5.9 6.0 7.0 6.7 4.9 5.0 5.5 5.3 13.1 10.1 19.9 22.4 9.1 15.5 20.0 22.4 43.6% 40.4% 33.3% 31.1% 13.9% 18.4% 20.5% 15.2%

Portugal Telecom 4.9 4.8 4.9 4.8 4.9 4.7 4.9 4.8 8.4 8.1 12.1 13.0 8.1 12.1 13.0 11.2 34.6% 33.4% 34.3% 34.4% 19.9% 20.0% 19.1% 17.9%

Sw isscom 6.6 6.7 6.8 6.8 6.6 6.8 6.9 6.9 6.9 10.6 12.8 13.2 13.1 13.0 10.6 12.8 40.0% 39.3% 38.1% 38.4% 18.3% 19.1% 21.3% 20.8%

TDC 5.5 5.8 6.0 6.1 5.3 5.5 5.7 5.8 12.3 13.2 11.2 10.6 15.6 11.3 12.2 11.2 41.6% 39.9% 40.0% 39.9% 13.0% 13.4% 14.3% 14.2%

Tele2 3.6 3.9 6.9 6.2 4.0 4.1 7.2 6.3 7.1 8.9 19.6 13.9 7.2 10.8 2.2 14.1 29.0% 25.1% 20.2% 22.5% 13.6% 10.5% 20.3% 15.2%

Telecom Italia (Agrg) 3.5 3.6 3.9 3.9 3.5 3.6 3.9 3.9 3.7 3.3 4.2 4.4 3.0 2.7 3.4 3.5 NA NA NA NA NA NA NA NA

Telecom Italia (Ords) 3.5 3.6 3.9 3.9 3.5 3.6 3.9 3.9 4.0 3.6 4.7 4.8 -2.2 -6.4 4.8 4.7 40.6% 39.8% 39.5% 39.5% 16.1% 17.4% 18.2% 18.5%

Telecom Italia (Savs) 3.5 3.6 3.9 3.9 3.5 3.6 3.9 3.9 3.0 2.7 3.4 3.5 -1.7 -5.1 3.9 3.7 40.6% 39.8% 39.5% 39.5% 16.1% 17.4% 18.2% 18.5%

Telefonica 5.1 5.3 5.9 6.0 4.9 5.1 5.7 5.8 8.2 9.4 11.1 11.0 8.4 11.5 9.2 9.4 35.1% 33.7% 33.5% 33.4% 14.5% 14.2% 14.4% 14.8%

Telekom Austria 4.7 5.0 5.5 5.8 3.9 4.1 4.5 4.8 18.6 17.6 16.5 20.3 18.6 17.6 16.5 20.3 34.3% 32.7% 31.3% 30.5% 16.6% 15.6% 16.3% 16.5%

Telenor 6.9 5.8 5.6 5.4 7.4 6.9 6.3 6.0 14.8 9.6 9.3 8.8 27.4 21.5 10.7 9.7 31.0% 32.3% 34.9% 35.6% 13.5% 16.6% 16.0% 13.6%

Teliasonera 5.8 5.8 6.1 6.0 6.6 6.8 7.0 6.9 10.1 8.9 10.0 9.5 10.1 9.3 10.7 9.9 35.5% 34.5% 35.0% 35.4% 16.7% 14.4% 14.6% 13.7%

Vodafone 5.6 5.6 5.5 5.4 8.0 8.7 8.9 8.7 12.1 11.5 12.0 11.2 12.1 11.5 12.0 11.2 31.2% 29.9% 29.5% 29.8% 13.8% 13.9% 14.3% 14.1%

Average 5.0 5.1 5.4 5.4 5.0 5.1 5.3 5.3 10.0 10.5 10.8 11.2 10.1 10.8 10.7 10.9 35.8% 34.6% 34.0% 33.8% 15.0% 15.7% 16.6% 15.9%

Weighted average 5.3 5.3 5.5 5.4 6.0 6.3 6.4 6.3 11.3 12.3 11.4 10.9 11.7 12.5 10.6 10.4 32.3% 31.4% 31.3% 31.4% 14.0% 14.3% 15.1% 14.6%


Belgacom 12.7% 14.5% 8.7% 8.7% 1.5% -0.3% -0.1% 0.0% 14.2% 14.2% 8.6% 8.7% 7.6 8.3 10.3 10.6 11.8% 12.6% 13.4% 12.3% 0.8 1.0 1.2 1.3

BT Group 2.4% 2.8% 3.2% 3.3% -0.1% 0.5% 1.2% 1.2% 2.3% 3.3% 4.3% 4.6% 11.7 10.6 9.7 9.5 13.6% 14.1% 14.0% 14.3% 1.4 1.3 1.1 0.9

Deutsche Telekom 8.0% 8.0% 8.0% 5.7% 0.0% 0.0% 0.0% 0.0% 8.0% 8.0% 8.0% 5.7% 8.2 8.7 10.6 10.8 2.8% 2.4% 2.2% 0.3% 2.4 2.4 2.4 2.4

France Telecom 18.9% 10.6% 10.9% 10.9% 0.0% 0.0% 0.0% 0.0% 18.9% 10.6% 10.9% 10.9% 5.1 8.3 7.9 7.3 14.4% 14.2% 15.9% 15.0% 2.1 2.5 2.3 2.3

KPN 54.9% 7.8% 0.0% 0.0% 43.4% 0.0% 0.0% 0.0% 98.4% 7.8% 0.0% 0.0% 6.4 8.4 9.8 11.7 16.3% 20.3% 19.5% 21.6% 2.3 2.5 2.4 2.6

Millicom 7.0% 7.3% 6.4% 8.4% 6.3% 2.4% 0.0% 0.0% 13.2% 9.7% 6.4% 8.4% 7.8 9.8 14.5 11.1 15.3% 23.1% 22.5% 15.7% 0.8 1.0 1.3 1.4

Portugal Telecom 20.9% 10.5% 10.5% 10.5% 19.9% 0.9% 0.0% 0.0% 45.0% 11.3% 10.5% 10.5% 13.1 13.4 12.5 11.4 25.3% 25.7% 20.4% 19.2% 3.1 3.5 3.6 3.6

Sw isscom 5.5% 5.5% 5.5% 5.5% 0.0% 0.0% 0.0% 0.0% 5.5% 5.5% 5.5% 5.5% 14.6 13.4 16.4 15.0 18.5% 19.3% 21.7% 21.2% 1.8 1.8 1.8 1.8

TDC 10.0% 10.5% 8.5% 8.5% 0.0% 2.1% 0.0% 0.0% 10.0% 12.7% 8.5% 8.5% 8.0 8.7 9.3 9.4 14.1% 11.3% 14.6% 14.5% 1.9 2.1 2.2 2.2

Tele2 16.4% 8.9% 45.3% 13.8% 0.0% 0.0% 0.0% 0.0% 16.4% 8.9% 45.3% 13.8% 6.9 6.2 152.9 17.6 17.5% 11.1% 20.6% 15.2% 1.0 1.4 1.3 1.5

Telecom Italia (Agrg) 12.1% 9.3% 4.6% 4.6% 0.0% 0.0% 0.0% 0.0% 12.1% 9.3% 4.6% 4.6% 5.9 6.6 7.2 7.3 16.8% 18.2% 19.0% 19.3% 2.5 2.4 2.5 2.3

Telecom Italia (Ords) 10.7% 8.1% 3.6% 3.6% 0.0% 0.0% 0.0% 0.0% 10.7% 8.1% 3.6% 3.6% 5.9 6.6 7.2 7.3 16.8% 18.2% 19.0% 19.3% 2.5 2.4 2.5 2.3

Telecom Italia (Savs) 16.0% 12.7% 7.1% 7.1% 0.0% 0.0% 0.0% 0.0% 16.0% 12.7% 7.1% 7.1% 5.9 6.6 7.2 7.3 16.8% 18.2% 19.0% 19.3% 2.5 2.4 2.5 2.3

Telefonica 13.0% 0.0% 7.5% 7.9% 0.0% 0.0% 0.0% 0.0% 13.0% 0.0% 7.5% 7.9% 8.6 9.2 10.3 10.8 20.2% 15.1% 14.6% 14.9% 2.6 2.4 2.6 2.5

Telekom Austria 7.3% 1.0% 1.0% 1.0% 0.0% 0.0% 0.0% 0.0% 7.3% 1.0% 1.0% 1.0% 9.6 11.5 11.7 12.6 19.2% 14.3% 34.6% 18.4% 2.2 2.2 2.3 2.3

Telenor 4.1% 5.0% 5.8% 6.6% 2.3% 2.1% 3.3% 3.5% 6.5% 7.0% 9.1% 10.1% 14.8 10.2 11.2 9.3 14.4% 22.5% 21.8% 14.2% 0.6 1.0 1.1 1.2

Teliasonera 6.6% 6.6% 7.9% 8.6% 5.4% 0.0% 0.0% 0.0% 12.0% 6.6% 7.9% 8.6% 12.5 10.0 10.7 10.0 15.2% 6.1% 13.4% 11.6% 1.7 1.6 1.5 1.4

Vodafone 7.5% 5.7% 5.7% 5.7% 6.2% 1.7% 1.0% 0.0% 13.7% 7.4% 6.7% 5.7% 8.8 9.1 8.7 8.6 14.4% 26.7% 15.2% 15.0% 1.7 2.0 2.0 1.7

Average 12.8% 7.3% 6.3% 6.4% 5.3% 0.6% 0.3% 0.3% 18.4% 7.9% 6.6% 6.7% 9.3 9.6 10.5 10.2 15.6% 16.5% 17.6% 15.4% 1.9 2.0 2.0 2.0

Weighted average 9.5% 5.7% 6.8% 6.3% 3.4% 0.8% 0.7% 0.4% 12.9% 6.5% 7.4% 6.7% 9.4 9.3 11.6 9.8 14.3% 17.6% 15.0% 13.7% 1.8 2.0 2.0 1.8

Berenberg Prop. EV/EBITDA Berenberg Adj. PE

Dividend Yield % Other Shareholder Returns Total Return to S/H Berenberg EV/OpFCF Net Debt/EBITDATotal Investment/Sales

EV/EBITDA EBITDA Margin (%) Capex to Sales (%)Standard PE


31

Exhibit 1.24: Cable valuation comparables

Source: Berenberg


Stuart Gordon +44 203 207 7858 Laura Janssens +44 203 465 2639 Usman Ghazi +44 20 3207 7824

Paul Marsch +44 203 207 7857 Barry Zeitoune +44 203 207 7859 Wassil El Hebil +44 20 3207 7862


Remaining

Div.

Yield % to

Total

Returns %

Berenberg


Berenberg

EV/OpFCF

Berenberg

Real

FCFE Yield

Normalize

d FCFE

Yield

Net Debt/

EBITDA

Berenberg

Organic

EBITDA

CAGR %


Kabel Deutschland Hold € 82.69 € 85.00 2.8% 7,320 3.0% 5.8% 11.1x 11.0x 23.9x 32.5x 1.2% 0.9% 3.1x 9.2%

Telenet Hold € 34.00 € 25.50 -25.0% 3,959 0.0% -25.0% 9.0x 9.0x 27.9x 15.9x 5.5% 3.5% 4.3x 6.7%

ZON Hold € 3.60 € 4.00 11.1% 1,113 0.0% 11.1% 6.5x 5.5x 23.8x 11.5x 8.5% 8.3% 1.6x 4.0%

Cable Sector Average -3.7% 1.0% -2.7% 8.9x 8.5x 25.2x 20.0x 5.0% 4.2% 3.0x 6.6%

Cable Weighted Average -5.3% 1.8% -3.6% 10.0x 9.9x 25.1x 25.3x 3.2% 2.4% 3.4x 7.9%


Sector Weighted Average 4.5% 1.4% 5.9% 5.8x 6.7x 18.2x 10.1x 9.0% 9.6% 2.0x 0.4%Note :

P&L items are calendarised (e.g. for UK March year-end operators, figures shown are for prior December year-end).

Comps do not use rolling market cap, or rolling EV.

NA - not available NM = not meaningful LC = Local currency

Weighted average uses market capitalisation converted to EUR.

*For Virgin Media, apart from the price, all other data are in Sterling

Calculation methods:

Dividend Yield is forward 12 months rolling to include interim dividends

Berenberg EV/EBITDA based on last reported Net Debt adjusted for certain items (i.e. contingent liabilities, pension obligations, certain provisions and guarantees)

EV/EBITDA based on as reported net debt and EBITDA

Berenberg PE includes mainly EBITDA adjustments flowing through to earnings

Berenberg EV/OpFCF based on adjusted EV (see above). OpFCF = EBITDA - Capex



Adjusted Market Cap: Minorities grossed up using current market cap were available else current year EV/EBITDA multiple; added PV of future investment spend


Based on Adjusted Market Cap (as above) and normalized FCFE (EBITDA - capex - net int - tax - FCF haircut related to specific issues such as cash flow repatriation)



Organic growth assumes constant currencies and constant perimeter (excluding acquired businesses)


32

Exhibit 1.25: Cable valuation comparables

Source: Berenberg



18-Jun-13 Rating Curr. Target (%) Rtn (%) LC mn LC mn LC mn LC mn LC mn SH Funds EBITDA Net Int. S&P Moody S&P/Mdy

Kabel Deutschland Hold EUR 2.8% 5.8% 7,320 2,985 3,025 10,305 10,372 -1.8 3.5 3.1 BB Ba2 S/S 6.6% 9.2%

Telenet Hold EUR -25.0% -25.0% 3,959 3,637 3,744 7,597 7,703 -5.1 4.7 2.4 B+ Ba3 S/NR 8.6% 6.7%

ZON Hold EUR 11.1% 11.1% 1,113 707 650 1,819 1,763 3.0 1.9 12.1 na na / 1.7% 4.0%

Average -3.7% -2.7% -3.5 3.4 5.9 - - - 5.6% 6.6%

Weighted average -5.3% -3.6% NA 3.7 3.7 - - - 6.8% 7.9%


Kabel Deutschland 1.5% 2.1% 1.2% 2.1% 1.6% 1.5% 0.9% 1.1% 1.2% 1.6% 0.7% 1.7% 3.2% 3.3% 2.6% 3.3% 2.3% 2.3% 2.1% 2.6% 3.0% 3.0% 2.3% 3.1%

Telenet 5.6% 5.7% 5.5% 6.9% 4.4% 4.4% 3.5% 4.6% 6.2% 6.3% 6.3% 7.8% 3.3% 3.5% 3.7% 4.3% 3.6% 3.9% 4.1% 4.8% 5.5% 5.8% 6.3% 7.2%

ZON 6.3% 8.4% 8.5% 8.2% 7.0% 8.3% 8.3% 8.3% 11.1% 12.3% 13.9% 13.2% 3.4% 5.2% 5.1% 5.7% 4.6% 5.8% 6.2% 6.6% 6.3% 7.4% 8.3% 8.5%

Average 4.4% 5.4% 5.0% 5.8% 4.3% 4.7% 4.2% 4.7% 6.2% 6.8% 7.0% 7.6% 3.3% 4.0% 3.8% 4.4% 3.5% 4.0% 4.1% 4.7% 4.9% 5.4% 5.6% 6.3%



Kabel Deutschland 13.3 12.2 11.1 10.1 13.2 12.1 11.0 10.1 55.7 23.8 23.9 21.8 55.7 23.8 23.9 21.8 46.1% 46.2% 47.4% 48.5% 23.3% 25.0% 28.7% 28.8%

Telenet 10.5 9.7 9.0 8.4 10.5 9.7 9.0 8.4 28.4 32.3 27.9 23.3 28.4 32.3 27.9 23.3 52.6% 52.2% 50.2% 50.0% 21.4% 21.4% 21.1% 19.7%

ZON 6.7 6.6 6.5 6.1 5.8 5.8 5.5 5.3 32.7 30.8 23.8 18.4 32.7 30.8 23.8 18.4 29.1% 31.1% 31.4% 32.5% 16.6% 14.3% 13.6% 14.0%

Average 10.2 9.5 8.9 8.2 9.8 9.2 8.5 7.9 38.9 29.0 25.2 21.2 38.9 29.0 25.2 21.2 42.6% 43.2% 43.0% 43.7% 20.4% 20.2% 21.1% 20.8%



Kabel Deutschland 0.0% 1.8% 3.5% 4.1% 0.8% 0.0% 0.0% 0.0% 0.8% 1.8% 3.5% 4.1% 30.6 30.1 32.5 26.1 25.2% 25.0% 28.7% 28.8% 3.7 3.5 3.1 3.0

Telenet 12.5% 23.3% 0.0% 0.0% 0.0% 0.9% 1.3% 0.0% 12.5% 24.1% 1.3% 0.0% 18.1 17.1 15.9 13.9 21.3% 21.4% 21.1% 19.7% 3.6 3.8 4.3 3.7

ZON 4.4% 4.4% 3.3% 7.2% 0.0% 0.0% 0.0% 0.0% 4.4% 4.4% 3.3% 7.2% 15.6 12.2 11.5 10.8 16.6% 14.3% 13.6% 14.0% 2.0 1.9 1.6 1.5

Average 5.6% 9.8% 2.3% 3.8% 0.3% 0.3% 0.4% 0.0% 5.9% 10.1% 2.7% 3.8% 21.4 19.8 20.0 16.9 21.0% 20.3% 21.1% 20.8% 3.1 3.1 3.0 2.7

Weighted average 4.4% 8.9% 2.4% 3.1% 0.5% 0.3% 0.4% 0.0% 4.9% 9.2% 2.8% 3.1% 25.3 24.4 25.3 20.8 23.2% 22.9% 24.9% 24.6% 3.5 3.5 3.4 3.1

Total Investment/Sales Net Debt/EBITDADividend Yield % Other Shareholder Returns Total Return to S/H Berenberg EV/OpFCF

Normalized Unlevered FCF Yield Standard Unlevered FCF Yield

Berenberg Prop. EV/EBITDA EV/EBITDA Berenberg Adj. PE Standard PE EBITDA Margin (%) Capex to Sales (%)

Berenberg Real FCFE Yield Normalized FCFE Yield Berenberg Real Unlevered FCF Yield

82.69 85.00

34.00 25.50

Standard FCFE Yield

3.60 4.00

Credit Metrics Organic Rev Growth Organic EBITDA Growth

Price Rating3 Yr fcast CAGR % 3 Yr fcast CAGR %

17-Jun Target


33

Exhibit 1.26: Other valuation comparables

Source: Berenberg


Stuart Gordon +44 203 207 7858 Laura Janssens +44 203 465 2639 Wassil El Hebil +44 20 3207 7862

Paul Marsch +44 203 207 7857 Barry Zeitoune +44 203 207 7859 Usman Ghazi +44 20 3207 7824


Remaining

Div.

Yield % to

Total

Returns %

Berenberg


Berenberg

EV/OpFCF

Berenberg

Real

FCFE Yield

Normalized

FCFE Yield

Net Debt/

EBITDA

Berenberg

Organic

EBITDA

CAGR %


C&W Communications Hold £ 0.40 £ 0.40 -0.3% 1,607 8.3% 8.0% 5.5x 5.6x 2311.5x 10.0x 40.9% 7.6% 1.3x -4.9%

Elisa Sell € 15.17 € 12.85 -15.3% 2,363 0.0% -15.3% 6.7x 6.7x 12.1x 11.2x 8.1% 8.6% 2.0x -0.6%

Iliad Hold € 159.40 € 150.00 -5.9% 9,386 0.2% -5.7% 10.0x 10.0x na na -2.2% -1.5% 1.5x 22.0%

Drillisch Hold € 12.11 € 14.00 15.6% 620 0.0% 15.6% 7.3x 8.3x 4.1x 11.8x 5.9% 6.4% 1.9x 16.6%

Freenet Hold € 16.76 € 18.00 7.4% 2,146 0.0% 7.4% 6.6x 7.5x 10.1x 7.7x 9.1% 10.2% 1.2x 1.7%

Jazztel Buy € 5.85 € 7.00 19.7% 1,467 0.0% 19.7% 6.9x 8.5x 25.6x na 0.5% -0.8% 1.3x 16.3%

Mobistar Sell € 16.30 € 16.00 -1.8% 978 0.0% -1.8% 3.8x 3.7x 7.8x 8.2x 7.6% 11.7% 1.1x -6.5%

QSC Buy € 2.69 € 3.00 11.5% 333 0.0% 11.5% 5.5x 4.9x 16.4x 7.6x 7.6% 5.9% 0.5x 13.3%

TalkTalk Sell € 223.10 € 160.00 -28.3% 2,102 6.7% -21.6% 8.2x 8.6x 17.0x 13.2x 7.0% 5.8% 1.2x -3.7%

United Internet Buy € 22.35 € 26.00 16.3% 4,365 0.0% 16.3% 11.1x 11.6x 21.2x 13.3x 4.9% 5.6% 0.3x 20.1%

Other Telcos Sector Average 1.9% 1.5% 3.4% 7.2x 7.5x 269.5x 10.4x 8.9% 5.9% 1.2x 7.4%

Other Telcos Weighted Average -1.5% 1.2% -0.3% 8.7x 9.0x 121.3x 6.5x 4.8% 3.6% 1.2x 12.3%


Sector Weighted Average 4.5% 1.4% 5.9% 5.8x 6.7x 18.2x 10.1x 9.0% 9.6% 2.0x 0.4%

Note : Calculation methods:

P&L items are calendarised (e.g. for UK March year-end operators, f igures show n are for prior December year-end). Dividend Yield is forw ard 12 months rolling to include interim dividends

Comps do not use rolling market cap, or rolling EV. Berenberg EV/EBITDA based on last reported Net Debt adjusted for certain items (i.e. contingent liabilities, pension obligations, certain provisions and guarantees)

NA - not available NM = not meaningful LC = Local currency EV/EBITDA based on as reported net debt and EBITDA

Weighted average uses market capitalisation converted to EUR. Berenberg PE includes mainly EBITDA adjustments f low ing through to earnings

*For CWC, apart from the price, all other data are in Dollar Berenberg EV/OpFCF based on adjusted EV (see above). OpFCF = EBITDA - Capex



*Virgin Media price slightly different than closing price of Friday Adjusted Market Cap: Minorities grossed up using current market cap w ere available else current year EV/EBITDA multiple; added PV of future investment spend


Based on Adjusted Market Cap (as above) and normalized FCFE (EBITDA - capex - net int - tax - FCF haircut related to specif ic issues such as cash f low repatriation)



Organic grow th assumes constant currencies and constant perimeter (excluding acquired businesses)


34

Exhibit 1.27: Other valuation comparables

Source: Berenberg



Rating Curr. Target (%) R e t urns (% ) LC mn LC mn LC mn Rtn (%) LC mn SH Funds EBITDA Net Int. S&P Moody S&P/Mdy

C&W Communications Hold GBP* -0.3% 8.0% 1,607 1,651 541 3,258 3,212 1.5 2.5 4.7 BB Ba2 N/N -4.9% -4.9%

Elisa Sell EUR -15.3% -15.3% 2,363 909 900 3,272 3,262 1.0 1.6 -12.7 BBB Baa2 S/S -0.2% -0.6%

Iliad Hold EUR -5.9% -5.7% 9,386 969 969 10,355 10,355 0.6 1.1 18.2 NA NA NA/NA 13.8% 22.0%

Drillisch Hold EUR 15.6% 7.4% 620 (42) (41) 562 563 -0.3 -0.7 na 0.0 0.0 / na 16.6%

Freenet Hold EUR 7.4% 7.4% 2,146 512 512 2,658 2,658 0.4 1.4 8.5 n/a n/a n/a/n/a 2.9% 1.7%

Jazztel Buy EUR 19.7% 19.7% 1,467 48 48 1,515 1,515 0.1 0.3 6.2 0.0 0.0 / 10.9% 16.3%

Mobistar Sell EUR -1.8% -1.8% 978 428 467 1,406 1,445 1.2 0.9 45.8 na na na/na 3.5% -6.5%

QSC Buy EUR 11.5% 11.5% 333 55 55 388 388 NA 0.7 33.8 na na na/na -1.0% 13.3%

TalkTalk Sell GBP -28.3% 11.5% 2,102 438 444 2,540 2,426 1.1 1.6 15.3 na na na/na 0.7% -3.7%

United Internet Buy EUR 16.3% 16.3% 4,365 257 149 4,623 4,419 1.3 0.8 28.7 na na na/na 10.4% 20.1%

Average 1.9% 5.9% 0.8 1.0 16.5 - - - 4.0% 7.4%

Weighted average -1.5% 2.9% NA 1.1 15.5 - - - 7.7% 12.3%


C&W Communications 5.8% 0.9% 40.9% 15.8% 11.7% 6.2% 7.6% 6.5% 19.2% 5.5% 10.5% 9.1% 10.3% 5.7% 4.5% 3.5% 12.1% 8.0% 7.5% 6.8% 11.9% 7.8% 7.4% 6.7%

Elisa 8.6% 6.5% 8.1% 8.9% 9.3% 9.1% 8.6% 9.3% 10.0% 8.9% 8.3% 9.3% 7.0% 5.4% 6.6% 7.2% 7.5% 7.3% 6.9% 7.5% 7.9% 7.1% 6.6% 7.4%

Iliad -4.2% -3.3% -2.2% 0.8% -5.5% -2.9% -1.5% 0.2% -4.8% -3.5% -1.5% 0.3% 0.7% -1.6% -0.5% -0.2% -2.1% -0.7% 0.5% 2.0% -2.1% -0.7% 0.5% 2.0%

Drillisch 6.0% -1.4% 5.9% 9.3% 9.3% 3.9% 6.4% 7.4% 6.6% 0.2% 5.9% 9.4% 3.0% 0.3% 5.8% 8.7% 3.6% 2.0% 5.9% 8.8% 3.6% 2.0% 5.9% 8.8%

Freenet 10.5% 10.6% 9.1% 10.8% 10.7% 9.7% 10.2% 9.7% 12.4% 11.7% 10.9% 11.1% 9.0% 9.8% 9.1% 10.2% 8.8% 8.7% 8.7% 8.8% 7.8% 7.8% 7.8% 7.8%

Jazztel 3.4% 1.9% 0.5% -1.4% 1.7% 2.3% -0.8% -4.3% 2.2% 3.1% -13.9% -13.9% 3.4% 1.7% 5.5% 4.2% 2.7% 3.6% 1.5% 0.2% 2.2% 2.2% 2.9% -8.6%

Mobistar 24.9% 17.6% 7.6% 5.8% 22.2% 21.4% 11.7% 10.2% 24.9% 12.5% 7.6% 5.8% 14.9% 12.1% 6.0% 4.9% 15.6% 14.5% 9.7% 8.6% 16.0% 14.9% 8.3% 7.2%

QSC 15.3% 8.4% 7.6% 9.6% 9.5% 7.1% 5.9% 10.7% 15.3% 8.4% 7.6% 9.6% 13.4% 7.6% 7.0% 8.7% 17.8% 10.5% 8.6% 10.6% 7.4% 4.5% 4.7% 9.7%

TalkTalk 6.3% 8.3% 7.0% 9.6% 5.5% 6.2% 5.8% 8.6% 7.1% 9.0% 7.5% 9.3% 5.6% 7.3% 6.2% 8.3% 4.9% 5.5% 5.2% 7.5% 6.3% 7.9% 6.7% 8.1%

United Internet 3.9% 5.7% 4.9% 7.1% 4.8% 4.4% 5.6% 7.0% 3.0% 4.1% 4.9% 6.9% 5.3% 5.6% 5.4% 6.9% 4.4% 4.1% 5.6% 6.9% 4.3% 4.0% 5.3% 6.6%

Average 8.0% 5.5% 8.9% 7.6% 7.9% 6.7% 5.9% 6.6% 9.6% 6.0% 4.8% 5.7% 7.3% 5.4% 5.6% 6.2% 7.5% 6.4% 6.0% 6.8% 6.5% 5.8% 5.6% 5.6%



C&W Communications 3.7 4.8 5.5 5.5 3.6 4.9 5.6 5.6 14.3 -88.1 2311.5 -49.1 14.3 -88.1 2,311.5 -49.1 31.4% 30.6% 30.7% 30.7% 12.9% 15.2% 13.7% 15.2%

Elisa 6.4 6.5 6.7 6.6 6.5 6.5 6.7 6.6 11.8 11.4 11.7 11.1 11.7 11.4 12.1 11.1 33.1% 32.3% 32.0% 31.9% 12.3% 12.2% 12.9% 12.5%

Iliad 12.4 12.1 10.0 8.1 12.4 12.1 10.0 8.1 35.9 56.1 37.1 24.0 35.9 56.1 37.1 24.0 39.3% 27.0% 27.2% 29.9% 55.0% 30.4% 25.2% 22.4%

Drillisch 9.1 8.6 7.3 6.5 10.9 9.1 8.3 7.3 17.5 17.3 15.2 13.7 16.8 28.2 4.1 14.4 14.7% 19.1% 23.2% 25.4% 0.8% 7.1% 5.3% 1.0%

Freenet 7.0 6.9 6.6 6.6 7.9 7.4 7.5 7.4 8.4 8.6 10.1 8.7 8.4 8.6 10.1 8.7 10.3% 11.6% 11.0% 10.9% 0.6% 0.7% 0.6% 0.6%

Jazztel 8.8 7.1 6.9 5.6 10.9 8.8 8.5 6.8 28.9 23.4 25.6 18.7 28.9 23.4 25.6 18.7 18.5% 19.0% 17.5% 19.7% 12.1% 12.0% 35.8% 35.1%

Mobistar 2.7 2.9 3.8 3.9 2.7 2.8 3.7 3.8 4.4 5.3 7.8 9.5 4.4 5.3 7.8 9.5 31.2% 29.0% 23.5% 23.6% 10.8% 9.0% 12.4% 13.5%

QSC 5.5 5.7 5.5 4.1 4.9 5.0 4.9 4.9 13.3 19.4 16.4 15.5 13.3 19.4 16.4 15.5 16.7% 16.2% 17.5% 17.2% -6.1% -6.9% -8.1% -8.3%

TalkTalk 8.8 8.0 8.2 7.1 8.1 8.4 8.6 7.5 13.9 14.8 15.2 12.0 16.9 19.3 17.0 12.2 16.1% 18.3% 17.2% 19.8% 6.3% 6.0% 6.5% 6.2%

United Internet 13.2 13.6 11.1 9.1 13.8 14.2 11.6 9.5 35.6 31.8 21.0 15.9 35.7 32.1 21.2 16.0 16.0% 13.6% 15.0% 16.4% 2.6% 2.7% 2.4% 2.2%

Average 7.8 7.6 7.2 6.3 8.2 7.9 7.5 6.8 18.4 10.0 247.2 8.0 18.6 11.6 246.3 8.1 22.7% 21.7% 21.5% 22.5% 10.7% 8.8% 10.7% 10.0%



C&W Communications 12.6% 8.3% 6.2% 6.7% 0.0% 0.0% 0.0% 0.0% 12.6% 8.3% 6.2% 6.7% 6.2 9.4 10.0 11.0 19.0% 16.4% 14.4% 15.2% 1.6 2.4 1.3 0.9

Elisa 11.2% 8.6% 7.9% 8.2% 0.0% 0.0% 0.0% 0.0% 11.2% 8.6% 7.9% 8.2% 10.3 10.5 11.2 10.8 12.0% 12.1% 19.2% 12.5% 1.6 1.7 2.0 1.9

Iliad 0.2% 0.2% 0.3% 0.2% 0.0% 0.0% 0.0% 0.0% 0.2% 0.2% 0.3% 0.2% -35.6 -119.8 1,781.1 28.2 54.4% 30.4% 25.2% 22.4% 1.2 1.5 1.5 1.2

Drillisch 5.4% 10.1% 10.1% 10.1% 0.0% 4.8% 1.3% 1.6% 5.4% 14.9% 11.4% 11.6% 19.7 34.3 11.8 8.0 0.9% 7.1% 5.3% 1.0% -0.2 1.8 1.9 1.9

Freenet 7.2% 8.1% 8.9% 8.9% 0.0% 0.0% 0.0% 7.2% 7.2% 8.1% 8.9% 8.9% 7.6 7.7 7.7 7.6 0.7% 0.7% 1.2% 0.6% 1.6 1.3 1.2 1.1

Jazztel 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 31.4 23.8 57.8 347.4 12.1% 12.0% 15.0% 19.3% 0.6 0.3 1.3 2.0

Mobistar 22.7% 11.0% 11.0% 11.0% 0.0% 0.0% 0.0% 0.0% 22.7% 11.0% 11.0% 11.0% 4.2 4.5 8.2 9.5 11.3% 11.1% 14.2% 15.4% 0.6 0.8 1.1 1.3

QSC 3.0% 3.3% 3.7% 4.1% 0.0% 0.0% 0.0% 0.0% 3.0% 3.3% 3.7% 4.1% 6.4 7.3 7.6 7.9 6.1% 6.9% 8.1% 8.3% 0.4 0.7 0.5 0.3

TalkTalk 3.5% 4.5% 5.6% 6.9% 0.0% 0.0% 0.0% 0.0% 3.5% 4.5% 5.6% 6.9% 14.5 11.8 13.2 10.4 6.3% 6.0% 6.5% 6.2% 1.6 1.3 1.2 0.8

United Internet 1.3% 1.3% 6.3% 2.7% 7.2% 0.0% 0.2% 0.0% 8.5% 1.3% 6.5% 2.7% 15.7 16.9 13.3 10.5 2.9% 5.6% 2.5% 2.2% 1.4 0.8 0.3 0.2

Average 6.7% 5.5% 6.0% 5.9% 0.7% 0.5% 0.2% 0.9% 7.4% 6.0% 6.2% 6.0% 8.0 0.6 192.2 45.1 12.6% 10.8% 11.2% 10.3% 1.0 1.3 1.2 1.1

Weighted average 4.0% 3.4% 4.2% 3.8% 1.2% 0.1% 0.1% 0.6% 5.2% 3.5% 4.3% 3.8% -4.5 -35.5 666.0 36.1 24.5% 16.1% 14.6% 13.0% 1.2 1.3 1.2 1.1

Total Investment/Sales Net Debt/EBITDADividend Yield % Other Shareholder Returns Total Return to S/H Berenberg EV/OpFCF

Normalized Unlevered FCF Yield Standard Unlevered FCF Yield

Berenberg Prop. EV/EBITDA EV/EBITDA Berenberg Adj. PE Standard PE EBITDA Margin (%) Capex to Sales (%)

Berenberg Real FCFE Yield Normalized FCFE Yield Standard FCFE Yield Berenberg Real Unlevered FCF Yield

2.69 3.00

22.35 26.00

159.40 150.00

16.76 18.00

16.30 16.00

12.11 14.00

223.10 160.00

5.85 7.00

0.40 0.40

15.17 12.85

Credit Metrics Organic Rev Growth Organic EBITDA Growth

Price Price Rating3 Yr fcast CAGR % 3 Yr fcast CAGR %

17-Jun-13 Target


35

Consolidation – hype or reality?

60-second summary

We think bullish expectations for in-market consolidation to rescue the sector are overdone. While we do not rule out consolidation in the sector, we do believe that the significant risk of remedies will deter operators from pursuing deals, or will significantly reduce, or even negate, the actual synergies available from in-market consolidation. Investors should not therefore rely on vain hopes of market repair via in-market consolidation to underpin a recovery in the sector’s fortunes.

Across the eight markets that we examine, one or two (ie France, Germany) do stand out as presenting the possibility for in-market consolidation with the potential for relatively light remedies, but these are special cases in our view, and any move towards consolidation in those markets should not be extrapolated across the sector. The UK would also appear well positioned to attract relatively light consolidation remedies were it not for the regulator’s clearly stated preference for four mobile networks.

In Italy, we think the combination of TI with Hutchison’s 3 will either not proceed, or will attract significant remedies that aim to offset the synergies of such a combination. In Spain, operators appear to have tried and failed to pursue in-market consolidation in a market that seems likely to attract relatively severe remedies. In Sweden and Denmark, we see material hurdles to consolidation arising from existing network-sharing JVs. The Netherlands is also likely to attract material remedies given the current state of market concentration and high pricing.

Our thinking is based on current regulatory attitudes to consolidation, which we believe is the best approach given the strong signalling component of the EC competition commission’s (CC) involvement and remedies proposed in the Austrian market with the combination of Hutchison’s 3 with Orange Austria. Hopes that regulatory attitudes towards in-market consolidation will shift to become more lenient as part of a European grand bargain to rescue the sector from crisis are seriously misplaced, in our view.

Why have we written this report?

It is simple – earlier this year, there was a wave of optimism among investors that European regulators would adopt a more positive or lenient attitude to in-market consolidation that would lead to market repair, a return of pricing power and change the outlook for the European mobile sector. Our view has been, and continues to be, that this view is misplaced. However, we do not, nor have we ever, ruled out the potential for in-market consolidation. Our view instead argues that while consolidation could potentially occur, the regulatory remedies would significantly offset the benefits of such moves. While this is our base case, we are also curious about the relative position of each market regarding the severity of remedies that could be imposed, and the potential for operators and regulators to misjudge matters, agreeing remedies that still leave scope for a degree of market repair. This report therefore tries to rank the eight main European markets with four or more mobile network operators according to their potential for consolidation to occur and to attract relatively light remedies.

How have we approached the subject of consolidation?

We find it difficult not to look at the prospect for in-market consolidation crudely


36

as a game between market participants and regulators. Market participants want to see market repair and synergies, and to regain control of pricing by using consolidation to remove a competitive force and excess capacity from the market.

Regulators want to protect consumers from price rises and ensure the continuation of competition and choice for the consumer. Regulators should not mind consolidation that results in the benefits of improved efficiency being passed on to consumers through lower prices or better services, or both. But regulators should mind where consolidation serves to protect or enshrine abnormal profits at the expense of consumers.

In the European mobile telecoms market, we have seen several instances of in-market consolidation, and so have a good impression of regulators’ likely attitude to further consolidation moves. The UK, Italy, the Netherlands and, most recently, Austria have all seen a reduction in the number of mobile networks competing in the market. The UK has seen five networks reduce to four with the merger of Orange and T-Mobile to form EE. Prior to TEL2B acquiring 4G spectrum, the Netherlands had seen the number of mobile networks reduce from five to four (via KPN’s acquisition of Telfort) and then from four to three (via the combination of Orange and T-Mobile). Italy saw the break-up of Blu with assets and customers allocated among the other four networks.

Regulators have typically resorted to remedies primarily focused on three areas.

Spectrum divestment: The divestment of spectrum creating scope for either rebalancing spectrum portfolios among the remaining players (as in the UK EE merger case and the KPN/Telfort case), or to facilitate the entry of a new entrant (as in the case of Austria).

MVNO remedies: Remedies to strengthen the MVNO segment (as was the case in Austria where MVNOs were almost non-existent as a competitive force).

Network access remedies: Remedies to ensure or enforce network access and/or roaming to benefit weaker or smaller players (as was the case in both the UK EE merger, and the Orange/Hutchison merger in Austria).

In this report, we therefore sought to identify which markets were more or less exposed to remedies of these kinds in the event of consolidation being proposed.

We have supplemented our analysis with an assessment of which markets were structurally more likely to benefit from consolidation by looking at relative pricing levels (assuming that lower priced markets might have more to gain from consolidation), recent market revenue growth trends (assuming that faster declining markets might have more to gain from consolidation), and the extent of post-paid tariff adoption (assuming that markets with more pre-paid customers could benefit from consolidation as a catalyst to drive the mix-shift towards post-paid tariffs).

In short, we have tended to assume that those markets that are currently declining the fastest, with the lowest prices and biggest bundles, the best developed MVNO and distribution segments, the most evenly balanced spectrum distribution and material network-sharing arrangements, and the most underdeveloped post-paid segments are the markets best positioned to mitigate regulatory remedies.

But is consolidation likely in the first place?

As mentioned earlier, consolidation has already happened in Europe on several


37

occasions, including in the UK, the Netherlands, Italy and Austria. Austria was an important development, coming as it did in the wake of VOD’s tentative approach to consolidation in Greece, which never materialised. Austria was also important given the rising volume of commentary from the industry and from the investment world in support of in-market consolidation. Technically speaking, since the merger of Orange and Hutchison in Austria resulted in a combination controlling less than a 25% share of the market, the deal should have been approved without any significant concerns.

However, we believe the EC CC used the review process for the Austrian merger to publicly reiterate its attitudes to in-market consolidation in European mobile. When looked at in this light, the Austrian consolidation example takes on a more important meaning as a benchmark for how the EC will likely behave with respect to future mobile in-market consolidation. As we discuss later in much more detail, remedies in the Austrian market included the creation of an MVNO sector, enforced access to the merged entities network infrastructure, and the divestment of spectrum for a new competitor – quite tough remedies for a combination that fell below the 25% combined market share threshold.

Moves in Spain by TLSN to sell Yoigo, possibly to one of the other operators, have so far come to nothing, suggesting operators are wary of overpaying in the face of uncertainty over the severity of remedies.

In Germany, the much-anticipated combination of KPN’s E+ with TEF’s O2 has yet to materialise, with talk of a merger rapidly turning to talk of a comprehensive network- and spectrum-sharing arrangement, rapidly turning to a concern that no such cooperation of any kind will actually materialise – which is a shame, as the German market seems to us to be one market where remedies could turn out to be relatively light, or at least where the scope for remedies to result in incremental competitive pressure is relatively low.

It could all hang on Italy!

Italy then could be the great test case for European in-market mobile consolidation! TI is currently deciding whether to pursue talks with Hutchison with regard to a combination of TIM Italia and 3 Italia. Being the combination of the market’s number one and number four operators, the proposal would seem likely to encounter material challenges and hurdles, particularly in terms of regulatory remedies, although the fact that the deal would likely be overseen by the AGCM, the local antitrust authority, rather than by the EC, may raise hopes of a more lenient attitude towards remedies. At the very least, significant spectrum divestments seem likely to be demanded in the event that the parties decide to proceed.

We expect the EC to remain vigilant

The consolidation bulls cite several key arguments to support their thesis: they say the EC is worried about the health of the sector, the EC is concerned that global technology companies have gained the upper hand over European telecoms operators, that the EC has signalled to industry CEOs a more positive attitude to consolidation, or that Neelie Kroes will push through a more accommodating attitude to in-market consolidation, or that the move to a pan-European telecoms market will facilitate more in-market consolidation.

We have had various discussions both with the EC, and with others involved in antitrust issues relating to European telecoms, and we have carefully scrutinised the


38

speeches of both the EC competition commissioner and the EC’s telecoms commissioner, and here is our take on each argument of the bull case in turn.

1) On the EC being concerned about the health of the sector: Specifically, the bull case argues that the EC is worried about the health of the European telecoms sector and sees in-market consolidation as a way of supporting the sector in return for more investment. We think this is not the case. While it is true that the CC would likely take into account an extreme situation where an industry was in real difficulties, our discussion with the EC suggests that the CC does not, in the vast majority of cases, take into account the direction of profits or revenues. Instead, the CC bases its view on market concentration, and the impact on the consumer, specifically asking whether the consolidation proposed is likely to lead to higher prices and be a bad outcome for the consumer. Our contacts at the CC actually stated that they saw investment in the European telecoms industry increasing at present anyway, so the evidence did not support the notion that the industry was not investing due to too much competitive pressure.

2) On the EC concern regarding global tech dominance: Specifically, the bulls argue that the EC is concerned about the balance of power between the European operators and global technology players like Apple and Google. We believe this remains the case, as is evident in much of the comment emanating from the EC. However, the CC has made the point very clearly that this does not affect judgements when it comes to competition policy and the attitude of the commission to the European telecoms sector.

3) On the CC meeting with sector CEOs: Specifically, the bulls argue that the EC CC, in a meeting with sector CEOs, agreed on a path forward which included a more supportive attitude towards in-market consolidation. We think this meeting has been misinterpreted, and that Mr Almunia’s message may have been misunderstood following his meeting with sector CEOs. The message intended was that so long as national markets existed, competition policy would be determined on a country-by-country basis (with consequent limitations on the potential for in-market consolidation). Mr Almunia proposed, however, that consolidation could be facilitated by the creation of a pan-European telecoms market.

This is where we think the misunderstanding has arisen. Bulls have taken these comments by Mr Almunia as a sign that the commission is softening its stance on consolidation. Equally, telecoms CEOs may have misunderstood the meaning of a pan-European telecoms market to be restricted to network-sharing and access. According to the commission, a pan-European telecoms market needs to go far beyond just network-sharing and access to an elimination of roaming charges and facilitation of consumers to buy a pan-European tariff from a single operator, for instance.

4) On Neelie Kroes pushing through support for consolidation: Specifically, the bulls argue that Neelie Kroes is discussing with her EC counterparts the case for in-market consolidation and is seeking to propose a case to allow in-market consolidation. We do not think this is the case. The main issues of debate within the EC relating to the telecoms sector at present are the proposals from Neelie Kroes made in July 2012 relating to future regulation of next-generation fibre networks, and forthcoming proposals by Kroes to accelerate the development of a truly pan-European telecoms market. With respect to the former, we expect to hear formal proposals by October 2013. With respect to the latter, we expect to hear formal proposals between July and


39

August with a view to the European parliament voting on the proposals in October 2013 and implementation by Easter 2014. Is there a split between Almunia and Kroes when it comes to in-market consolidation in telecoms? Not according to the CC, which states that investors should not underestimate Kroes’s attachment to the principles of competition, an attachment that leaves her more closely aligned with Almunia than perhaps some observers would recognise.

5) On the move to a pan-European telecoms market: Specifically, the bulls argue that the move to a pan-European single market for telecoms services will lead to a change in the definition of markets used in assessing consolidation, allowing more in-market consolidation to occur. Long-term, this may happen, but we do not believe this is a near- or medium-term prospect. Even in the event of a formal proposal to move to a single market for telecoms services across Europe, from our discussion with EC contacts, it is clear that even this would not automatically lead to the kind of in-market consolidation that investors would want to see. Instead, we have the impression that a single market is more likely to be price deflationary, resulting in the end of European roaming rates, and offering the potential, for example, for TEF consumers in Spain to utilise their domestic tariff in TEF’s other European operations (ie, the UK, Germany, the Czech Republic). More radical is the idea that, say, UK users could go to Italy to buy their mobile service from 3, for use in the UK and elsewhere in Europe with no difference in tariffs. The CC does not seem perturbed by the inherent cross-subsidies from high-cost markets to low-cost markets that this model would likely involve.

And if that is not enough to convince, hear it in the words of the EC’s competition commissioner, taken from recent speeches or publications this year.

Mr Almunia’s words are discouraging for consolidation bulls

We see no reason for optimism when we read over the text of Mr Almunia’s latest speeches. On 28 February 2013, the commissioner gave a very informative speech entitled “Relying on the single market for the future of Europe”, a general speech which contained his response to calls for the European mobile industry to be allowed to consolidate. The commissioner was evidently well aware of the industry’s desire to consolidate, and to create regional champions.

“I still hear from time to time that our merger control is too strict and prevents the creation of so-called ‘European champions’. I also hear that our market definitions are too narrow and fail to take account of global competition. I think these criticisms are simply unfounded. And when markets are national – as in the mobile-telephony sector to which I will refer later – we define them as such. We ought to make sure that companies do not carry out their plans at the expense of European consumers, of their business partners, and of the competition conditions in the EU.”

Mr Almunia is clearly proud of the EC’s track record in dealing with issues of “foreclosure by dominant firms”, and “tearing down” barriers to entry that make it impossible for new competitors to access a market. Here is how those thoughts relate to the European telecoms market:

“We often have to deal with these issues in… our work on mobile telephony markets. I have to say ‘markets’ – using a plural noun – because there are 27 of them in the EU; each member state has its own regulatory


40

environment and each awards its own spectrum licences. To make matters worse, in many countries there are only few network operators and the barriers to entry are high. This is because the spectrum is limited and building a network expensive. This is the reality of mobile telephony markets in Europe today. And this is the context in which some operators keep asking us to favour consolidation in the industry.”

Consolidation bulls should take special note of the commissioner’s use of “only” when discussing how few actual network operators there are in each market, and also, that reference to high barriers to entry? He goes on:

“I fully share the call for a single market for telecommunications made by Neelie Kroes. The industry would do well to consolidate across national borders, if that meant lower prices and new and better services. But let us not forget that around 80% of mobile phone subscriptions are with four leading European groups. This market structure has not brought us closer to a single market for cell phones; one where we can actually buy and use mobile services freely across the EU. Instead, roaming fees remain high and prices vary a great deal. Research shows that users pay up to 10x more to use their smartphones in those member states where there is no challenger to the big European operators.”

Okay, so the commissioner mentions consolidation, but it is the wrong kind, and would be expected to lead to “lower prices”. Lower prices, newer services and more investment are the results the EC competition commissioner would like to see arising from any consolidation across national borders. Finally, the commissioner is very clearly alert to the potential for operators to “game” the EC:

“Moreover, as long as markets remain fragmented along national borders, there is no evidence that operators will invest more if they scale up. So, the risk here is that some incumbents could play an asymmetric game in which they can move in the single market without restrictions whereas their users and business customers cannot escape from their national borders.”

Mr Almunia’s Austrian comments further undermine the bull case

After much deliberation, on 12 December 2012, the acquisition of Orange Austria by Hutchison was allowed. In granting the approval, the competition commissioner again bemoaned the “national” and “highly concentrated” nature of Europe’s mobile phone markets. The approval press release also explicitly responded to operators’ demands to be allowed to consolidate within markets in Europe. The commissioner responded thus:

“What we cannot do in any event is to give some companies a blank cheque to consolidate within their national borders and increase prices for consumers on the basis of mere promises of further investment.”

The Austrian approval provides valuable insight into the mindset of the EC when it comes to in-market consolidation, given it reflects a market that previously consisted of four network operators, consolidating down to three. The EC applied material remedies to the Hutchison/Orange acquisition, including an obligation on Hutchison to make spectrum available for a new network operator to enter the Austrian mobile market, to make wholesale access to its network available to up to 16 virtual operators for the next 10 years, and to not complete the acquisition of Orange until it had entered into a wholesale MVNO agreement approved by the commission. The commissioner noted that MVNOs were practically non-existent in Austria.


41

Even Neelie Kroes is toning down consolidation expectations

Neelie Kroes is often cited by the consolidation bulls as supporting in-market consolidation for European mobile operators. What has been clearly stated by Ms Kroes is her recognition that the European telecoms industry is in “stormy weather” at a moment when Europe “badly needs a healthy telecoms sector”.

While there has been little sign of explicit support for in-market consolidation, to be fair to the bulls Ms Kroes did pour oil on the fire of consolidation expectations when in February at the Mobile World congress she was asked by a Wall Street Journal reporter which type of consolidation would come first – in-market or at the Europe-wide level? Ms Kroes replied:

“I am very much in favour of not limiting whatever category, but I think we should indeed make the step forward and say, if it is a single market, if we have single market rules and transparency, then it should be absolutely possible that it is not only ring-fenced within a member state, for we do have so many telecoms operators, more than 250 of them, or even more than that, and when you compare that with other parts of the world it is not a guarantee that it is better for the consumer.”

It is not exactly a ringing endorsement of in-market mobile operator consolidation, and it is not entirely clear that Ms Kroes is explicitly referring to European mobile markets. For instance, her reference to more than 250 operators compares with the EC competition commissioner’s view that 80% of European mobile consumers are already served by one of four European mobile operators (which we assume refers to VOD, TEF, DTE and FTE).

In a more recent interview on the subject with the Wall Street Journal, Ms Kroes was attributed as suggesting that telecoms companies need to invest in building networks as well as just acquiring each other to create economies of scale and prevent the European telecoms market from falling behind. Ms Kroes went on to say:

“It is not about consolidation, it is about taking away fragmentation. It is a fragmented market, there are 27 markets with own regulation, own legislation, and so on.”

And in her most recent interview in the Financial Times, she went as far as ruling out consolidation within a national market, saying:

“I am a strong believer that competition keeps you awake.”

It is increasingly clear that Ms Kroes’s objective now is to accelerate the move towards a single market for telecoms services in Europe – a move that seems to us more likely to be price deflationary than to lead to market repair through in-market consolidation. As for her comments on consolidation, we find them a slim foundation on which to build a bull case for the sector given that Ms Kroes has no direct influence on how such consolidation would be treated by the EC – that responsibility sits clearly with Mr Almunia at the EC CC.

So, where is the attitude change?

On in-market consolidation, we fail to see it. Commissioner Kroes for sure would like to see faster development of a single market for telecoms. But this does not imply in-market consolidation (note Mr Almunia’s reference to the fact that 80% of European mobile users are already served by one of four European operators), and seems to us to be potentially price deflationary given the commission’s historical


42

desire to see prices aligned with costs. Commissioner Almunia likewise acknowledges the lack of a true single European market for telecoms services, particularly in mobile. But his actions in Austrian mobile (and in Greece in 2011!), and his words in recent speeches, suggest to us that the market’s expectations of a more progressive attitude to in-market consolidation will turn out to be a case of hope over reality.

Bull thesis on consolidation firmly refuted

In reality, there is nothing preventing operators proposing in-market consolidation deals today. The rules are clear, the actions of the CC are there for all to see, and in fact most proposed deals are actually eventually approved. However, they are approved with remedies.

In the distant future, Europe may eventually move towards a single market, but we think this will be evolution not revolution, and will likely involve further price deflationary pressure on the sector as roaming revenues are given the termination rate treatment. Change to market definition may well arise, but as the CC points out – it believes that 80% of mobile consumers are already served by one of the big four providers anyway, implying, we think, that the scope for material benefits from consolidation are limited, especially when significant remedies are the likely outcome. Interestingly, when we calculate EU mobile customer share, of the large four European operators, it comes in at 59%, significantly below the EC estimates.

National versus EC jurisdiction should not matter

As we explain in detail later, most of the obvious mobile market combinations across our eight focus markets would fall within the EC CC’s jurisdiction should they actually materialise as firm merger proposals. In theory, it should not make any material difference whether a merger proposal is handled at the local national level or at the EC level – European rules govern how local anti-trust authorities should treat any such proposals. In reality, it remains possible that a local authority could adopt a more lenient attitude to a merger proposal than the EC might do.

In Italy, for example, many observers have pointed to the fact that jurisdiction for any firm merger proposal between TI’s TIM Italy and Hutchison’s 3 Italia would fall within the remit of Italy’s anti-trust authority, the AGCM, and that this will result in any merger proposal standing a greater chance of approval with less severe remedies than would be imposed were the EC to have jurisdiction. Ultimately, we will only know the reality in the event of an actual merger proposal being put forward.


43

Understanding the EC remit for evaluating deals

Understanding who evaluates which deals

Whether consolidation falls under the EC’s remit, is about whether a deal forms a “concentration” that comes with a “community dimension”.

What is a concentration (ECMR Article 3)? This is where a change of control results in an undertaking that will “perform on a lasting basis all the functions of an autonomous economic entity”. A concentration arises, when a change of control results from:

i) a merger of two or more independent undertakings;

ii) the acquisition of one or more persons already controlling at least one undertaking, or by one or more undertakings, whether by purchase of securities or assets, by contract or by any other means, of direct or indirect control of the whole or parts of one or more other undertakings;

iii) the creation of a JV performing on a lasting basis all the functions of an autonomous economic entity.

Thus the Austrian deal, being an acquisition, would have satisfied as a concentration under ii), while the UK JV between T-Mobile and Orange (Everything Everywhere) was a concentration under iii).

What is a community dimension (delineated in ECMR Article 1)? For the purposes of looking at a community dimension, the EC considers various revenue thresholds of the combination. Typically, the combination is assessed on the combined revenues of the new entity. This means that:

in the case of an acquisition, the EC will look at the acquiring groups revenues, and the revenues of the entity being acquired (so if it is a subsidiary, it will be that subsidiary which is evaluated);

in the case of a JV or merger, it will be the group revenues of both entities that form the basis of the evaluation.

Concentrations are of a community dimension either where the merging parties’ (the “undertakings concerned”):

i) combined worldwide turnover is more than €5bn and each of at least two of the merging parties realised more than €250m turnover in the EU; or

ii) Combined worldwide turnover is less than €2.5bn; their combined turnover is more than €100m in each of at least three member states; in each of those three member states the turnover of at least two of the merging parties is more than €25m; the community-wide turnover of at least two of the merging parties is more than €100m;

unless each of the merging parties obtains more than two-thirds of its EU turnover in one member state. In Exhibits 2.1 and 2.2, we provide a picture to guide investors through the process.


44

Exhibit 2.1: Understanding when the EC evaluates a deal

Source: Berenberg

COMBINED

WORLDWIDE TURNOVER

(EUR) EU TURNOVER (EUR)

EU TURNOVER IN

ONE MEMBER STATE

1 >5bn >250m not >2/3 YES

2 >5bn not >250m not >2/3 GO TO EVALUATION 2

3 <5bn >250m not >2/3 GO TO EVALUATION 2

4 <5bn not >250m not >2/3 GO TO EVALUATION 2

5 >5bn >250m >2/3 NO

6 >5bn not >250m >2/3 NO

7 <5bn >250m >2/3 NO

8 <5bn not >250m >2/3 NO

WORLDWIDE TURNOVER

(EUR)

TURNOVER IN EACH

OF AT LEAST 3

MEMBER STATES (EUR)

TURNOVER IN EACH

THOSE 3 MEMBER

STATES (EUR)

COMMUNITY WIDE

TURNOVER (EUR)

EU TURNOVER

IN ONE

MEMBER STATE

1 >2.5bn >100m >25m >100m <2/3 YES

2 <2.5bn

3 >2.5bn not >100m

4 >2.5bn >100m not >25m

5 >2.5bn >100m >25m not >100m <2/3 NO

6 >2.5bn >100m >25m >100m >2/3 NO

EVALUATION 1

EVALUATION 2

COMBINED FOR EACH MERGING PARTY (AT LEAST TWO)

COMMUNITY

DIMENSION

EVERY COMBINATION IS NOT A COMMUNITY DIMENSION



FOR EACH MERGING PARTY

COMMUNITY

DIMENSION


45

Exhibit 2.2: Flow chart to help understand when the EC evaluates a deal

Source: Clifford Chance

A history of deals, and why they fell under the EC’s remit

We go through a list of deals evaluated by the EC in the last eight years, discussing the above criteria, and the reasons why they were considered a concentration with a community dimension.

T-Mobile and Tele.ring (Austria 2006)

The transaction involved T-Mobile acquiring the shares in EHG Eubkaufs-und Handels GmbH, which was the sole owner of the Tele.ring group. As such, the deal constitutes a concentration under Article 3(1)(b).

Because the combined entity (DTE and Tele.ring) had:

• an aggregate worldwide turnover of more than €5bn

and because each of DTE and Tele.ring each had:

• over €250m of EC turnover and,

• neither entity generated over two-thirds of its EC revenues in one member state,

the deal was considered to have a community dimension.

Combined worldwide turnover over €5bn

EU-wide turnover of each of at least two undertakings

over €250m

Combined worldwide turnover over €2.5bn

EU-wide turnover of each of at least two undertakings

over €100m

Combined turnover in each of at least three Member States

over €100m

In each of at least three of these Member States turnover

of each of at least two undertakings over €25m

2/3 rule met

EUMR applies EUMR does not apply

No

No

No

No

No

No

No Yes

Yes

Yes

Yes

Yes

Yes

Yes


46

T-Mobile and Orange (Netherlands 2007)

The acquisition formed a concentration with T-Mobile buying the shares of ONL and ONB (with ONL being the MNO and ONB being the broadband unit).

It counts as having a community dimension because:

• the combined worldwide turnover was more than €5bn (DTE’s was €61.3bn and ONL/ONB had €715m);

• each had a European turnover of more than €250m (DTE’s was €45.9bn while ONL/ONB had €715m of revenues);

• while ONL/ONB had over two-thirds of revenues in one EC market, DTE did not.

Everything Everywhere (UK 2010)

The JV is a full-function JV venture performing on a lasting basis all the functions of an autonomous economic entity.

The JV was deemed to have a community dimension because:

• the combined worldwide turnover of FTE (€53.5bn) and DTE (€61.7bn) was over €5bn;

• each of FTE and DTE had European turnover of more than €250m;

• neither had over two-thirds of their EC revenues coming from one member state.

Hutchison Orange (Austria 2012)

The EC has not yet published the detailed document on the deal. However, because Hutchison had over €5bn of worldwide revenues, and because each of Hutchison and Orange Austria had over €250m of EC-wide revenues, this would satisfy the community dimension criteria, and although Orange Austria had over two-thirds of its revenues from Austria, Hutchison would not be able to satisfy the more-than-two-thirds-from-one-member-state criteria.

In addition to these deals, there are also deals that could avoid EC scrutiny and instead fall under the national regulator’s mandate. A likely example of this would be the proposed deal between TI and Hutchison.

TI and Hutchison (Italy TBC)

The combination between TI and 3 Italia would have over €5bn of revenues, and each have over €250m of European revenues. However, because TI has more than two-thirds of its EU turnover in one member state, and because 3 Italia does too, the deal would fall under national jurisdiction rather than EC jurisdiction.

Which acquirers would face EC Directorates-General (DG) competition scrutiny?

Given that we know the process to determine whether a deal comes with a community dimension, we can estimate which of them may fall under an EC remit and which may fall under the remit of the national regulator.


47

• In Exhibit 2.3, we show that only ILD, TDC and TEL2B have worldwide revenues of less than €5bn.

• In addition, TDC has more than two-thirds of its European revenues from one member country. This is also the same for all the French operators.

• This would suggest that any consolidation in France would be unlikely to fall under the remit of the EC, and would instead fall under the French regulators concern.

• In addition, any acquisition by TDC in the Danish market would fall under the DCC, not the EC.

• Vimpelcom and TI are also interesting. Because over two-thirds of their European revenues are based in Italy, if either Wind or TI would like to consolidate the Italian market, this will fall under AGCM’s jurisdiction.

• TEL2B is the less obvious one, as it has revenues of less than €5bn but more than €2.5bn, meaning that it falls under the secondary analysis. We believe that a TEL2B acquisition of 3 in Sweden would thus fall under the KKV, not the EC (because each merging party will not have over €25m in at least three member states).

• Finally in Exhibit 2.4, we show that each fourth player in all the four-player markets has over €250m of revenues, meaning it could fall under the first point as being under EC jurisdiction (assuming the other points do too).

• So we can for example conclude that an O2 acquisition of E-plus, or vice versa, would have a community dimension and be scrutinised by the EC.

Exhibit 2.3: European four-player market participants and the relevant metrics for evaluating whether deals fall under EC remit of having a community dimension

Source: Berenberg ** European revenues estimated based on published data

Company

Worldwide

revenues 2012 (€)

European

revenues 2012 (€)

>2/3 of EC revenues

in one member

state

Bouygues 33,547 25,851 YES

Deutsche Telekom ** 58,169 37,144 NO

France Telecom ** 43,515 39,157 YES

Hutchison Whampoa ** 23,801 5,778 NO

Illiad 3,153 3,153 YES

KPN ** 12,708 12,708 NO

TDC 3,505 3,505 YES

Tele2 (ex-Russia) 3,589 2,765 NO

Telecom Italia 29,503 18,386 YES

Telefonica 63,356 29,995 NO

Telenor 13,366 3,507 NO

TeliaSonera 12,248 8,654 NO

VimpelCom 17,470 5,427 YES

Vivendi 28,994 19,348 YES

Vodafone 52,227 35,865 NO


48

Exhibit 2.4: All fourth players have over €250m of revenues, meaning they could, if all else satisfies, fall under the EC remit of having a community dimension

Source: Berenberg

What we really want to hear

Over the rest of this year, we will be hearing from Neelie Kroes on her package of reforms to accelerate the development of the single market, and on the final shape of proposals for regulating next generation networks. To the extent that there are implications for the consolidation bull thesis, we will be judging these pronouncements and policies by asking whether they will lead to:

in-market consolidation – not just any kind, but the kind that takes excess capacity out of the market and reduces the scope for disruptive players to exploit their underutilised networks to price the incremental “byte” at marginal cost;

fewer MVNO deals – network operators have got to get real and realise that offering crazy deals to MVNOs just makes the market worse, regulators have to understand that if a requirement for network operators to offer cheap access to their networks is introduced, then this risks undermining future network investment;

less spectrum, not more – it is the raw material of network capacity, and despite governments’ and regulators’ desire to see even more spectrum made available to the industry, the reality is that a spectrum shortage would be good for prices: smaller operators would soon see their empty networks filled up as their disruptive offers were exploited by high-volume users, and operators would have less capacity available to offer crazy MVNO deals: the result would be an industry-wide move to more rational pricing based on the tension between growing demand and limited supply – nirvana!

We should not get our hopes up. Notwithstanding all of the above, it remains possible that operators will pursue in-market consolidation, taking a hopefully carefully considered risk that the anti-trust authorities will misjudge the need for remedies, leaving enough scope for market repair an synergies on the table to encourage a deal to be consummated. Later in this report, we assess which market might be best positioned to consolidate with relatively light remedies, or to mitigate the impact of remedies. First, we revisit the lessons from a case study on Austrian, the most recent example of in-market consolidation in European mobile, as well as what can be learnt from the Everything Everywhere tie up in the UK.

Country Fourth operator

Local revenues

2012 (€)

Denmark Three 316

France Illiad 3,153

Germany E-Plus 3,404

Italy Three 1,965

Netherlands Tele2 606

Spain Yoigo 960

Sweden Three 740

UK Three 2,289


49

The EC merger control regime and telecoms case studies

We follow our assessment of the parameters for EC intervention with an investigation into how the European Commission (EC) has applied the EU merger control regime in mobile merger cases. In particular, we examine the remedies that have been sought and what impact these deals have had on market competition. We also identify, from a regulatory perspective, the key lessons on how competition authorities address deal activity and what their key remedies to maintain competitive balance have been.

Key lessons

1. The relevant markets: The authorities consider the impact that MVNOs have on the market in their assessment of market concentration. However, they consider the retail market a single market for mobile services to end-users and make no distinction between private and business users, or pre-paid or post-paid subscribers.

2. Lessening of competition as opposed to dominance as a criterion: Some firms have more of an influence on the competitive process than their market shares suggest. For instance, a merger of two companies selling close substitutes would eliminate a significant competitive constraint, even though the pro-forma combined market share of those entities would not be dominant. The EC is permitted, in these cases, to prohibit the deal or seek commitments from the notifying parties. The most vivid example of this is the recent Austrian deal between Orange/H3G in which the EC intervened despite a subscriber share for the combined entity which was below the 25% safe harbour limits of the Commission’s own Horizontal Merger Guidelines.

3. Remedies for safeguarding mavericks: If a transaction takes an active price-aggressive operator out of the market or has the potential to threaten it, the EC is unlikely to approve the deal without significant infrastructure concessions from the merging parties. Similarly, in markets where MVNOs exercise a competitive constraint, the EC will consider the impact of the transaction on the wholesale access market.

4. Spectrum concentration in 1.8GHz an issue: The EC views this band as important for speed to market for long-term evolution (LTE) services, as was evident in the T-Mobile/Orange JV transaction in the UK.

5. The EC seems sceptical that network cost efficiencies will be passed on to consumers. In appraising problematic concentrations under the Merger Regulation, the EC states that it will consider the “efficiencies defence”, which supports deals that see consumers benefiting from asymmetries that make the larger firms more efficient. However, in practice, the EC has tended to view merger efficiencies defence with scepticism (T-Mobile/Tele.ring, 2006).

6. Approach to future technologies is fluid: The most stunning example of this was the T-Mobile/Tele.ring deal of 2006, for which the EC did not give any credit to Hutchison’s 3G spectrum (2.1GHz band) when assessing capacity constraints, arguing that the case for its utilisation was unproven. However, a year later in 2007, one of the factors that the EC


50

used to justify that MVNO access would not be compromised following the merger of T-Mobile/Orange in The Netherlands was that the operators had significant amounts of underutilised 3G spectrum.

Remedies The EC’s prescriptions have tended to be most severe when it is remedying the loss of a maverick operator or trying to preserve the role of the maverick in the market. Since the introduction of the new EU merger control rules, we have had two transactions that have resulted in the takeovers of maverick operators (Tele.ring by T-Mobile in Austria, Orange by H3G in Austria) and one in which the maverick could have been marginalised (T-Mobile/Orange in the UK). In each case, the EC placed extensive remedies on the notifying parties to improve the position of an existing challenger or to facilitate a new market entrant to replace the price aggressor that was acquired in the transaction. The remedies were a combination of:

(i) divestitures of mobile radio access sites and spectrum; and

(ii) behavioural undertakings, ranging from regulated MVNO access commitment at preferential rates to national roaming and network-sharing obligations.

In the case of H3G/Orange Austria, the regulator went one step further and reserved 20MHz of 800MHz spectrum for a new entrant at a low price of €0.27/MHz/pop, then bolstered the case by offering the entrant lower coverage requirements than the other operators.

EC has also demanded up-front commitments to ensure its remedies are effective and appointed monitoring agents to oversee compliance with such commitments. T-Mobile had to sign a legally binding framework agreement with H3G before it obtained clearance from the EC for its acquisition of Tele.ring in 2006. Similarly, H3G had to enter an MVNO agreement with a party approved by the EC before it received merger clearance for its acquisition of Orange in 2012.

Of the four case studies into how the EC has reacted to in-market consolidation proposals, three had spectrum remedies applied, with the Netherlands being the odd one out, where the deal between T-Mobile and Orange was deemed not to need remedies, due in part to a strong MVNO presence. However while spectrum remedies have in retrospect addressed the EC’s competitive concerns and helped a maverick to continue being price aggressive, the lack of this remedy in the Netherlands has seen prices remain high, with the MVNO structure alone not acting as a sufficient competitive constraint. This was likely to be the driver behind the Dutch authorities introducing a new entrant into the Dutch mobile market, and in turn taking the spectrum HHI back below 3,000 (as was the case with the EC remedy in each of the other EC evaluated deals). It is in this context that we see the EC’s actions with the Austrian deal between Hutchison and Orange as setting a new precedent on how the EC intends to treat M&A. With the Dutch example failing to protect consumers, the requirement to divest spectrum to an existing maverick (3 in the UK) or a new maverick (the new entrant in Austria) suggests that the EC sees its original remedies in the T-Mobile/Tele.ring Austrian deal back in 2006 as a better way to go than the remedy-free approach it took in the Netherlands in 2007. While clearly the Dutch example differed from the UK and Austria, due to its developed MVNO segment, we doubt that this alone, without addressing spectrum holdings, will be the approach that the EC will take in future. It is in this context that we evaluate European M&A opportunities in this note.


51

A brief overview of the EU merger control regime

Before 2004, merger control tests insisted on “dominance” being either necessary or sufficient to prohibit a merger. Dominance was understood as: “single dominance”, where a merger led to a single firm having a market share of 40% or more and, as a result of the deal, was able to raise prices, reduce choice or cut levels of innovation without losing customers; and “collective dominance”, where the merger facilitates tacit collusion between a small group of players. However, by making dominance a defining requirement, this led to under-enforcement in cases when, for instance, a merger of two companies selling close substitutes would eliminate a significant competitive constraint, even though the pro-forma combined market share of those entities was not dominant.

In 2004, therefore, the EU introduced a new merger control regime that broadened the scope of the EC to intervene in situations in which a merger/JV did not result in single/collective dominance but nevertheless raised serious competition concerns. An economic framework, named the Horizontal Merger Guidelines, set a safe-harbour limit of 25% market share. The broader concept here is that some firms have more of an influence on the competitive process than their market shares suggest. A merger involving such a firm could change the competitive dynamics in a significantly anti-competitive way, in particular when the market is already concentrated. The EC is allowed, in these cases, to prohibit the deal or seek commitments from the notifying parties.

In appraising a merger that is likely to be anti-competitive, the guidelines identify two main effects that may significantly harm competition.

(i) Non-coordinated effects (unilateral effects): The merger eliminates competitive constraints on one or more firms, which consequently enjoy greater market power. A key concern in unilateral-effect cases is that the notifying firms could raise their prices post-merger. These effects are strongest when the firms are close competitors in a market segment that has limited alternatives.

(ii) Coordinated effects: The pre-/post-merger market structure is examined. If it is oligopolistic (eg limited to three or four players), the EC will consider whether the merger will facilitate tacit collusion resulting in higher prices, reduced output or other harmful effects.

In appraising problematic concentrations, the EC will also consider the following factors.

• The efficiencies defence: If the notifying parties can put forward substantiated and verifiable evidence that the deal will result in cost savings or other efficiencies, the EC will consider if the merged entity is better placed to act pro-competitively for the benefit of consumers (thereby counteracting the adverse effects on competition that the merger might otherwise have).

• The failing company defence: In exceptional circumstances, the EC may conclude that an otherwise problematic merger is nevertheless compatible if one of the merging parties will “fail” in the near future because of financial difficulties, leading to an exit from the market and its market share accruing to the notifying part.


52

Case study: T-Mobile/Tele.ring, Austria (2006)

What happened? On 26 April 2006, the European Commission approved the merger of T-Mobile and Tele.ring. The case was a five-to-four-player merger between the fourth-smallest market player (Tele.ring) and second-largest (T-Mobile). Although the merger increased T-Mobile’s market share to slightly north of one-third of the market, the company remained smaller than Mobilkom – the market leader owned by TKA. This would have been a difficult case to handle under the old EC Merger Regulation as the deal did not result in a dominant market position for T-Mobile. Therefore, its handling and result was significant for the application of the new merger control rules.

Exhibit 3.1: Subs share pre-deal Exhibit 3.2: Subs share post-deal

Source: Berenberg Source: Berenberg

What was the EC CC’s approach? For the examination of the competitive assessment, the EC focused on the analysis of “unilateral” effects. Owing to the issues listed below, the EC expressed serious concerns and opened an in-depth investigation.

• The transaction would remove a maverick from the market: The EC considered after a review of market share, pricing, past behaviour and number portability data, that Tele.ring was a substitute for Mobilkom (number one) and T-Mobile (number two) in the market. Hence, it exercised considerable competitive pressure and played a crucial role in restraining Mobilkom and T-Mobile pricing behaviour. A merger involving Tele.ring would, therefore, remove a maverick from the market, which could significantly impede competition. This was particularly problematic given:

o the lack of competitive constraint from MVNOs and service providers, which were estimated to account for less than 5% of the market (of which the Orange-owned Yesss! brand was the most significant); and

o the inability of other network operators to replace Tele.ring as a challenger in the market.

• Other network operators could not replace Tele.ring: One (number three) was considered unlikely to take over Tele.ring role as it had not acted as a price aggressor before. Even though Orange had Yesss!, its discount brand, which was proving to be reasonably successful, the EC regarded its focus on the pre-paid market as problematic in terms of offering a substitute to a mainly post-paid discount brand, such as Tele.ring. In the eyes of the EC, the problem for Hutchison (H3G, number five) was that it relied on a roaming agreement with Mobilkom for 2G and its 3G coverage was only 50% of pops.

A1, 39%

T-Mobile, 24%

tele.ring, 12%

One , 21%

H3G, 3%

A1, 39%

T-Mobile, 36%

One , 21%

H3G, 3%


53

• The transaction would constrain network capacity: The EC only considered the GSM frequencies for its analysis as the potential for UMTS frequencies was unproven. H3G did not own any GSM frequencies and relied on a national wholesale roaming agreement with Mobilkom. The concern was that after the transaction, there would only be three operators with GSM frequencies and with the elimination of the Tele.ring network, One would be the only operator with significant spare capacity. It was concluded that the significant reduction in spare capacity would disincentivise the network operators from attracting new customers by offering lowering prices.

• T-Mobile’s cited merger efficiencies would not benefit consumers: There is a concept prevalent in the Merger Regulation that efficiencies from a merger are passed on to consumers and that higher concentration should have no significant effect on prices. This was the case put forward by T-Mobile when it substantiated the efficiency gains it expected from the deal, primarily from better capacity utilisation and lower network maintenance costs. The EC argued, however, that savings from fixed costs are less likely to be passed on to consumers and that T-Mobile was disincentivised to pass these savings on to consumers due to the backbook-repricing impact it would have on the existing subscriber base.

What were the key remedies required by the EC? The case was cleared after T-Mobile offered specific remedies and the EC was convinced that these remedies would create a player that would likely fill a similar role to that vacated by Tele.ring.

The EC considered H3G as the most likely to take over the role of Tele.ring in the market, given its past pricing behaviour, similar price positioning and smaller customer base, provided that it had a better network and one that was not dependent on a 2G roaming agreement with Mobilkom. The rationale was that this would make H3G independent and lower its variable cost base, which in turn would have a direct impact on H3G’s pricing. Furthermore, as H3G would not be able to achieve any significant economies of scale from its new network unless it won over new customers, it would be incentivised to offer cheap pricing.

T-Mobile had to divest Tele.ring’s two packages of UMTS frequencies and a significant number of mobile telephony sites (including the necessary technical equipment). The EC wanted certainty that Hutchison would purchase the spectrum frequencies and sites and, therefore, T-Mobile entered a legally binding framework agreement with H3G before the EC granted clearance.


54

Exhibit 3.3: Adjusted spectrum share post-deal, pre-remedies

Exhibit 3.4: Adjusted spectrum share post-deal, post-remedies

Source: Berenberg, adjusted so that 900MHz band is 2-2.5x more valuable than 1.8-2.1GHz


What was the result in the market?

• Pricing? Competition has been intense since Tele.ring was bought by T-Mobile. It did not start well when Orange, being prepared for sale at around the same time as the T-Mobile/Tele.ring deal, had launched an attack on the T-Mobile acquired subscriber base in April 2005 by launching discount no-frills brand Yesss!. TKA, not wanting to cannibalise its subscriber base, responded with its own discount no-frills brand, Bob, in 2006.

• In fact, with the exception of 18 months of relative stability over the period between end-2007 and mid-2009, there has been a series of price wars. The main aggressor over 2007-2010 was One (Orange) in mobile voice and smartphone tariffs, with H3G focusing its attention on cheap mobile broadband services. In 2011, Hutchison became aggressive in the smartphone category having vowed in 2010 to pass back the reduced costs from a sale and lease back transaction of its 3G network with ZTE to customers.

• The incumbent, TKA, attempted to reverse this trend in April 2011 when it introduced annual service charges and activation fees in the hope that competitors would follow. Orange/Hutchison, while implementing similar charges, have also been cutting prices.

• Competitive tensions? Tensions are high with a marked difference in the pricing gap between the high- and low-end tariffs (please see exhibits 5.2 to 5.4). Moreover, handset subsidies have risen significantly in 2013 as Mobilkom (TKA’s high-value brand) has gone toe-to-toe with Hutchison to prevent a further loss of high-value customers.

• Market share shifts? Since the Tele.ring deal., T-Mobile has lost around 4% of subscriber market share to H3G. H3G meanwhile has taken 8% of market share from T-Mobile and Orange. TKA has been able to maintain its market share due to the success of its discount no-frills brand, Bob.

A1, 33.4%

T-Mobile, 24.9%

tele.ring, 13.5%

One , 23.8%

H3G, 4.3%

A1, 33.4%

T-Mobile, 34.2%

One , 23.8%

H3G, 8.6%


55

Exhibit 3.5: Remedies appear to have worked in strengthening the role of the maverick


Exhibit 3.6: Subscriber share (m) evolution

Source: Berenberg

Exhibit 3.7: Service revenue share (€m) evolution

Source: Berenberg

2200

2400

2600

2800

3000

3200

3400

3600

2400 2600 2800 3000 3200 3400 3600 3800 4000

Se

rvic

e r

ev

en

ue

HH

I

Spectrum HHI

Spectrum concentration vs market concentration

Market is too concentrated withrevenues and spectrum in the hands of too few players. Regulators are likely to demand significant remediesin the event of consolidation, suggesting consolidation is unlikelyto happen.

Market is relatively uncompetitive in terms of revenues but spectrum ownership is more fragmented suggesting later entrants have so far failed to exploit spectrum portfolios. Consolidation efforts are more likelyto attract non-spectrum remedies that improve the lot of smaller players, maybe through improved MVNO terms or network access and sharing provisions.

Market is more balanced andrelatively competitive both in termsof revenues and spectrum ownership. Consolidation which shifts the market into one of the other quadrants remains likely to result in remedies, albeit these are likely to be relatively less burdensome.

Market is more competitive albeitfuture competitive intensity may bejeopardised by an imbalance in the ownership of spectrum, possibly suggesting the market needs remedies to resolve spectrum hoarding or spectrum accumulationin the event of consolidation.

Dec05: Pre tele-ring deal

Apr'06: Post deal, pre remedies

Dec'12: pre H3G/Orange

Dec'06: post remedies

2200

2400

2600

2800

3000

3200

3400

3600

2300 2800 3300 3800

Su

bs

hare

HH

I

Adj. Spectrum HHI


2007 2008 2009 2010 2011 2012

3 0.51 0.65 0.87 1.09 1.34 1.71

Orange 2.05 2.12 1.95 2.27 2.35 2.39

T-Mobile 3.27 3.40 3.45 3.78 4.06 4.10

TKA 3.96 4.50 4.83 5.11 5.27 5.38

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

Su

bsc

rib

ers

(m

)

2007 2008 2009 2010 2011 2012

3 190 169 174 207 222 242

Orange 624 587 569 548 500 477

T-Mobile 1182 1085 989 926 863 802

TKA 1399 1442 1371 1311 1240 1155

0

500

1000

1500

2000

2500

3000

3500

4000

Se

rvic

e r

ev

en

ue

s £

m


56

Case study: T-Mobile/Orange, the Netherlands (2007)

What happened? On 20 August 2007, the EC approved the merger of T-Mobile and Orange in the Netherlands without expressing any serious concerns or need for remedies. The case was a four-to-three-player merger between the fourth-largest market player (Orange Netherlands) and the third-largest (T-Mobile). The merger increased T-Mobile’s market share to slightly north of 25% of the market, just ahead of VOD (22.3%) but significantly below KPN (48.5%).



What was the EC CC’s approach? For its examination of the competitive assessment, the EC focused on the analysis of “unilateral” effects in the retail as well as wholesale markets, given the importance of MVNOs in the Dutch market. It made the following key observations.

• Orange was not a maverick: The EC examined whether Orange was the closest substitute to T-Mobile and if this exerted a strong competitive restraint on T-Mobile. Orange was found to be more focused on the pre-paid segment in terms of subscribers and revenue share. As the post-paid market represented 80-90% of overall service revenues in the Dutch retail market and more than 80-90% of T-Mobile overall retail revenues, the EC held that Orange was not a major competitive constraint. This was reflected in number portability data and a review of pricing, which showed that several other providers (such as T-Mobile/Telfort – a brand of KPN, and MVNOs such as TEL2B) were cheaper than Orange in the post-paid segment. The EC concluded, therefore, that T-Mobile was unlikely to increase tariffs for Orange customers on the basis that a large number of those customers would, in any event, switch to T-Mobile.

• Competitive constraint from MVNOs would remain significant: The EC remarked that the subscriber share of the 50 MVNOs in the Netherlands was underestimated. Two of the most significant of these, TEL2B and Debitel (since acquired by KPN), were already active in the post-paid market and accounted for 9% of Dutch mobile retail revenues, according to estimates collected by the EC. The conclusion was that competitive tension from MVNOs would ensure any tacit collusion between the three remaining largest operators in the market was impossible after the T-Mobile/Orange deal.

• Wholesale market was unaffected by capacity constraints: After an analysis of the current and future network utilisation rates of the MNOs, the EC deemed that the access conditions for MVNOs and service providers were unlikely to change as MNOs would continue to use unutilised spectrum, particularly 3G spectrum with the roll-out of UMTS, to host existing and additional MVNOs. Other factors that the EC considered were KPN’s past

KPN, 51%

Vodafone, 22%

T-Mobile, 15%

Orange, 12%

KPN, 51%

Vodafone, 22%

T-Mobile, 27%


57

behaviour, which had accommodated an MVNO strategy to recover retail revenue share, and the fact that Orange was not a significant wholesale provider in the market, with a market share of 10-20% versus KPN’s 50% plus.

Exhibit 3.10: Adjusted spectrum share pre-deal, pre-remedies




Key remedies required by the CC? None.

What was the result in the market? The EC had hoped that competitive pressures from MVNOs would offset the negative impact from an increase in market concentration post the T-Mobile/Orange deal. However, as can be seen below, market concentration has remained high and benchmarking mobile pricing in the Netherlands versus peers shows it to be at the more expensive end of Europe (see Exhibit 5.2-5.4). There are changes afoot, however, that began after TEL2B acquired 2x10MHz of 800MHz spectrum at the auction in December 2012. Since, then it has ramped competitive activity and is now taking the bulk of subscriber additions in the Dutch market.

KPN, 42.4%

Vodafone, 21.3%

T-Mobile, 18.6%

Orange, 17.7%

KPN, 43.6%

Vodafone, 19.1%

T-Mobile, 37.3%

Exhibit 3.12: The EC approach did not work in reducing market concentration


2200

2400

2600

2800

3000

3200

3400

3600

2400 2600 2800 3000 3200 3400 3600 3800 4000

Se

rvic

e r

ev

en

ue

HH

I

Spectrum HHI






Dec'06: Pre T-Mob/Or deal

Dec'07: post deal, post remedies

Dec'12

2200

2400

2600

2800

3000

3200

3400

3600

3800

4000

2200 2700 3200 3700 4200

Su

bs

share

HH

I

Adj. Spectrum HHI



58

Case study: Everything Everywhere, UK (2010)

What happened? On 1 March 2010, the EC gave the green light for Orange (the number three player in the market) and T-Mobile (the number four player) to implement a JV. The JV reduced the market from five to four players and emerged as the new number one player in the market. On the infrastructure side, the number of networks were reduced to three (the JV, VOD and O2) and after taking network-sharing agreements into account, the EC argued one could even make the case that there were only two networks remaining, with T-Mobile/Orange/3UK on the one hand, and VOD/O2 (which had also entered a network-sharing agreement) on the other.



What was the EC CC’s approach? For its examination of the competitive assessment, the EC focused on the analysis of unilateral effects in the retail as well as wholesale markets, given the importance of MVNOs in the UK market. The EC made the following observations. As result, it expressed serious concerns and opened an in-depth investigation.

• Competitive constraint from MVNOs would remain significant: The EC concluded that the JV parties were not close substitutes to each other, were not the cheapest providers in the market and had not been price disruptors in the past. Instead, it saw pricing pressure in the market to be driven by the almost 50 MVNOs and service providers in the UK market. It expected these to continue to exercise a significant competitive restraint in the retail market, even after the JV. The EC also concluded that despite the planned rationalisation of mobile sites by T-Mobile/Orange after the JV, the potential for capacity constraints to harm MVNO access deals was likely to be low. However, this assessment was contingent on safeguards for the maverick, 3UK, which had a 3G site-sharing arrangement with one of the parties of the proposed JV, T-Mobile.

• The maverick had to be protected: When the EC was notified of the JV, 3UK had a 2G national roaming agreement with Orange and a 3G mobile site-sharing agreement with T-Mobile, under which both parties agreed to merge their existing sites to create a single 3G network. The concern voiced by 3UK, echoed by OFCOM (the UK regulator) and agreed by the EC was that after the transaction, T-Mobile might have the ability to terminate the RAN-sharing agreement early, or at least compromise it, to the detriment of 3UK. As a consequence, the JV transaction could, in the short to medium term, lead to a “five- to three-” player market given that it could force 3UK to exit.

• 1.8GHz spectrum concentration could adversely affect future competition: The EC undertook an analysis of spectrum holdings and

Orange, 23%

T-mobile, 14%

Vod, 26%

O2, 29%

3, 8%

EE = Orange/T-

mobile, 36%

Vod, 26%

O2, 29%

3, 8%


59

concluded that in the absence of the JV, more than one LTE network could emerge in the UK market. However, if the JV was allowed, the JV concentration of spectrum in the 1.8GHz band would give it a significant time advantage in launching LTE, given the uncertainty around the 800MHz auction for the other operators and the time needed to clear the sub 1GHz spectrum for 4G services. In addition, the JV larger holdings would mean that it could offer superior network quality.

What were the key remedies required? T-Mobile and Orange had to divest 2x15 MHz of spectrum in the 1.8GHz band by the end of 2011. Of the divested spectrum, 2x10MHz had to be cleared by 2013 at the latest and a further 2x5 MHz by 2015. In addition, the existing network-sharing deal with 3UK was reinforced.





What was the result in the market? Please see a detailed analysis in the section entitled How has the Everything Everywhere JV performed?.

EE JV, 41.4%

Vodafone, 27.2%

O2, 25.2%

3UK, 6.2%

EE JV, 36.5%

Vodafone, 29.5%

O2, 27.3%

3UK, 6.7%


60

Case study: H3G/Orange, Austria (2012)

What happened? On 12 December 2012, the EC approved the merger of Hutchison 3G Austria (H3G) and Orange Austria. The case was a four- to three-player merger between the two smallest in-market players (H3G/Orange) which was dependent on a side-deal involving Orange Austria’s divestment of its no-frills discount brand Yesss! to TKA, which has the number one position in the market. Given the connected nature of the transactions, the Austrian Federal Competition Authority (FCA) tried to make the case that local interests would be better served if it reviewed both the deals. However, the EC overruled and undertook the review of the main deal between H3G/Orange, allowing the Austrian FCA to assess the side-deal between Yesss!/TKA.



What was the EC CC’s approach? After conducting an early-stage probe, the EC opened an in-depth investigation in July 2012 to “make sure that the reduction in market concentration in an already highly concentrated market does not lead to higher prices for end-consumers”. We do not have the full text of the EC decision but the approach was unusual given the following factors.

• Pro-forma market share was below the safe harbour limit: Unlike the deal between T-Mobile/Tele.ring in 2006, which resulted in a close symmetry in market shares between T-Mobile (with a 36% market share) and TKA (38%), this transaction resulted in a combined subscriber share of below 25% for H3G, materially below its remaining competitors’ 75% and below the safe harbour limit of the Commission’s own Horizontal Merger Guidelines.

• H3G could be considered a maverick: If Orange could be considered a maverick in the market by the commission, so could H3G. Its pricing was consistently at the cheap end, particularly since 2011. Note that H3G on various occasions during the process publicly stated that it was targeting a 30% subscriber market share by 2015 (from 22% pre-merger post-divestment of Yesss!) and that it would use the synergies from the Orange acquisition to fund an aggressive LTE build to bolster an already strong mobile network in Austria (ranked number one in Austria by Connect Trade magazine in 2011/2012).

• A failing-company defence for Orange: There were reports during the merger process that Orange Austria faced very high debt maturities in 2013 and, as a result, could be financially constrained going into the 800MHz spectrum auction.

A1, 40%

T-Mobile, 30%

One , 18%

H3G, 13%

A1, 45%

T-Mobile, 30%

H3G, 25%


61

What were the key remedies required? The concessions that H3G had to give to get the deal approved were the most extensive seen yet. We detail them below.

• A commitment to divest 2x10 MHz in the 2.6GHz frequency band (at no minimum price) to the acquirer of the 2x10 MHz of spectrum in the 800 MHz frequency band that would be reserved by the Austrian regulator for a new entrant in the upcoming spectrum auction to take place in Q3 2013. If the auction spectrum is not acquired or the acquirer does not wish to also acquire the divestment spectrum, H3G shall not be required to divest the divestment spectrum, and can continue to use it.

• H3G must also offer national roaming for six years to the new entrant on the basis of the same charges and substantially the same terms as contained in the reference offer for MNVO wholesale access. In addition, H3G must by September 2013 divest on commercial terms at least 2,000 sites that it does not require for its network to the potential new entrant and offer preferred co-location on H3G’s existing sites on standard market terms if technically feasible.

• A commitment to make wholesale MVNO access available to at least 16 MVNOs on regulated terms over a 10-year period. This included an upfront commitment to sign an MVNO agreement with an MVNO approved by the EC prior to receiving merger clearance. The EC approved the upfront MVNO arrangement with UPC (Austria’s largest cable operator) on 18 December 2012.

• Appointment of a monitoring trustee to oversee implementation of the above commitments.



Source: Berenberg, adjusted so that 800/900MHz band is 2-2.5x more valuable than 1.8-2.1GHz

Source: Berenberg. The chart assumes that A1/T-Mobile/Hutch maintain same spectrum holdings in 900/1800/2100 MHz bands post-September 2013 auction

What was the result in the market?

• Pricing? Since the deal was cleared, both TKA and T-Mobile have taken steps to increase headline prices in the market with the introduction of new tariffs in Q1 2013. However, both operators have heavy promotions on these tariffs to minimise churn to H3G which continues to maintain its aggressive price points. TKA hopes that H3G will (eventually) respond to the higher price points. H3G’s CEO meanwhile gave an interview to local press (Kurier.at, 28 March 2013) stating that the company intends to use the synergies from the Orange combination to introduce new tariffs. When asked about the direction of these new tariffs, the CEO said that the company will continue to maintain attractive prices for consumers and hopes to upsell subscribers on low price points over

A1, 33.4%

T-Mobile, 33.1%

Hutch, 33.4%A1, 35.2%

T-Mobile, 30.8%

Hutch, 25.4%

Divestment for new

entrant, 8.6%


62

time as demand for better mobile transmission (ie speed) grows. On current momentum, he expects the combined H3G/Orange entity to become the second-largest mobile operator in Austria with a 30% market share in three years’ time.

• Competitive tensions? Tensions remain high, with Hutchison now entering the 18th month of a 50% off promotion on SIM-only smartphone tariffs. Its subsidies on high-value handsets continue to be at the high end as well. The company also has a €7 tariff for 1,000 min, 1,000 SMS and 1GB data, presumably to prevent churn of the Orange subscriber base on the popular €7.50 tariff (offering a similar 1,000 min, 1,000 SMS and 1GB data). In Q4 2012, the combined H3G/Orange entity captured 47% of subscriber net additions despite the incumbent, TKA, significantly increasing handset subsidies in the quarter. If the momentum at H3G continuous, it looks set to become the second-largest mobile operator in Austria by early 2015 surpassing T-Mobile, which reported a market share of 30% in Q4 2012.

Exhibit 3.21: Austrian market would have been highly concentrated without spectrum remedies

Source: Berenberg Source: Berenberg. The chart is designed to show the EC desired scenario, ie

that a new entrant takes the 800MHz reserved spectrum in September 2013

2200

2400

2600

2800

3000

3200

3400

3600

2400 2600 2800 3000 3200 3400 3600 3800 4000

Se

rvic

e r

ev

en

ue

HH

I

Spectrum HHI






Dec'05: Pre tele-ring

Apr'06: Post deal, pre remedies

Dec'12: pre H3G/Orange

Dec'06: post remedies

H3G/Orange pre remedies

H3G Orange post remedies

2200

2400

2600

2800

3000

3200

3400

3600

2300 2800 3300 3800

Su

bs

share

HH

I

Adj. spectrum HHI



63

How has the Everything Everywhere JV performed?

In our case study section (page 59), we have outlined the remedies that were enforced by the EU in order for approval for the FTE/DTE Everything Everywhere (EE) JV. In this section, we look at EE’s performance, both operational and financial, based on the initial expectation of the JV, as presented to the market in September 2010.

In a nutshell, the JV has been a success for the partners, albeit on a relative basis. This is due to the fact that competition in the UK market has been such that top-line aspirations have been curtailed and that despite EE extracting the anticipated synergies, profitability and cash flow are unlikely to match the initial medium-term outlook. That said, the JV partners have fared much better than O2, thanks to the synergy benefits derived, and modestly better than VOD. This supports the shareholders’ decision, even though it is perhaps not creating the value they had first imagined due to competitive pressures.

EE – the ambition

The ambitions of EE when it launched were simple and are outlined in Exhibit 4.1.

Exhibit 4.2: EE targets and performance assessment since consolidation

Target Result

Become the network of choice. Positively accelerated customer migration to post-paid.

Market leader in service revenue market share

Successfully retained leadership.

Market-leading profitability EE has productively captured synergies and has improved profitability on a relative basis to its UK peers.

Deliver cash to shareholders EE has successfully upstreamed significant dividends and management fees to the JV partners.

Source: Berenberg estimates, company data

Has EE become the network of choice? Overall, EE has had a mixed performance, losing subscriber and service revenue market share in the last two years. While some of this is understandable, with 3 (the last entrant into the market) still growing, subscriber additions were the weakest of the four operators. This narrows somewhat in terms of service revenue share, and this reflects EE’s focus on post-paid subscribers.

The post-paid subscriber segment is where EE can justifiably suggest that it has met its operational target of being the network of choice, comfortably adding more post-paid customers than any other UK operator.

In 2010, when the JV started trading, the base was a little over 27m, which at the time was a 35.7% share of the UK market by subscribers. Since launch, however, EE has lost around 1m subscribers, with its market share falling by just over 2% – the weakest performance of the four operators if we just look at total subscribers. This leaves EE’s market share at 33.4% at the end of March 2013.


64

Exhibit 4.2: UK mobile subscribers and market shares

Source: Company data, Berenberg

Exhibit 4.3: UK mobile subscribers’ market share


The story on service revenues has been better, with EE’s share falling by just 0.5% in the last two years and rising slightly over the last six months. The big loser here has been O2, whose market share of service revenues has fallen by 3% over the last two years.

Exhibit 4.4: UK mobile market share service revenues


Mar-11A Jun-11A Sep-11A Dec-11A Mar-12A Jun-12A Sep-12A Dec-12A Mar-13A

3 mobile 7.24 7.54 7.66 7.54 8.11 8.45 8.75 9.05 9.29

EE 26.98 26.80 26.76 26.83 26.49 26.33 26.19 26.15 25.74

O2 22.29 22.14 22.21 22.17 22.33 22.37 22.48 22.86 22.91

Vodafone 19.15 19.01 19.33 19.33 19.17 19.07 19.31 19.54 19.22

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

Su

bsc

rib

ers

mln


3 mobile 9.6% 10.0% 10.1% 9.9% 10.7% 11.1% 11.4% 11.7% 12.0%

EE 35.7% 35.5% 35.2% 35.4% 34.8% 34.5% 34.1% 33.7% 33.4%

O2 29.5% 29.3% 29.2% 29.2% 29.3% 29.4% 29.3% 29.5% 29.7%

Vodafone 25.3% 25.2% 25.4% 25.5% 25.2% 25.0% 25.2% 25.2% 24.9%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


3 mobile 9.3% 9.8% 9.8% 9.9% 10.4% 10.8% 10.7% 11.2% 12.4%

EE 34.5% 34.4% 34.5% 34.5% 34.5% 33.8% 33.7% 33.9% 33.9%

O2 31.0% 30.5% 30.3% 29.5% 29.1% 30.4% 30.6% 29.3% 28.1%

Vodafone 25.2% 25.3% 25.4% 26.1% 25.9% 25.0% 25.0% 25.6% 25.7%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Se

rvic

e r

ev

en

ue

sh

are


65

If we look at service revenues in absolute numbers, EE has seen its revenues fall by around £400m pa since launch, or 3% pa compound. This is less than the 4% declines seen for the UK sector.

Exhibit 4.5: UK service revenues by operator


Part of the reason for an improved performance at EE is the focus on post-paid customers, and this may be a better metric with which to assess its success at being the network of choice. It has risen from being the operator with the least number of post-paid subscribers as a percentage of the base to number three, but has seen the highest increase of all operators over the last two years.

Exhibit 4.6: Proportion of post-paid subscribers


This change in dynamic is also demonstrated when one looks at absolute additions by operator. Here EE has been by far the strongest performer in terms of net contract additions. In this regard, it can be argued that the company has delivered in terms of pushing itself as the network of choice.


3 mobile 416 430 438 440 453 477 477 485 520

OrangeTmobile 1,541 1,503 1,542 1,526 1,503 1,486 1,496 1,467 1,422

O2 1,383 1,334 1,356 1,305 1,266 1,336 1,360 1,267 1,180

Vodafone 1,125 1,104 1,138 1,155 1,129 1,102 1,112 1,109 1,078

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

Se

rvic

e R

ev

en

ue

s £

mln


Vodafone 50.40% 51.90% 52.00% 52.90% 53.80% 55.00% 55.80% 56.40% 57.80%

O2 47.79% 48.21% 48.46% 49.35% 50.00% 51.02% 51.68% 52.05% 53.04%

EE 44.88% 46.05% 46.82% 47.86% 49.05% 49.92% 51.15% 51.99% 53.45%

3 mobile 53.86% 54.03% 53.84% 55.26% 55.54% 57.26% 57.72% 58.14% 58.14%

50.4%

51.9% 52.0%52.9%

53.8%55.0%

55.8%56.4%

57.8%

40%

42%

44%

46%

48%

50%

52%

54%

56%

58%

60%


66

Exhibit 4.7: Net post-paid additions by operator since the launch of EE

Source: Berenberg, company data

To be the leader in service revenue and profitability by extracting synergies? Overall, EE has maintained its market leadership in terms of service revenues, with the gap between EE and the number two player, O2, widening despite the drop in market share at EE. So it can point to success on this front. In addition, the company has been successful in extracting the available synergies, with the company delivering a run rate of £369m (83% of the target) at the end of 2012, just over £100m ahead of target, which will likely lead to the 25% target EBITDA margin being met.

However, we feel that its overall targets on revenues, EBITDA and OpFCF, are looking increasingly stretched, which we believe is caused by an increase in competitive intensity, which has affected revenue expectations, given the success of EE in terms of the delivery of synergies.

The company has pointed to double-digit CAGR cash flow growth between 2009 and 2014, without a step-change in capex. By working backwards and using this last leg of the outlook, we can assess the implications for OpFCF, EBITDA and revenues.

While we were never provided with full-year numbers for 2009, we had comparable nine-month figures. Using this as a barometer for full-year performance, it implies that OpFCF was, we estimate, £850m to £900m. As an educated starting point, we can get to the original target for 2014 OpFCF of £1,350m to £1,450m. If we assume that capex synergies come through, implying medium capex of around £500m, then EBITDA is expected to be between £1.8bn and £2.0bn. With a target margin of 25%, this implies revenues of £7.2bn to £8.0bn. As we detail below, these targets look challenging.

For revenue, it is always difficult to gauge the extent to which regulatory headwinds are factored into expectations, but we estimate that EE will see a regulatory impact on revenues to the tune of between £750m and £800m between the launch and the end of 2014. With the discussion on the current MTR glidepath well advanced at the time of the launch of EE, we expect that this was well understood when the targets were issued.

However, the company had pointed to some top-line initiatives, namely a strengthening of its leadership in UK mobile and new business opportunities (fixed line, B2B and new initiatives). The new business opportunities were expected to

1,650

1,700

1,750

1,800

1,850

1,900

1,950

2,000

2,050

2,100

2,150

Vodafone O2 EE 3 mobile

Po

stp

aid

ad

ds

sin

ce l

au

nch

of

EE

th

d


67

add around £500m to revenues by 2014. In combination, while not explicit (as we outline above), the rest of the outlook statement suggests the company was anticipating top-line growth. We believe that this ranged from a CAGR of between 0% and 2.1% pa, compared to the decline of 2.6% pa in the past three years.

If we break down the revenues into the elements of mobile service revenues and other revenues, we can build a picture for how the top line has evolved. Service revenues have been affected by regulatory headwinds, but have grown modestly on an underlying basis driven by the focus on the post-paid segment and churn reduction.

Exhibit 4.8: EE mobile service revenue waterfall chart, 2010 to 2012


The greater disappointment, we suspect, is in other revenues, which saw material contraction in 2011 before recovering somewhat in 2012. We suspect that this has been partly driven by weaker-than-expected performance in the Home segment, although this has now stabilised.

Exhibit 4.9: EE other revenues evolution 2010 to 2012

Source: Company data

6,296

6,1676,112

5,952

(132)3 (55)

(316)

156

5,500

5,750

6,000

6,250

6,500

Servicerevenues

2010

MTR UnderlyingChange

Servicerevenues

2011

AccountingAdj

Servicerevenues

2011

MTR UnderlyingChange

Servicerevenues

2012

753

617 672 705(136)

55 33

0

100

200

300

400

500

600

700

800


68

With revenues set to fall short of expectations, EE has also still to translate the extraction of the synergies into improved profitability, although the 2014 target margin of 25% is achievable. The margin is creeping up: having troughed in 2010 at 19.6%, it exited 2012 at 21.2% (22% in H2 2012), and is on the right trajectory to hit the outlook. This was to be driven by synergies that had been identified, totalling £545m pa.

Exhibit 4.10: EE synergies identified


At the end of 2012, the company was ahead of target with 83% (£369m) of the identified synergies delivered. In Exhibit 4.11, we have estimated how EBITDA has progressed over the first two full years of EE. Considering the fact that synergies are well ahead of expectations, and notwithstanding the impact of regulation, this is a frustrating performance.

Exhibit 4.11: EE EBITDA waterfall chart

Source: Berenberg estimates

201

445

545

83

161

100

0

100

200

300

400

500

600

Networkand savings

Distributionand

Marketing

Other CostSavings

OpexSavings

Capex TotalSavings

1,382 1,416 1,410

(265) 66 139

94

(127) 158

112

(149)

0

250

500

750

1,000

1,250

1,500

1,750

2010EBITDA

Revenuedecline

MTRCosts

Synergies Other 2011EBITDA

Revenuedecline

MTRCosts

Synergies Other 2012EBITDA


69

As we say, both the margin and synergies are on target, which further emphasises that the bane of all incumbent telecoms operators, the top line, has been the root cause of why EE is under pressure to deliver its medium-term outlook. We feel this reflects moderately worse-than-expected regulatory headwinds and a material uptick in competitive intensity.

If we look at the other operators, O2 has experienced a 6.6% compound decline in EBITDA over the 2009 to 2012 timeframe. VOD has fared better but still saw a compound decline of 1.8%, worse than our estimated 1.6% compound decline at EE. Both have experienced top-line pressure over this period, particularly O2.

In conclusion, we believe that there is reasonable evidence to suggest that despite EE’s parents delivering on cost-cutting at the time of the merger, either the remedies which is seeing potential spectrum constraints ease at 3 and/or intense competition continuing despite consolidation has eroded most of the commercial benefits. This is clearly shown by Exhibit 4.12 which shows the evolution of the HHI through the consolidation process and that concentration is heading back to the pre-consolidation levels.

Exhibit 4.12: HHI changes pre- and post-consolidation


Dec'09: Pre EE JV

Dec'10: post EE JV deal, post

remedies

Dec'10: postEE JV deal

pre remedies

Dec'12

2200

2400

2600

2800

3000

3200

3400

3600

2200 2700 3200 3700

Serv

ice r

ev

en

ue H

HI

Spectrum HHI



70

Market reviews – key conclusions

In this section, we aggregate our thinking from our eight country market assessments together with our review of consolidation case studies to rank the eight markets according to their scope for consolidation remedies to have a relatively moderate effect.

To be clear, all our judgements are relative – we are simply trying to assess which markets are most or least likely to attract relatively light consolidation remedies or are best/worst positioned to mitigate the impact of consolidation remedies.

Key findings

Which markets look least likely to benefit from in-market consolidation?

We think that Sweden, Denmark and the Netherlands are the three markets where the scope for remedies appears relatively more significant, and therefore we see these markets as relatively poorly positioned to mitigate the effects of remedies that would be likely to accompany in-market consolidation.

Sweden: Swedish consolidation is exposed to MVNO remedies and faces a significant hurdle given the existing network-sharing JVs and the scope for reallocation of spectrum to a new entrant. The existing network JV between TEL and TEL2B provides synergies to both players making consolidation with 3 less likely. On other criteria, Sweden ranks in the middle ground on pricing, spectrum concentration and market concentration. Finally, Sweden is one of the better performing mobile markets in our sample group and so the “opportunity cost” of not seeing consolidation appears low. In summary, we think consolidation in Sweden is unlikely, but should it happen, we would expect to see challenging remedies put in place.

Denmark: Spectrum concentration and the existing network JV between TLSN and TEL are the main hurdles to consolidation in Denmark. The most obvious consolidation pairings would result in an over-concentration of spectrum (eg TLSN and 3 would own 46% of spectrum, while TEL and 3 would own 38% of spectrum on an adjusted basis), raising the threat of effective spectrum reallocation remedies. While the Danish Competition Council may have applied a more lenient approach to the TLSN/TEL network JV than the EC might have done (eg allowing spectrum-sharing), the implication is that TLSN and TEL opted for a JV because a full merger would have been opposed or would have attracted even more significant remedies.

Netherlands: The Dutch mobile market has high prices and is the most concentrated of the eight markets we review, and while spectrum distribution is broadly in line with the average of the eight markets, effective network-sharing beyond the existing legal obligation is limited. Anti-trust authorities would see scope for remedies including the divestment of customers and infrastructure, possibly the divestment or partitioning of spectrum for use by competitors or new entrants, and increased obligations on MNOs to provide network access to MVNOs. The Netherlands’ track record of allowing consolidation (Orange and T-Mobile, KPN and Telfort) has resulted in a market that provides limited further scope for consolidation without material offsetting synergies.


71

Exhibit 5.1: Markets most likely to see consolidation benefits exceed impact of remedies

Source: Berenberg

What about Spain? Similar to the Netherlands, the Spanish market is characterised by relatively high prices (even with the recent reductions) and a relatively concentrated market (even with the recent rise of competition from MVNOs), and while spectrum distribution is broadly in line with the peer group, consolidation would likely require some level of spectrum divestment or reallocation. As with the Netherlands, our analysis implies that anti-trust authorities would respond to consolidation with remedies, possibly including increased obligations to provide network access to MVNOs, and divestment of excess spectrum.

Which markets might potentially benefit from in-market consolidation?

We think Germany and France are the markets where the scope for consolidation remedies appears relatively less significant, and so these markets appear to us to be relatively better positioned to mitigate the effects of anti-trust remedies.

France: France already has very low prices, a strong MVNO presence, and existing network-sharing arrangements between three of the four network operators. Furthermore, given recent market performance in the wake of the entry of ILD, the potential gains from consolidation could be material. So with relatively less scope for remedies to be effective and a relative large amount to gain, we see France as among the best positioned market to benefit if consolidation were to arise.

Germany: While German prices generally remain above average, the market does appear to be the most fragmented if the strong MVNO presence is reflected in the assessment. Spectrum is already evenly balanced in quantity, although E+’s lack of 800Mhz could be an issue addressed by regulators if consolidation omitted E+. The lack of network-sharing means this is one area where remedies could act to offset the benefits of market repair, with incumbent MNOs being asked to provide access to potential new entrants, or existing players with weaker network positions. On balance, though, the characteristics of the German market suggest it is among the best positioned in terms of the potential remedies that could result from consolidation.

CriteriaPrices &

bundles

Market

concentration

MVNO

presence

Consumer

choice

Spectrum

balance

Network

sharingGrowth

Regulatory

attitudeTotal Score

Score from 1 to 10

where 10 indicates

most positively

exposed, and 1

indicates least

postiively exposed

…prices are

already

relatively low,

bundles are

big

…the market is

relatively more

fragmented

...MVNOs

have a strong

presence

already

…independent

retailers or

resellers have

a strong

position in the

market

…spectrum

will be evenly

balanced post-

consolidation

…network

sharing

presents

limited scope

for

incremental

remedies

…the

opportunity

cost of not

seeing

consolidation

is high

…regulatory

authorities are

not overtly

opposed to a

reduction in

networks

Weighting of factor 10% 15% 15% 5% 15% 10% 5% 25% 100%

FR 10 5 9 6 4 7 3 5 6.10

GE 4 10 10 10 2 2 3 5 5.80

UK 8 9 6 9 4 9 3 1 5.40

IT 3 7 3 7 1 5 10 5 4.55

ES 1 3 4 7 3 6 10 5 4.30

NDL 1 2 8 5 4 2 7 5 4.25

DK 6 8 5 2 3 3 7 1 4.00

Swe 5 4 2 7 3 3 2 5 3.85

AU 10 0 1 5 4.5 4 10 3 3.73

Consoldation would be more likely to gain approval, and remedies less likely to have an impact, if…….:


72

Two special cases – the UK and Italy

UK – special case: Ofcom is widely quoted as stating in its 4G auction consultation that it desires to see at least four network operators in the UK market, suggesting very limited scope for MNO consolidation to occur, which is a shame, as our analysis shows that the UK market is among the best positioned to cope with the likely remedies that would result from network consolidation. The UK market is relatively low priced, and is among the least concentrated, with a meaningful MVNO presence and a strong independent retail channel presence. Network-sharing is already widespread, limiting the scope for incremental damage from network access remedies. With these characteristics, we think anti-trust remedies would be challenged to increase the level of competitive intensity in the UK market. The exception relates to spectrum. While spectrum is evenly balanced currently, any deal would be likely to fall foul of Ofcom’s spectrum cap, resulting possibly in the reallocation of spectrum across the other players, or even to a new entrant (note BT’s presence in the market, with only 2,600Mhz spectrum at present).

Italy – significant remedies seem likely: Italy’s two Achilles’ heels are the lack of a strong MVNO segment and the likelihood that a deal between TI and 3 would result in a significant spectrum remedy. While most of the current speculation has focused on a TI and 3 combination, we could also not rule out a combination between Wind and 3. MVNOs in Italy have generally struggled, with a collective market share of only 4-5%. One MVNO, PosteMobile, has gained about half of that share, illustrating the importance of extensive high street distribution. So while Italy ranks in the middle ground of our overall ranking of eight markets, we still feel that a combination of TI and 3 would face significant remedies, including the reallocation of spectrum, potentially to invite in a new entrant, and the possibility of better terms and conditions to strengthen the MVNO sector.

A walk through our criteria

Pricing levels/bundles: Put simply, we assume that where prices are already low, consolidation, even with remedies, may stand a better chance of resulting in market repair as low price levels will disincentivise new entrants.

MVNO presence: Where MVNOs are already a powerful force in the market, there will be less scope for regulators to use MVNO remedies to counterbalance consolidation. Where MVNOs with strong distribution channels are not already present, the introduction of MVNOs could present a good counterbalance to consolidation.

Herfindahl-Hirschman Index (HHI) (pre- and post-MVNOs): Where HHI is low, the market would appear to be already very competitive, so the consolidation of two players may be more acceptable, and may leave little scope for new entrants to make an impact, or for the introduction of more consumer-friendly MVNO privileges.

Extent of network-sharing: Where network-sharing is prevalent, there should be less scope for regulators to enforce access to shared infrastructure as a remedy. However, too much network-sharing, or arrangements that are too complex, could act to prevent operators pursuing full consolidation. Also, network-sharing without consolidation at the point of sale to the customers (to remove actual competitive tension in the market) should serve simply to allow operators to extract efficiencies which will then just be passed onto consumers


73

in the form of lower prices, sustaining competitive tension. This can be particularly important where smaller operators share infrastructure (as in Sweden and Denmark) to gain enough scale to compete with the incumbent.

Strength of competition in distribution: Where the customer has lots of choice at the point of sale, there should be less scope for remedies to make a difference. The presence of strong reseller/retail distribution channels (such as Carphone Warehouse in the UK) should add to the impression that the consumer already has a broad choice of alternatives. Likewise, if the consolidation of operators can also reduce the level of competition at the point of sale without being offset by regulatory remedies, this should be good for market repair.

Spectrum balance among competitors: Where there is an equal distribution of spectrum among competitors (post-consolidation), there is less scope for regulators to enforce spectrum reallocation remedies. Likewise, where there is a significant imbalance between operators’ share of subscribers and spectrum, the scope exists for one or more residual competitors to price disruptively, possibly irrespective of consolidation. We ask whether consolidation could remove potentially disruptive excess capacity from the market, and whether spectrum divestments could create such an opportunity for a new entrant (ie for MVNOs’ ongoing own-network, or for wireline competitors looking to enter mobile).

The opportunity cost of no-consolidation: Maybe consolidation will not result in a more rational market, but perhaps it could result in a less disruptive one – it is all relative – so can we identify which markets have a high opportunity cost of not pursuing consolidation? Put another way, there may be markets where consolidation is worth pursuing, even with significant remedies, simply because the current competitive environment is so irrational that any level of improvement would be welcome. We compare recent mobile market service revenue growth trends with our mid-term expectations for growth in order to assess which markets face a high opportunity cost of not seeing consolidation.

Price levels/bundles

Conclusion: Across the eight markets that we review, our analysis suggests that prices are the lowest in Austria, France and the UK. Our work suggests that prices remain highest in the Netherlands and Spain. Prices in Germany and Italy can appear above-average, but much depends on which spend level is being considered. For example, while German contract prices in general appear above-average (but not the highest), German pre-paid prices offer relatively cheap per minute voice and per message SMS prices. A similar picture emerges in Italy where pre-paid tariffs dominate, and where headline offers appear above-average versus the peer group, but where competition in the low-end pre-paid space presents the scope for access to very low per minute or SMS rates once recharge levels increase above €1-2 per week.


74

Source: Berenberg

Source: Berenberg

Source: Berenberg

A note of caution is deserved when considering the comparison of prices across national borders. While we have tried to consider comparable tariffs at comparable spend levels, the sheer variety of offers facing the consumer across the eight markets that we examine does present a difficulty. For example, tariffs in Denmark tend to offer larger bundles of usage such that even a relatively low volume user is

Exhibit 5.2: SIM-only prices appear lowest in France, Austria and the UK; the Netherlands, Spain and Italy appear above-average for SIM-only prices

Exhibit 5.3: Including handsets in the cost does little to change to least costly markets; Austria, Sweden, the Netherlands and the UK appear to be subsidising iPhones the most, France the least

Exhibit 5.4: For pre-paid tariffs, Austria, Sweden and Germany appear to be cheapest, while the Netherlands and the UK appear to be the most expensive

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

FR UK DK NDL AU Avg. IT ES GE Swe

€p

cm

Typical SIM-only bundle prices €pcm

10.0

20.0

30.0

40.0

50.0

60.0

70.0

FR AU UK DK Avg. Swe GE IT ES NDL

€p

cm

Typical unlimited (voice & SMS) SIM-only big bundles

30.0

40.0

50.0

60.0

70.0

80.0

90.0

100.0

AU UK FR NDL DK Avg. Swe GE IT ES

€p

cm

Typical unlimited (voice & SMS) SIM+iPhone 5 (16Gb)

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

AU Swe NDL UK GE Avg. IT DK ES FR

€p

cm

Implied iPhone recovery embedded in ARPU

0.0

10.0

20.0

30.0

40.0

50.0

AU Swe GE IT ES Avg. NDL UK

€p

cm

Prepaid price/min €c

0.0

5.0

10.0

15.0

20.0

25.0

AU Swe GE Avg. ES IT UK NDL

€p

cm

Pre-paid SMS rates - entry level


75

better off up-tiering to a big-bundle or unlimited bundle rather than trying to find a tariff with a usage profile that suits (we note that low usage tariff structures are increasingly rare). Nevertheless, we think the broad impression created by our price comparisons is sound.

We present our pricing comparisons in Exhibits 5.2 to 5.4. We compared three post-paid (or equivalent) tariff structures and then entry level pre-paid per minute and per SMS rates. Our comparisons show:

1. SIM-only post-paid tariffs for a usage profile based on 100mpm of voice, unlimited SMS and 250Mb of data;

2. SIM-only “big-bundle” tariffs incorporating unlimited any-network voice minutes, unlimited any-network SMS and 1Gb of data;

3. an iPhone 5 16Gb post-paid tariff with the same big-bundle used in 2) above (in addition to enabling us to compare post-paid tariffs using a standard high-end phone, this approach also enables us to assess which operators are utilising the highest handset subsidies);

4. per minute voice rates for entry level pre-paid offers that require a small recharge (typically €5) – ie a level of recharge that does not provide access to significant volumes of “free” or bundled minutes;

5. per SMS rates for those entry level pre-paid offers that require a small recharge (typically €5) – ie a level of recharge that does not provide access to significant volumes of free or bundled SMS volumes.

On our analysis, we would question whether remedies resulting from in-market consolidation would be effective in markets such as Austria, France and the UK given the existing low level of prices. At the very least, we would have to question the motives of any new entrant seeking to take advantage of remedies such as the ring-fencing of spectrum, enforcement of roaming and access to tower infrastructure, given the existing low level of prices. Vice versa, in higher priced markets such as the Netherlands and Spain, there would seem to be scope for consolidation-related remedies to facilitate further price competition.

HHI pre- and post-MVNOs

Conclusion: Using a traditional HHI calculation based on service revenue market share, the least competitive (most concentrated) of our eight markets are the Netherlands, France and Spain, while the most competitive (least concentrated) are the UK, Italy and Denmark.

However, in the middle chart of Exhibit 5.5, we show the HHI ranking based on a broader definition of market participants that includes the MVNO segment as an additional competitor. Information on MVNO presence in the eight markets that we reviewed is scarce, so generally we were unable to disaggregate incumbent network operators’ own MVNOs in order to conduct this exercise. Even so, at least for illustrative purposes, the comparison with the traditional HHI ranking is informative.


76

Exhibit 5.5: When the presence of MVNOs is taken into account, Germany, the UK and Denmark appear relatively less concentrated, perhaps implying the best chance of avoiding tough remedies

Source: Berenberg

When MVNOs are treated as a separate competitive force, the conclusion is most different in Germany, which moves significantly up the ranking to take the place as the most competitive (least concentrated) market. France also moves up the ranking by two places, while Italy slips by two places due to its underdeveloped MVNO segment. The Dutch market sees a significant reduction of almost 400 points in its HHI score when MVNOs are included, yet still retains the position as the least competitive (most concentrated) of the eight markets we examine. On this analysis, it would appear that Germany, France and the Netherlands offer the least scope for MVNO remedies to counter the beneficial effects of any consolidation that might take place.

Spectrum share versus customer share

Conclusion: Austria says it all. Take a look at the last chart in Exhibit 5.7 which shows spectrum and revenue market concentration for Austria pre- and post- remedies. Before remedies, the combination of Orange and Hutchison left Austria actually relatively well positioned versus our eight focus markets, at least from the standpoint of spectrum concentration. But look at the effect of spectrum remedies – Austria is shifted way to the left of our eight-market peer group. The implications seems clear; if future in-market consolidation is treated in the same way as the EC treated Austria, then all markets should expect some degree of spectrum remedy to avoid increasing spectrum concentration.

0 1000 2000 3000 4000

UK

IT

DK

GE

Swe

Avg.

ES

FR

NLD

HHI using only MNO's

0 1000 2000 3000 4000

GE

UK

DK

IT

Avg.

FR

Swe

ES

NLD

HHI using MNO+MVNO's

0 200 400 600 800

GE

FR

NLD

UK

Avg.

DK

ES

IT

Swe

Reduction in HHI when MVNOs are incorporated


77

Source: Berenberg

Looking at specific markets, on first glance Germany and the Italy stand out as outliers at opposite ends of the range of outcomes.

• Germany looks good: As already noted, Germany has a relatively low service revenue HHI, implying a relatively low level of market concentration – a relatively favourable position to be in when it comes to in-market consolidation remedies. However, Germany also has the lowest spectrum HHI when higher frequency spectrum bands (above 1Ghz) are adjusted to equalise frequency spectrum bands. Germany stays an outlier when our HHI calculations are adjusted for an assumed combination of O2 and E+ – while the spectrum HHI does increase, suggesting some degree of spectrum remedies would be imposed, the service revenue HHI remains well below comparable in-market consolidation pairings in other markets.

• Italian consolidation looks challenged: Without an assumption of in-market consolidation, Italy appears in line with the peer group average on service revenue and spectrum HHI. However, our analysis suggests that a combination of TI and Hutchison’s 3 would result in an obvious outlier on both service revenue and spectrum HHI. Significant remedies would be likely in the event that talks between TI and 3 progressed to a consolidation proposal.

Exhibit 5.6: Comparing market concentration using service revenue share and spectrum share

2200

2400

2600

2800

3000

3200

3400

3600

2400 2600 2800 3000 3200 3400 3600 3800 4000

Se

rvic

e r

ev

en

ue

HH

I

Spectrum HHI







78

• UK and French markets look relatively well placed: The UK is an outlier which looks relatively well placed for consolidation, according to our analysis. Even when factoring in the most obvious deal (VOD and 3) the UK still compares favourably on service revenue and spectrum HHI – ie it remains among the least concentrated markets. However, the UK faces a clearly outlined spectrum cap published by Ofcom as part of the 4G spectrum auctions, which undermines the UK’s ranking on this factor. France is currently in line with the peer group average on service revenue and spectrum HHI, but moves to a more favourable position in the event of the most obvious combination (Bouygues and ILI) when compared with similar consolidation pairings in other markets.

Our conclusions are based on the idea explained in Exhibit 5.6 and illustrated in Exhibit 5.7. In summary, we plot on a dual axis chart the HHIs for our eight peer markets with one axis plotting the service revenue HHI (the traditional measure of market concentration) and the horizontal axis plotting the spectrum HHI (as a measure of the distribution of spectrum in the market).

The exhibits use the same scale for four different scenarios to illustrate how changes to assumptions result in the markets moving around the charts. We also use a service revenue HHI that is adjusted for the presence of MVNOs in the market. Our consolidation scenarios (charts in the right column of Exhibit 5.7) assume either the most talked about or most obvious consolidation scenario in each market.

When looking at these charts, a presence in the top-right quadrant is relatively bad, and a presence in the bottom-left quadrant is relatively good. Markets in the lower-left quadrant are characterised by higher levels of real competitors and/or more equal market shares. Markets in the top-right quadrant are characterised by fewer real competitors and/or unequal market share and spectrum distribution.

The top-left and bottom-right quadrants are perhaps more interesting. Markets occupying the top-left quadrant have a more concentrated market (ie they are relatively less competitive) but a more equal distribution of spectrum, suggesting either that smaller competitors have so far been unable to leverage their spectrum position to gain significant market share, or that these smaller competitors are quite recent entrants to the market. In the event of consolidation in these markets, we might expect anti-trust authorities to seek non-spectrum-based remedies to support smaller players, perhaps via the introduction of more favourable roaming terms, or access to incumbent infrastructure.

Markets in the lower-right quadrant are relatively more competitive but the sustainability of competition may be jeopardised by an uneven distribution of spectrum. Consolidation in Germany (assuming O2 and E+) results in this situation, shifting Germany from the lower left quadrant to the lower right quadrant. We feel that anti-trust authorities are much more likely to seek spectrum-related remedies in this scenario.


79

Source: Berenberg

The extent of network-sharing

Conclusions: We think extensive network-sharing agreements in the UK and France, followed by less extensive but still significant sharing arrangements in Denmark, Sweden and Spain undermine to some degree the ability of regulators to use enforced network-sharing as an effective remedy to counter in-market consolidation.

We detail the main network-sharing arrangements in our eight focus markets in Exhibit 5.8. As mentioned in our section introduction, network-sharing could be a doubled-edged sword when it comes to consolidation. It seems logical to believe that markets with extensive existing network-sharing offer less scope for network-sharing to be used as an effective consolidation remedy than markets without existing network-sharing agreements.

Exhibit 5.7: Italy appears likely to face the greatest spectrum-related remedies; Germany similarly has above-average spectrum concentration, but a strong MVNO segment makes it most competitive

NDL

Swe

FR

DK

ES

UK

IT

GE

Avg.

2200

2400

2600

2800

3000

3200

3400

3600

2400 2900 3400 3900

Se

rvic

e r

ev

en

ue

HH

I

Spectrum HHI


AU

NDLSwe

FR

DK

ES

UK

IT

GE

Avg.

2200

2400

2600

2800

3000

3200

3400

3600

2400 2900 3400 3900

Se

rvic

e R

ev

en

ue

HH

I

Spectrum HHI


(with consol.)

AU

NDL

Swe

FR

DK

ES

UK

IT

GE

Avg.

2200

2400

2600

2800

3000

3200

3400

3600

2400 2900 3400 3900

Ser

vic

e R

even

ue

HH

I

Spectrum HHI

Adjusted spectrum concentration vs market concentration

AU

NDL

Swe

FR

DKES

UK

IT

GE

Avg.

2200

2400

2600

2800

3000

3200

3400

3600

2400 2900 3400 3900

Se

rvic

e R

ev

en

ue

HH

I

Spectrum HHI

Adjusted spectrum concentration vs market concentration (with consol.)

AUAU incl.remedies


80

Exhibit 5.8: European network-sharing agreements

Source: Berenberg

Date Country Operators Sites Back-haul Full RAN Spec-trum Core n/w Aggregate cost saving identified Comments/rationale

May-11 Austria Orange/T-Mob Y Y Y €30m per partner or 5-7% of opex 10% increase in coverage; joint network deployment

Jun-11 Denmark Telia/Telenor Y Y Y Y n/a Better coverage/technologies; catch up to network leader (TDC).

Feb-10 France SFR/By/Or Y Sign framwork agreement for sharing 3G sites from 11 Feb 2010. According

to some estimates 20-40% of total sites are already shared depending on the

operator.

Apr-09 Germany* TEF/VOD Y VOD: 10% of opex Vodafone has sublet towers to O2 on an agreed price per tower

incorporating c400 sites as per 2009/2010.

Nov-08 Italy VOD/TIM Y n/a Site sharing agreement for new deployment and for consolidation of

existing sites to achieve CAPEX avoidance for passive infrastructure and

OPEX savings by reducing rental spend.

Jul-09 Italy TIM/3 Y 30% of network costs (3 yrs) or 7-8% of opex Boost coverage; reduces cost of site leases, decommissioning of antennaes.

Jan-00 Netherlands KPN/Vod/DT Y n/a Dutch regulation enforces widespread site sharing.

Jul-12 Netherlands KPN/Vod n/a Piloting passive sharing of facilities around sites such as air conditioning

and power supplies. No cost savings noted, and project remains in the pilot

phase.

Oct-07 Spain Vod/Orange Y Y Y Orange: €200m (5 yrs) or 6% of opex 3G sharing in less dense areas; reduces sites needed by 40%. Agreement

was extended to 2G in June 2012.

Apr-09 Spain* TEF/VOD Y VOD: 10% of opex Joint build;share 2G/3G; economic situation was a catalyst:

• Spain: extend 2007 site sharing agreement to power/cabinets/mast.

Apr-09 Sweden Tele2/Telenor Y Y Y Y n/a Improved coverage, cost efficient investments for better mobile internet

access and to improve scale and efficiency relative to the incumbent.

Nov-08 UK T-Mob/3 Y Y Y £2bn (10 yrs) in total for both partners Merging 3G networks, 5k sites (out of 18k) closed. Sharing agreement was

retained as a regulatory requirement for the merger of Orange and T-Mobile

to form EE.

Apr-09 UK* TEF/VOD Y VOD: 10% of opex Joint build;share 2G/3G; economic situation was a catalyst:

• UK: focus on joint build , consolidation of 2G/3G sites

Feb-10 UK Orange / T-Mob Y Y Y Y Y £154m annual opex savings by year 4 (25% of network opex, 4% of total

Opex), and £100mln pa capex savings (20-25%) by year 4. £700mln+ capex

savings over 10 years.

Resulting from the merger of Orange and T-Mobile to form Everything

Everywhere, citing opex and capex savings through joint network

operations. Aim to reduce site count from 28k to 18k+.

Jun-12 UK TEF/VOD Y Y Y n/a 40% increase in coverage to match leader EE; accelerates 4G coverage.

Network sharing


81

Exhibit 5.9: The opportunity cost of not seeing in-market consolidation is greatest for Italy and Spain, then the Netherlands and Denmark


However, existing sharing could act as a hindrance to consolidation occurring in the first place. In Sweden, for example, the network-sharing arrangement between TEL2B and TEL could present an obstacle to any move to consolidation by either of the JV partners with a third-party (eg 3).

In addition, without actual consolidation of choice at the point of supply to the customer, it remains possible that efficiency savings from network-sharing would simply be reinvested back into the market share battle between the market participants.

Market service revenue trajectory

Conclusion: Italy and Spain face the highest opportunity cost of not seeing in-market consolidation. The UK, Germany and Sweden face the lowest opportunity cost of not seeing in-market consolidation.

We explore recent trends in underlying (ex-MTR) market service revenue growth, and compare these trends with our three-year forecast growth rates. We think the markets seeing the fastest declines are those that could most benefit from consolidation, albeit the finer nuances of the market structure may result in significant remedies as anti-trust authorities seek to protect consumers against the risk of being disadvantaged by consolidation. Even so, it might still be worthwhile

IT

FR

Swe

ES

NDL

GE

DK

UK

Avg.

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

-12 -10 -8 -6 -4 -2 0 2 4

2013e-1

6e C

AG

R

Q1/13 % yoy (ex MTRs)

Market serevice revenue growth -Q1/2013 vs 2013e-16e CAGR


82

for market players in Spain and Italy to seek consolidation on the basis that the outlook is so challenging that any form of consolidation would be better than no consolidation. There is no doubt that the situation in both Italy and Spain is affected by the weak domestic economic backdrop in each market. High prices in Spain, and an intense pre-paid promotions culture in the Italian market, also play a part in the weak performance and outlook for these two markets.

This is not to suggest that the UK, Germany and Sweden would be wasting their time pursuing consolidation, although there are other factors in both the UK (Ofcom’s objection to less than four networks) and Sweden (existing network JVs) that could prohibit or constrain consolidation moves in those markets.

Regulatory attitudes

All of our above analysis may of course be meaningless in the face of an explicit aversion to further network consolidation on the part of either the national regulatory body or national/European competition authority.

At the EC level, operators seeking to pursue network consolidation can take some comfort from the fact that the EC competition authority has approved such consolidation recently (eg in Austria, and previously in the UK and the Netherlands).

At the national level the situation is less clear.

Explicit opposition

o UK regulator opposed to fewer than four networks: Ofcom in the UK has clearly expressed its view in recent consultation documents that its preference is for a minimum of four infrastructure-based network operators in the UK market. As far as we are aware, there is no similar viewed expressed by the UK Competition Commission, although in any deal overseen by the UK Competition Commission, Ofcom’s view would be invited.

o Danish authority concerns focus on wholesale access and spectrum hoarding: The Danish Competition Council (DCC) in its review for the TLSN/TEL JV highlighted the importance of keeping the core networks separate to maintain effective market competition. In addition, they were concerned about the ability of each party to obtain spectrum and offer it to the JV. One of the remedies is that the JV has obliged itself to accept any MVNO/Service Provider], as long as it is on commercial terms. The fact TLSN and TEL opted for a JV and not to integrate the core network (and thus take competitive intensity out of the market) is probably because they did not expect an approval to such a deal. The fact the DCC accepted spectrum-sharing, even though this goes beyond the level of network-sharing the EC is happy with, suggests a more lenient approach compared to the EC.

o Swedish Competition Authority has seen little of size: The Swedish Competition Authority has had very few merger notifications related to telecoms and network consolidations, with the 1999 proposed deal between TLSN and TEL falling under the old EC remit, while the combination between Telia and Sonera that today makes up TLSN, also fell under the EC remit, and was not an in-market deal anyway. Typically the fibre deals done with network operators buying municipality operators, have fallen below the revenue thresholds relevant for the authority.


83

Danish consolidation is much needed, but looks unlikely

Will the market consolidate in the next two years and why?

3’s recent comments regarding potential Nordic consolidation mean that consolidation is very much in focus. An analysis of spectrum holdings as well as data usage shows that while 3 holds only 12% of the spectrum on an adjusted basis, it accounted for 37% of the data traffic in Denmark in H2 2012. So 3 could well be running close to its capacity limits, arguably increasing its appetite for M&A. But with a deal with TDC likely to be impossible, this leaves a deal with either TLSN or TEL as the most likely possibility. However, TLSN and TEL now pool spectrum together within their new JV, which would therefore complicate any deal with 3. In addition, a deal would likely be scrutinised by the EC, and given the multiple-brand approach, we would expect required divestments could add another layer of complexity. In short, while there may be appetite, the hurdles may be too challenging to overcome.

Which combinations are most likely to gain approval?

We would view a combination between TDC (the largest operator) and 3 (the smallest) to be unlikely. Such a deal would result in 54% service revenue share and would effectively leave Denmark with two networks. This in itself may not be a deal-breaker (given that the DCC applied mandatory wholesale obligations to the TLSN/TEL JV); however, the close competition across TDC’s low-end brands (more than half its consumer mobile base) with 3 and its low-end brand Oister would indicate competitive concerns for the low end of the market. This could be alleviated by TDC selling either Telmore or Onfone, but we do not believe this would be appealing. In conclusion, we do not consider a TDC/3 tie-up to be credible.

Arguably the most credible deal would be between TLSN (number three in the market) and 3 (number four). This would result in only a 363-point increase in the subscriber HHI, taking it to 2,941, and would leave the combined entity with a 28% subscriber share and a 34% service revenue share. However, in being left with 46% of the spectrum in the market (41% on an adjusted basis), it is feasible that spectrum may have to be handed back. If the Austrian precedent was followed, this could be auctioned to a new entrant. Denmark has some characteristics similar to the Netherlands, in that it is a flat and urbanised geography. So this new entrant spectrum could be seen as attractive. In addition, 3 and TLSN operate very similar pricing points in the Danish market, competing across a similar market segment. This would be seen as a negative for the EC in trying to come to an approval for the deal.

A deal between TEL and 3 would also come with many of the same issues. However another issue between TEL or TLSN combining with 3 is that they operate together through a JV. The question then is why would either operator want to walk away from a JV with a more valuable spectrum portfolio, into a tie-up with an inferior spectrum portfolio? The TLSN/TEL JV in Denmark has 53% of the spectrum in the market – putting it in an incredibly strong position, without being forced to hand spectrum back. Risking that position in an uncertain consolidation scenario makes little sense. In addition, the costs of breaking up the JV could be high, and both TLSN and TEL are already benefiting from the network synergies that come with the JV.


84

Could imposed remedies be a deal-breaker?

Denmark already has a mature MVNO market, so there is unlikely to be regulatory pressure to force MVNO access. However, although there are four operators in Denmark, there are three networks. A reduction to two networks is unlikely to be acceptable. So any further consolidation in the Danish market would require either the break-up of the current TLSN/TEL JV, which could be costly, or/and spectrum being auctioned to a new entrant. As we highlight in the scenarios in Exhibit 6.1, another remedy would likely be a divestment of part of a subscriber base, adding a further complication to a deal. This could potentially make any deal very costly – given that the ability to buy spectrum cheaply, roll out in a small flat territory and buy a subscriber base cheaply could add a lot of credibility to a new entrant threat.

Exhibit 6.1: Market structure makes consolidation in Denmark a tough ask

Source: Berenberg

Potential deal Market issues to consider Potential remedies Conclusions Deal likelihood (1-10)

Telia + Three

1) Telia already shares

network in JV with Tele2

2) It also shares spectrum - so

an agreement between Telia

and Three would also

require an agreement

between Telenor and Telia

3) It could be costly to leave

the JV relationship

4) Telia already has network

synergies through its JV with

Tele2.

1) We do not believe that Three

could simply be brought into the

Net4Mobility JV, mainly because

it would leave two main

networks in Denmark, and the JV

would be left with 71% of market

spectrum 2)

Telia + Three would be left with

46% of the spectrum and 34% of

service revenues. So spectrum

divestments are a reasonable

remedy 3) Per the

DCC, there is already a remedy

for the JV to have to provide

wholesale access to anyone on

reasonable terms. So something

more severe could be imposed in

a full consolidation, such as

tower sharing or roaming

agreements

4) Because Telia and Three

compete closely, subscriber

divestments could also be forced.

DEAL IS UNLIKELY: A deal

would be expensive to execute

given the current JV structure,

and would also come with more

limited incremental synergies.

While the combined market

shares are unlikely to be a big

issue, the combined spectrum

share might. Spectrum allocated

to a new operator in Denmark

could be a credible threat.

3

Telenor + Three

Much the same market issues

as a deal between Telia and

Three.

Much the same market issues as

a deal between Telia and Three

although spectrum divestments

would potentially be less onerous

as there is less of a mismatch

with subscriber and service

revenue share.

DEAL IS UNLIKELY: For many of

the same reasons as a Telia +

Three tie up

2

TDC + Three

From a market perspective

this deal would make the

most sense - as it would

boost the spectrum holdings

at TDC and remove an

unwanted competitor.

1) The combination would be left

with just under 50% spectrum

share. While in isolation not an

issue, the fact that Denmark

would be left with two networks

would be a problem.

2) For this reason, spectrum

would almost certainly have to be

divested to a new entrant.

3) We could also expect a

wholesale obligation on the joint

network. 4)

TDC would have to divest one of

its low end brands (Onfone or

Telmore), potentially to the new

entrant.

DEAL IS UNLIKELY:With 54%

service revenue share this would

be a tough deal to get approved.

In addition, the remedies would

likely be severe enough, that

some of the logic to the deal may

not work.

1


85

Market overview

The Danish market has a history of intense competition across a disparate set of brands. This has led to consolidation of brands, although with a six-month contract rule in place and structurally higher churn, it remains an intensely competitive market.

• With a population of just over 5.5m people, Denmark has 8.3m mobile subscriptions.

• Of these, 1m are dedicated mobile broadband subscriptions. This equates to 18.6% of the population taking a dedicated mobile data package.

• In total, 65.5% of SIMs in Denmark are using data, with 17% of regular SIMs (so excluding dedicated mobile broadband) taking a dedicated add-on data package.

• The average dedicated mobile broadband subscription in Denmark used 2.2GB of data per month in H2 2013. The average mobile data user was only using 0.4GB of data.

• The usage varied widely across operators, though. The average mobile data user on 3 used 0.77GB per month, while the average user on TLSN used 0.38GB. In contrast, the average TDC user consumed only 0.2GB of data.

In Exhibit 6.2 below, we show that since end-2007:

• TDC has lost 1ppt of subscriber market share, having a share of 39.5% as at end-2012 (this was aided by the acquisitions of M1 and Onfone);

• TLSN has been the main loser of market share, losing 5.3ppt since end-2007;

• TEL lost 2.3ppt;

• the main gainer of market share has been 3, which has added 5.5ppt of share, taking it to 10.3% subscriber share;

• in addition, the MVNOs in the market have gained 3ppt of share.

In Exhibits 6.3 and 6.4, we show service revenues split by operator. Overall, the market has started moving backwards since 2010. This was started by an aggressive offer from MVNO Onfone, which was subsequently taken out by TDC. Since then, though, the environment has been categorised by a number of aggressive price wars, with TLSN and 3 being the most aggressive players in the latest round of competition. On our calculation, the market has lost 14% of its service revenues since end-2010, with only 3 managing to sustain strong growth.

In conclusion, the Danish market remains a very competitive space, and although the median pricing levels are not high from a European context, especially when we account for the large data bundles available, the range of pricing is wide, suggesting the market is going through another period of disequilibrium. In this process, TDC looks particularly exposed.


86

Exhibit 6.2: Limited subscriber growth left in the Danish market (subscribers quoted in thousands)

Source: Berenberg, Erhvervsstyrelsen (Danish regulator)

Exhibit 6.3: Service revenue growth declines start from 2010 (data in DKKm)

Source: Berenberg

2007 2008 2009 2010 2011 2012

Other 323 444 336 590 539 687

3 304 408 485 647 782 854

Telenor 1,680 1,813 2,038 2,014 2,007 2,015

Telia 1,449 1,493 1,460 1,450 1,426 1,462

TDC 2,551 2,704 3,102 3,089 3,334 3,274

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

2007 2008 2009 2010 2011 2012

Other 477 582 547 619 700 660

3 1,035 1,288 1,644 1,973 2,370 2,356

Telenor 4,123 4,422 4,570 4,649 4,118 3,558

Telia 4,220 4,678 4,346 4,042 3,506 2,945

TDC 6,703 6,750 6,660 6,707 6,626 5,966

-

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

20,000


87

Exhibit 6.4: TDC service revenue performance is masked by MVNO acquisitions (data is in DKKm)

Source: Berenberg

Market concentration

We use both market share and HHI to look at market concentration in the Danish market, and we do this looking at both subscriber share and service revenue share.

• Subscriber share (Exhibits 6.5 and 6.6): If we just look at market share of the MNOs (with MVNO and service provider subscribers consolidated within the MNO network subscribers), we find that TDC has a 42% share, TEL 27%, TLSN 18% and 3 a 10% share. This equates to an HHI of 3,134 – broadly in line with the average for other markets. However, the right way to do this in our view is to split out MVNOs from the market split. This does drive a material impact to HHI. The largest MVNO, DLG Tele, has only 100,000 subscribers, and the other MVNO share in the market is spread over a large number of operators. So although MVNOs account for c8% of the market, the spread over a number of operators drives a lower HHI of 2,578.

• Service revenue market share (Exhibits 6.7 and 6.8): We repeat the above analysis but for service revenues. Including MVNOs, TDC has a 38.5% market share, followed by TEL at 23%, TLSN at 19% and 3 at 15%. The HHI on this metric is very similar to the subscriber HHI at 2,609, which puts it as the third lowest of the eight four-player markets that we examine.

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

2007 2008 2009 2010 2011 2012

TDC Telia Telenor 3 Other


88

Exhibit 6.5: Denmark MNO subscriber HHI at 3,135

Exhibit 6.6: Denmark MNO/MVNO subscriber HHI at 2,578


Exhibit 6.7: Denmark MNO service revenue HHI at 2,883

Exhibit 6.8: Denmark MNO/MVNO service revenue HHI at 2,609


How concentration evolves under various consolidation scenarios, and the implications for competition, are key EC metrics for analysing what remedies to put in place. It is worth mentioning that according to the definition given by the EC, most combinations in the Danish market should qualify as a “concentration” with a “community dimension” – meaning it falls under EC scrutiny rather than that of the national regulator (as shown by the criteria laid out in Exhibit 6.9). The exception to this rule is when TDC is an acquirer, as it has 83% of its EU turnover in one member state.

Exhibit 6.9: EC criteria for community dimension shows all consolidation scenarios should fall under the EC mandate given acquirer’s position

Source: Berenberg (i) combined worldwide turnover is more than €5bn and each of at least two of the merging parties realised more than €250m turnover in the EU, or (ii) combined worldwide turnover is more than €2.5bn; their combined turnover is more than €100m in each of at least 3 member states; in each of those 3 member states, the turnover of each of at least two of the merging parties is more than €25m; the community-wide turnover of each of at least two of the merging parties is more than €100m (iii) unless each of the merging parties obtains more than two-thirds of its EU turnover in one member state.

TDC, 1,957

Telenor, 736

Telia, 336 Three, 106

TDC, 1,559 Telenor, 591

Telia, 311

Three, 106 DLG Tele, 1

TDC, 1,679 Telenor, 597

Telia, 376

Three, 231

TDC, 1,485

Telenor, 528

Telia, 362


Worldwide turnover (€) EU turnover (€) Danish turnover (€)

Turnover

>€100m in 3 EC

states (i) (ii) (iii)

TeliaSonera 12,248 8,654 650 YES YES YES NO

TDC 3,505 3,505 2,911 YES NO YES YES

Telenor 13,366 3,507 738 YES YES YES NO

Hutchison (telco only for EU) 23,801 5,778 316 YES YES YES NO


89

When we look at the various consolidation options, we conclude that 3 is the asset that would most likely have to form part of the consolidation, were it to happen.

• TDC and TEL or TDC and TLSN: We do not view either of these two scenarios as credible. TDC and TEL would result in a 64% subscriber share and a subscriber HHI of 4,497. A TDC and TLSN tie-up would result in 57% subscriber and service revenue share.

• TDC and 3 (Exhibits 6.10 and 6.11): This would result in a subscriber market share of 50% (with MVNOs separated out) and a service revenue market share of 54%. However, with TEL and TLSN operating a JV (including spectrum), this would leave Denmark with only two networks. We doubt that this would be seen as a suitable outcome.

• TLSN and TEL (Exhibits 6.12 and 6.13): With TEL and TLSN operating a full and extensive network JV, this deal would come with low execution risk. But given that the JV is already realising network synergies, incremental synergies would be limited to head office, procurement and distribution. The impact on HHI is not too dissimilar to the TDC and 3 tie-up. It would effectively leave the two largest operators in the market with about a 40% share each (TLSN and TEL would have 42%), 3 would have a 10% share and MVNOs would make up the difference. It is hard to make a case that this would not have an impact on competitive intensity. In addition, we believe that TLSN and TEL opted for a network JV rather than the full merger because it would not have expected a full merger to be approved.

• TLSN and 3 (Exhibits 6.14 and 6.15): From an HHI perspective, this is the deal that would attract the least regulatory scrutiny, as the subscriber/service revenue HHI of 2,941/3,187 does not look prohibitively bad from a European context, when compared with HHI of other European post-consolidation scenarios. It would leave the combined business as a new number two, with a 34% service revenue share and a 28% subscriber market share.

However, we see a number of issues to this potential deal. Firstly, TLSN is currently benefiting from network synergies from the JV with TEL. So doing this deal would mean not only that it would have to forgo those synergies, but it would also have to incur the costs of a break-up. Secondly, it would have to negotiate with TEL, and face regulatory approval for what happens to the 800MHz spectrum. If it could not come to an agreement on this, it would arguably be doing a deal to move to a lower quality network. Finally, recent pricing changes and the current pricing points suggest that TLSN and 3 are competing for the same segments of the market – which could prove an issue requiring subscriber divestments for a deal being evaluated. In conclusion, we see all of these as good enough reasons for the deal to not happen.

• TEL and 3 (Exhibits 6.16 and 6.17): This would suggest all the same issues as per a deal between TLSN and 3 as discussed above, and it also comes with a modestly worse impact on HHI calculations. Such a scenario would effectively leave the market with a very strong number two player, comparable to TDC in scale. The issue from a subscriber perspective is that CBB (TEL’s low-end brand) and Oister (3’s low-end brand) are in direct competition with one another. Most likely, the combination would divest the Oister brand. In our view, though, with headwinds to the deal from the current JV set-up, we see the potential for a transaction as unlikely.


90

Exhibit 6.10: Denmark MNO/MVNO subscriber HHI for TDC and 3 at 3,391

Exhibit 6.11: Denmark MNO/MVNO service revenue HHI for TDC and 3 at 3,781


Exhibit 6.12: Denmark MNO/MVNO subscriber HHI for TLSN and TEL at 3,435

Exhibit 6.13: Denmark MNO/MVNO service revenue HHI for TLSN and TEL at 3,483


Exhibit 6.14: Denmark MNO/MVNO subscriber HHI for TLSN and 3 at 2,941

Exhibit 6.15: Denmark MNO/MVNO service revenue HHI for TLSN and 3 at 3,187


TDC, 2,478

Telenor, 591

Telia, 311 DLG Tele, 1

TDC, 2,888

Telenor, 528


TDC, 1,559

Telenor, 1,758


TDC, 1,485

Telenor, 1,764


TDC, 1,559

Telenor, 591

Telia, 780

DLG Tele, 1

TDC, 1,485

Telenor, 528

Telia, 1,172

DLG Tele, 0


91

Exhibit 6.16: Denmark MNO/MVNO subscriber HHI for TEL and 3 at 3,079

Exhibit 6.17: Denmark MNO/MVNO service revenue HHI for TEL and 3 at 3,308


We should also not discount the possibility of 3 being an acquirer, although this would do little to address any of the above issues. TEL is unlikely to be a seller, although TLSN could be. However for TLSN to be a seller, it would probably need to break certain agreements with TEL for its network JV, which could kill the possibility of any potential deal. Also, if 3 could not get access to TLSN’s tower infrastructure, given that it is now in a JV, we struggle to see how the deal could happen in reality.

In Exhibit 6.18 below, we summarise the impact on HHI.

Exhibit 6.18: A combination of TLSN and 3 has the smallest impact on HHI

Source: Berenberg

Low prices and big bundles

Denmark is a mobile market that has been characterised by aggressive competition for a number of years now. Although not a constant, sporadic price wars have resulted in a step-down in pricing that usually becomes hard to reverse. We had seen the first period of stability for a number of years in the second half of 2012; however with 3 and TLSN’s new pricing points in Q1, we can expect another permanent step-down in pricing – as the market moves to “all-you-can-eat” voice/ SMS.

Historical context: The Danish bundles evolved with lots of data (which was typically not used) given away in the bundle, and with unlimited SMS also given away – so what consumers were really paying for was voice. However, with the recent 3/TLSN tariff moves, this has changed. There are a number of reasons why

TDC, 1,559

Telenor, 1,197


TDC, 1,485

Telenor, 1,459


HHI

Subscriber

market share

(MVNOs

separated out) Change

Service revenue

market share

(MVNOs


Current 2,578 2,609

TDC + Three 3,391 813 3,781 1,172

Telia + Three 2,941 363 3,187 579

Telenor + Three 3,079 501 3,308 699

Telenor + Telia 3,435 857 3,483 874

TDC + Telia 3,970 1,392 4,074 1,466

TDC + Telenor 4,497 1,919 4,379 1,771


92

we can describe Denmark as a particularly aggressive mobile market.

1. The first is that competition is spread over a number of different brands (as shown in Exhibit 6.19). Although many of these brands are owned by the MNOs, there has been a constant migration to the lower-end, no-frills brands, which today account for more than half of the TDC subscriber base. So a focus on low-end brands has leant itself to increased competitive intensity.

2. A second reason for the higher competitive intensity in Denmark is that no contract can be longer in duration than six months. This has been the case for many years and means that not only is the Danish market a structurally higher-churn market but also a market where the consumer is used to shopping around for the best deal on a regular basis.

3. A third reason is the strong MVNO presence in the market – which has partly come about as a result of the six-month contract limits. This has resulted in MVNOs Telmore and Onfone gaining rapid share in the past, only to be bought out by TDC.

4. Finally, Denmark has moved away from handsets subsidies (although TEL is moving back to them). This increases the transparency of offers, allowing subscribers to buy a handset and a contract and then move to a cheaper offer six months later.

While pricing in Sweden has moved lower, TDC has not followed, especially when it has come to its main TDC brand. Our key concern with TDC’s positioning in the Danish market is that in not following the competitive pricing points lower, TDC’s pricing is at a greater premium to its recent historical levels. TLSN and 3 are now in the world of unlimited voice, while TEL is offering this through a special promotion. As we show in Exhibit 16.20, in which we highlight the recent pricing history in Denmark, the TDC brand is now at a three-year high in terms of its pricing differential against peers (since we have been monitoring it), while the differential in voice minutes offered is also at an all-time high.

Exhibit 6.19: Brand subscriber split in H2 2012

Source: Erhvervsstyrelsen (Danish regulator)

TDC27%

Telmore8%

Onfone3%

M12%

Fullrate0%

YouSee0%Telenor

17%

CBB Mobil [3]8%

Telia16%

DLG Tele1%

Hi3G10%

Others8%

TDC

Telenor

Telia

Hi3G

Others


93

Exhibit 6.20: Pricing history in the Danish market

Source: Berenberg estimates (note we assume 2,000 minutes for an unlimited package in the voice discount calculation)

• At the low end (all packages have unlimited SMS), TDC costs DKK229/month for 120 minutes and 3GB of data. In contrast, with TLSN customers receive 180 minutes and 1GB for DKK129, and with 3 it costs DKK159 for unlimited voice and 1GB. Telmore is the most competitive TDC low-end brand, costing DKK99 for 180 minutes and 500MB.

• As can be seen in Exhibit 6.20, the TDC brand screens as very expensive to peers at the mid-end. At the high end, it is the same story, given that it will cost DKK499 with TDC to replicate the TEL DKK299 offers and the 3 DKK199 offer.

Clearly the TDC brand has a pricing issue; however, we think the problem is deeper than that. The only parts of the TDC subscriber base that are growing are the Telmore and Onfone brands, which have benefited from strong traction and aggressive pricing. However, when we look at current pricing, the mainstay 3 and TLSN brands are more attractive. So, in our view, the current pricing weakness not only indicates continued subscriber losses at TDC, but also slower growth at Onfone/Telmore. In addition, with some pricing response coming from both brands, the migration away from the main TDC brand is only likely to accelerate.

TDC brand Jan-11 Jun-11 Jan-12 Jun-12 Jan-13 March Current

Minutes 240 480 480 240 360 360 360

SMS Unlimited Unlimited Unlimited Unlimited Unlimited Unlimited Unlimited

Data 5GB 10GB 10GB 5GB 5GB 5GB 5GB

Price 399 349 349 299 299 299 299

Telenor brand Jan-11 Jun-11 Jan-12 Jun-12 Jan-13 March Current

Minutes 480 600 600 600 600 Unlimited Unlimited


Data 1GB 3GB 3GB 3GB 3GB 10 GB 10 GB

Price 399 280 280 280 280 299 299

Telia brand Jan-11 Jun-11 Jan-12 Jun-12 Jan-13 March Current




Price 399 270 270 199 199 179 199

3 brand Jan-11 Jun-11 Jan-12 Jun-12 Jan-13 March Current




Price 399 399 179 149 149 179 179

Onfone brand Jan-11 Jun-11 Jan-12 Jun-12 Jan-13 March Current

Minutes 900 900 900 660 660 660 600


Data 1GB 1GB 1GB 1GB 1GB 1GB 500MB

Price 249 208 249 199 199 199 149

TDC price premium to average 10.4% 20.7% 42.7% 44.6% 44.6% 39.7% 44.8%

TDC minutes discount -62.8% -13.5% -11.1% -51.5% -27.3% -78.4% -78.2%

TDC data discount -20.0% 110.5% 185.7% 42.9% 42.9% 17.6% 17.6%


94

Exhibit 6.21: Cost of an iPhone 5 16GB for high-end package (€)

Source: Berenberg

The pricing problem is even more serious than we had originally thought. In Exhibit 6.21, we show how much it costs to buy an iPhone 5 16GB with each operator. This demonstrates that TLSN and TEL have started subsidising handsets again, and surprisingly so has TDC on its main brand for the very high end. Therefore, not only are the likes of TLSN/TEL cheaper than TDC for the SIM, but buying a phone through them is also more attractive, unless for the most expensive TDC package. This is pertinent to TDC, given that it realised cDKK200m of cost savings from lower marketing and SAC/SRC costs in 2012. The risk is that competitive pressures could see some of these benefits reverse.

In Exhibit 6.22, we show the pricing for 3 user profiles in Denmark:

1) a user wanting 100 minutes, 500 SMS and 200MB of data;

2) a user wanting unlimited voice/SMS and 1GB of data;

3) the same as 2) but with an iPhone 5 16GB.

The results are particularly interesting.

1) At the low end, pricing is relatively flat for the online brands at about DKK100, and TEL is in line with that level. To replicate the equivalent profile on TDC would cost double that amount, however. Relative to the rest of Europe, pricing at this end of the market screens favourably.

2) There is a very wide variance of pricing for this second profile, especially with 3 and TLSN moving to all-you-can-eat voice and SMS across most of its tariff propositions. As a result, this exaggerates the value gap between TDC and its peers, especially when we look at the main brands.

The cheapest package for this user profile is the DKK159 package available with 3. Although the DKK299 TEL special offer is more expensive, it does come with unlimited data. The outlier is TDC, which thus far has the highest pricing point we have seen in Europe to replicate this package, at DKK499 (= EUR67) – although it comes with a significant amount of data at GB20. A more comparable TDC package for 12 hours of voice, unlimited SMS and 10GB of data still costs DKK349 or EUR47 – still expensive in a European context.

0

100

200

300

400

500

600

700

800

TDC Telenor Telia 3 Onfone Telmore CBB Oister


95

Exhibit 6.22: Pricing analysis of the Swedish market (prices in DKK)

Source: Berenberg

3) The inclusion of an iPhone 5 16GB does not change the picture. In fact, with TEL and TLSN subsidising the iPhone 5, the picture is even exaggerated. TDC’s pricing looks to be in a different league.

In conclusion, the Danish market remains very competitive, with one of the widest pricing ranges we have seen in Europe. This only goes to highlight the backbook issues at TDC, which in our view will manifest themselves first in higher subscriber losses, and second in potential pricing cuts. There have been many false dawns in Denmark in terms of improving the competitive environment, and it seems the stability we saw in the second half of 2012 is just another example of that.

The MVNO threat in Denmark

According to the regulatory data, as at the end of 2012 we estimate the total MVNO share of the Danish mobile market to have been 8.3%, with the largest MVNO DLG Tele having 110,000 subscribers, equivalent to a 1.2% market share. The Danish mobile market remains heavily fragmented and MVNOs continue to appear, with Nettalk (representing supermarket Netto) and Coop Mobile contributing to additions for TDC.

100 minutes + 500 SMS +200MB TDC Telenor Telia 3 Onfone Telmore CBB Oister Fullrate

Changes

Brand name Mobilpakke Fri tale 4Fun 3FriSpeech 1GB 2 timer 3 timer 11 timer 3 timer 5 timer

Voice calls (in min) 100 100 100 Unlimited 100 180 180 180 300

pecularity

Free TDC to TDC

calls // free music

with TDC play

Unlimited - - - - Free Oister to Oister

Price/ extra min 0.75 0.75 0.70 0.75 0.59 0.59 0.60 0.59

Call charges 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Data 500MB 300MB 1GB 1GB 300MB 500MB 1GB 3GB 200MB

Price/1 extra GB na na na na

SMS free SMS & MMS 300SMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS

Subscription terms 6 months 6 months 1 month 1 month 6 months 1 month 1 month 1 month 1 month

Monthly Flat rate tariff 204 102.5 169 159 129 99 99 99 99

Unlimited voice / SMS + 1GB TDC Telenor Telia 3 Onfone Telmore CBB Oister Fullrate

Changes

Brand name Mobilpakke Fri tale

4EVERYTHING

LIGHT 3FriSpeech 1GB FriTale FriTale FriTale Free Unlimited

Voice calls (in min) Unlimiited Unlimited Unlimited Unlimited Unlimited Unlimited unlimited Unlimited Unlimited

pecularity -

Free music / free

facebook -

Free to other CBB

customers

Price/ extra min 0.75 0.75 0.79 0.50 0.59 0.60 0.60 0.59

Call charges 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Data 20GB unlimited 1GB 1GB 3GB 3GB 4GB 6GB 1GB

Price/1 extra GB na na na na

SMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS

Subscription terms 6 months 6 months 1 month 1 month 1 month 1 month 1 month 1 month 1 month

Monthly Flat rate tariff 499 299 179 159 249 249 199 179 169

Unlimited voice / SMS + 1GB with iphone 16GBTDC Telenor Telia 3 Onfone Telmore CBB Oister Fullrate

Changes Price cut

Brand name Mobilpakke Fri tale

4EVERYTHING

LIGHT 3FriSpeech 1GB FriTale FriTale FriTale Free

Voice calls (in min) Unlimiited Unlimited Unlimited Unlimited Unlimited Unlimited unlimited Unlimited

pecularity -

Free music / free

facebook -

Free to other CBB

customers

Price/ extra min 0.75 0.75 0.79 0.50 0.59 0.60 0.60

Call charges 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Data 20GB unlimited 1GB 1GB 3GB 3GB 4GB 6GB

Price/1 extra GB na na na

SMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS free SMS & MMS

Subscription terms 6 months 6 months 1 month 1 month 1 month 1 month 1 month 1 month

Monthly Flat rate tariff 499 299 179 159 249 249 199 179

Upfront 3560 5161 5299

Creating 199 0 99 364 99

Monthly (24 months) 155 165 208

Monthly (12 months) 416 392

Monthly equivalent (24 months) 656 454 348 374 472 461 395 400


96

Exhibit 6.23: MVNO/distribution summary in Sweden

Source: Berenberg

Distribution and choice

Online distribution plays a much more important role in Denmark than in other European markets, given that the main brands account for 70% of the market, while online brands predominantly account for the remainder. In Exhibit 16.23, we provide a summary of distribution in the Danish market. TDC is the only operator that has any real presence with third-party distributors, while TDC and TEL dominate the MVNO and service provider market.

3

52 storesDirect

n/aNo

MVNOs

Only sold through own

stores and the Fonachain

Reseller/

Retail

35 storesDirect

TeliaSonera

Small MVNO and is only a minor player

1MVNO

No external resellersReseller/

Retail

TDC

Telenor

DLG and Lycamobile

23 MVNOs /

SPs

Elgiganten, Atea, TDC business centre

Reseller/Retail

52 TDC shopsDirect

Lebara, Firmafon and Ipinion are the largest

21 MVNOs & SPs

No external resellers, only telemarketing

Reseller/Retail

100 storesDirect

MNOs MVNOs, Resellers,

C

u

s

t

o

m

e

r

s


97

Pre-paid versus post-paid mix

At the end of 2009, post-paid accounted for 79.9% of market subscribers. By Q1 2013, this has increased to 85.4%. We estimate that post-paid accounted for 92% of service revenues at end-Q1 2013, compared with closer to 89% in 2009.

Rapid post-paid growth in Sweden has been driven by the strong uptake in smartphones, with some commentators estimating smartphone penetration in Sweden to be over 80%. Post-paid continues to take share within the Danish market, as subscribers move to bundles and those bundles start to contain more and more usage. In addition, bundle pricing has moved down so rapidly, that the relative value versus pre-paid continues to look more and more attractive.

The high post-paid share means that any EC analysis of market concentration would be heavily focused on the post-paid segment.


In Exhibit 6.24, we show how spectrum is split between operators and the JVs through which they operate. When we look at the spectrum HHI (2,570 – but 3,209 for the less than 1GHz spectrum), it is close to the subscribers’ share and service revenue HHI at c2,600. However, when we look at the market shares, we see that 3 is left with excess spectrum relative to its market share, while TLSN is left with too little.

However, this analysis is based on absolute spectrum amounts, which fails to take account of differing spectral band quality. So we calculate adjusted spectrum holding based on 1MHz of 800MHz/900MHz spectrum being able to cover1:

• 2x the area of 1800MHz spectrum;

• 2.5x the area of 2,100MHz spectrum; and

• 9x the area of 2,500MHz spectrum.

Exhibit 6.24: Danish spectrum split

Source: Berenberg

Exhibit 6.25: Total spectrum share

Source: Berenberg * assumes JVs split spectrum equally among owners

1 Based on data provided by Ofcom

FDD Spectrum TOTAL Share TOTAL <2GHz Share TOTAL <1GHz Share

TDC 172 31.9% 102 36.5% 58 45.0%

Telenor/Telia 286 53.2% 146 52.7% 61 47.2%

3 80 14.9% 30 10.8% 10 7.8%

TOTAL 538 100.0% 278 100.0% 129 100.0%

TOTAL Spectrum TOTAL Share Spectrum HHI Subscriber market share

TDC 177 29.3% 858 44.2%

Telenor 151 25.1% 630 27.1%

Telia 165 27.4% 749 18.3%

3 110 18.2% 333 10.3%

TOTAL 603 100.0% 2,570 100.0%


98

Exhibit 6.26: Danish spectrum split with spectrum amounts adjusted for coverage


The spectrum adjustments calculation is particularly relevant to Denmark, where 3 lacks much spectrum below 1GHz (it only has 8% of that spectrum) but has high amounts of high frequency FDD and TDD spectrum. So if we adjust the spectrum holdings for this as we do in Exhibit 6.26, we can see that 3’s share of spectrum holding is only marginally higher than its current market share at 10.3%. However, what is most stunning about this result is that when we analyse the data traffic provided by the regulator in Denmark, it shows that although it only operates with 14% of spectrum in Denmark on an adjusted basis, it has 37% of the data traffic. In theory, all else being equal, 3 is likely to be running closer to capacity limits than any of the other operators in Denmark, arguably increasing the appetite for M&A.

Implications for consolidation: Looking at the implications of a TLSN/3 tie-up, we see it as quite likely that spectrum would have to be handed back to the regulator, given that the combination would have 46% of the spectrum in the market (41% on an adjusted basis) and a 29% subscriber market share (counting MVNOs on the network).

TLSN and 3: In carrying out this analysis, we assumed that the deal would drive a split in two of the 800MHz spectrum that TLSN holds through its JV with TEL. The result of this is shown in Exhibit 6.27 for unadjusted spectrum and Exhibit 6.28 for adjusted spectrum. The spectrum HHI left over is one of the highest from a consolidation scenario in Europe, although this looks more average when we look at adjusted spectrum HHI.

Exhibit 6.27: Danish spectrum split – assumes TLSN acquires 3 Sweden


Exhibit 6.28: Danish spectrum split with spectrum amounts adjusted for coverage – assumes TLSN acquires 3 Denmark


TOTAL adjusted

Spectrum TOTAL Share HHI Subscriber market share

Total data used in H2

12 (GB)

Data market

share

TDC 97 35.3% 1,244 44.2% 3,956,685 17.0%

Telenor 66 23.8% 568 27.1% 4,586,002 19.7%

Telia 75 27.2% 738 18.3% 6,200,966 26.6%

3 38 13.7% 189 10.3% 8,591,952 36.8%

TOTAL 275 100.0% 2,738 100.0% 23,335,604 100.0%


TDC 177 29.3% 858 44.2%

Telenor 151 25.1% 630 27.1%

Telia 275 45.6% 2,080 28.6%

3 0 0.0% - 0.0%

TOTAL 603 100.0% 3,568 100.0%

TOTAL adjusted



12 (GB)

Data market

share

TDC 97 35.3% 1,244 44.2% 3,956,685 17.0%

Telenor 66 23.8% 568 27.1% 4,586,002 19.7%

Telia 113 40.9% 1,674 28.6% 14,792,918 63.4%

3 0 0.0% - 0.0% - 0.0%

TOTAL 275 100.0% 3,485 100.0% 23,335,604 100.0%


99

• Subscriber HHI would increase by 363 points to 2,941 as a result of a TLSN and 3 tie-up.

• Service revenue HHI would increase by 579 points to 3,187.

• Spectrum HHI would increase by 998 points to 3,568, which is obviously a far greater increase than seen in the subscriber or service revenue HHI, and would leave the combined business with 46% of the spectrum and only a 29% market share.

• Adjusted spectrum HHI would increase by 747 points to 3,485.

TEL and 3: We repeated the above analysis for a TEL/3 combination, which would come with many of the same issues. The result of this is shown in Exhibit 6.29 for unadjusted spectrum and Exhibit 6.30 for adjusted spectrum. In essence, we would once again expect the combination to have to divest spectrum (this assumes the 800MHz is split between JV partners), although the spectrum divestment is arguably less pressing than it was for a TLSN and 3 tie-up.

• Subscriber HHI increases by 501 points to 3,079 due to TEL and 3 tie-up.

• Service revenue HHI would increase by 699 points to 3,308.

• Spectrum HHI would increase by 916 points to 3,486.

• Adjusted spectrum HHI would increase by 655 points to 3,393.

Given the scale of either deal, spectrum divestments would likely form a key remedy. This would partly depend on how the 800MHz spectrum in the JV would be treated. So, for example, if it was left to the other JV partner, arguably spectrum would not need to be divested. However, the question then would be: why would an operator want to walk away from a JV with a more valuable spectrum portfolio, into a tie-up with an inferior spectrum portfolio? The TLSN/TEL JV in Denmark has 53% of the market spectrum, putting it in an incredibly strong position. Risking that, in an uncertain consolidation scenario does not make a lot of sense.

Exhibit 6.29: Danish spectrum split – assumes TEL acquires 3 Sweden


Exhibit 6.30: Danish spectrum split with spectrum amounts adjusted for coverage – assumes TEL acquires 3 Denmark



TDC 177 29.3% 858 44.2%

Telenor 261 43.3% 1,879 37.4%

Telia 165 27.4% 749 18.3%

3 0 0.0% - 0.0%

TOTAL 603 100.0% 3,486 100.0%

TOTAL adjusted



12 (GB)

Data market

share

TDC 97 35.3% 1,244 44.2% 3,956,685 17.0%

Telenor 103 37.6% 1,411 37.4% 4,586,002 19.7%

Telia 75 27.2% 738 18.3% 14,792,918 63.4%

3 0 0.0% - 0.0% - 0.0%

TOTAL 275 100.0% 3,393 100.0% 8,542,686


100

In addition, we do believe the risk of a new entrant buying up available spectrum exists. Denmark’s geography is not too dissimilar to the Netherlands (urbanised and flat) and although prices are not high, it is a market that operates increasingly through online channels – meaning distribution channels are less a barrier to entry.

Network-sharing

Following on from the positive experiences of Sweden, TLSN and TEL entered into a network-sharing deal in Denmark back in June 2011. The sharing includes:

• RAN – the RAN-sharing agreement comprises sharing of the physical RAN infrastructure (masts and antennas);

• frequency – the sharing agreement also includes all frequencies, such as 2G, 3G and LTE (the way the agreement works is that each company leases all its spectrum to the JV, although the company still owns that spectrum; however the JV owns the recently purchased 800MHz spectrum);

• geography – the agreement covers the entire Danish territory.

The agreement does not include the core network, which is where the different services and customer data are defined, and it also does not include some parts of the transmission capacity. From a regulatory perspective, it is thus important that the two operators remain separate for the purpose of retail and wholesale markets (where TEL was at the time the number one wholesale provider in Denmark).

On 29 February 2012, the DCC approved the deal, noting that the improved network/coverage would be beneficial to consumers. The DCC had raised six anti-competitive issues which are discussed in Exhibit 6.31. This showed its concerns from an anti-competitive perspective were based on:

• the potential for reduced wholesale access;

• the potential for the JV to have excess spectrum relative to market share;

• the potential for collusion to impact retail prices.

In our view, the DCC would not have looked favourably on a full merger between TEL and TLSN’s Danish businesses (even if it is an EC decision). However, the spectrum point is interesting, as the JV now has 53% of the spectrum in the market, and a combined market share of less than 40%, so it is already in a very strong position. However, the EC has set new precedents through its remedies in Austria, and it would be reasonable to expect similar remedies to any deal in Denmark as well.

In addition, the DCC did not look at spectrum as a component of the JV unfavourably, while the EC prefers network-sharing to exclude spectrum. The DCC could therefore arguably be considered as more dove-ish than the EC. In going for a network share rather than full consolidation, TLSN and TEL made sure they dealt with DCC rather than the EC.


101

Exhibit 6.31: DCC concerns with Denmark JV

Source: Berenberg, Danish Competition Commission

Anticompetitive concern Commitment

1Risk of collusion in the Wholesale

market

The parties will accept ALL requests

from wholesale customers - based on

market conditions

2

The tariff structure for recovering JV

costs could change the underlying cost

structure - converting fixed to variable

costs - which couls reduce incentives

to compete for new custtomers

The tariff structure for buying capacity

from the JV will always reflect the

underlying cost structure of the radio

access network

3

The JV could buy excess spectrum that

could significantly exceed competing

operators

The parties can only by spectrum

through the JV, to prevent the ability

to build up of spectrum share

4

Reduction of masts/antennas could

create coverage problems for

competitors that rent the masts

The parties are obliged to sell or let an

antenna site that is surplus to

requirements

5Increased risk of exchange of

commercially strategic information

Memebers of the JV management and

board are restricted in the information

they can share with the JV parties

6Reduces competition on coverage and

technology roll out. No grounds for action on this point


102

French consolidation has potential, but is politically sensitive

Will the market consolidate in the next two years?

Overall, France has some potential for consolidation, with limited risk due to the pricing environment and also the structure of both the mobile and fixed markets. We doubt that any proposed remedies would be sufficient to offset the potential synergies. The central debate would more likely be political in our view.

With the newest MNO, ILD, having only launched its service in January 2012, it is in some ways odd that we are already addressing this question. There have, however, been conflicting political views expressed about the possibility of the French market being consolidated back to three operators.

On the one hand, the head of the competition authority warned against such a thought, saying a return to three players was not advisable. However, there have also been rumblings that the launch of Free and the impact of this on prices have led to significant job losses that could be reversed if we saw the market return to a three-player market. The convergence momentum in France and the fact that the four MNOs are also the four main broadband players would add further layers of complexity as both markets would likely consolidate from four to three players.

Our thinking is that from a spectrum perspective, France would become too concentrated with a merger, and this is the main reason why we highlight Germany and the UK as the best options. However, if we look at France based on the HHI of service revenues, then we see a situation emerging which suggests that the market would remain borderline competitive. This is particularly the case when we incorporate MVNOs into the equation.

Therefore while we would likely see some stiff remedies in terms of ensuring the continued success of the MVNO operators, we think that given the pricing environment in France this could be done with limited impact on the current market. Furthermore, given the momentum that ILD has in terms of winning market share, we expect that by the end of 2013, it is feasible that we could see consolidation in France that would leave the subscriber HHI (and possibly the service revenue HHI) under the ideal 2,800 threshold.

Which combination is most likely to gain approval?

On spectrum split, there is not a lot of difference in terms of a combination of any of the three legacy MNOs with Free, and as we say, this would leave spectrum allocation sitting well above the perceived ideal threshold.

However, when we nail this down to either subscribers or service revenue market share, then the only likely combination would be Bouygues and Free. However, with SFR having lost 5% market share in service revenue since 2007 and 4% over the last three years, there is an argument that SFR’s market share would only revert to where it was previously.

As can be seen, the HHI sits not far from the magic number of 2,800 and we believe that by the end of 2013, it is feasible that this will have fallen below 2,800. This would particularly be the case if we remove M2M SIMs from the equation.


103

Exhibit 7.1: HHI for subscribers (left) and service revenues (right) post-Bouygues/ILD merger


What remedies would be imposed

In fairness, the French consumer is enjoying somewhat of a boon, with prices collapsing on the back of the launch of ILD. We would expect a merger between ILD and Bouygues to face modest remedies. As we say, we think the single biggest issues will be the potential for the return of spectrum, which could be issued to another new entrant and assurances on MVNOs.

However, we think the current pricing regime and the big bundles already on offer in France should help mitigate the risks posed by remedies of this nature, even if they are fairly harsh. Indeed, we would argue that with the synergies being available to the operators that have traditionally been the disruptors in the market and which would still trail handsomely in terms of market share, we suspect that the risk here would sit with the other two MNOs, Orange and SFR.

This may make it worth SFR considering whether it could make an audacious move for ILD, which has ben rumoured. Either way, Orange is out the equation on all fronts, and would therefore be the most at risk in the event that some of the synergy savings were used to invest in winning share.

Market overview

France has four MNOs (Orange, SFR, Bouygues and Free) and a number of MVNOs (the list could include up to 65 additional operators). Despite a significant number of operational MVNOs , the MNOs dominate the market with around an 89% market share, according to the domestic regulator ARCEP. In total, the market has just over 71m subscribers, although within this there are 3.4m dongle customers and 5.3m machine-to-machine SIMs, which lowers the human connections to 62.4m, which means penetration is approaching 100%. There are 56m consumer connections, excluding dongles, implying penetration of 88%.

Bouygues, 608

Orange France, 1,326

SFR, 791

MVNO's, 117

HHI = 2,842

Bouygues, 596


SFR, 881

MVNO's, 81

HHI = 2,920


104

Exhibit 7.2: French mobile subscribers by operator

Source: Company reports

The market in France has changed dramatically over the last two years and continues to evolve following the launch of ILD’s Free Mobile in January 2012. This has had the twin impact of increasing subscriber elasticity in the market, while materially altering the pricing structure. MNO service revenues had reflected the oligopolistic nature of the market through to the middle of 2011, with underlying service revenue growth. However, the launch of ILD has reversed this trend with the market exiting 2012 on a run rate of around 6.4% service revenue declines. This has been despite a doubling of net additions in 2012, compared to the previous two years and an acceleration in the move to post-paid from pre-paid.

In terms of the total service revenues in the market, these peaked in 2010, when it is clear the three incumbent MNOs began pricing for the launch of ILD.

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Iliad 0.000 5.205

SFR 8.070 10.160 12.555 13.547 14.724 15.820 17.198 17.883 18.766 19.652 20.395 21.303 21.463 20.690

Orange France 10.926 14.311 17.823 19.216 20.329 21.241 22.430 23.268 24.226 25.202 26.334 26.929 27.090 27.190

Bouygues 3.650 5.190 6.467 5.823 6.630 7.468 8.131 8.722 9.256 9.594 10.352 11.084 11.304 11.251

0.000

10.000

20.000

30.000

40.000

50.000

60.000

70.000


105

Exhibit 7.3: French service revenue by operator (EUR 000’s)


Exhibit 7.4: French service revenue total


-1000

1000

3000

5000

7000

9000

11000

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Bouygues Orange France SFR Iliad

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Iliad 0 720

SFR 5,232 5,525 6,270 6,431 6,808 7,479 7,935 8,391 8,576 8,510 8,420 7,885 7,006

Orange France 5,190 6,215 7,082 7,371 7,933 9,131 9,238 9,320 9,714 10,037 9,958 9,832 9,305

Bouygues 2,046 2,905 3,028 3,050 3,449 3,827 4,241 4,464 4,696 4,863 5,060 5,082 4,631

0

5000

10000

15000

20000

25000


106

Exhibit 7.5: France HHI with and without MVNOs, subscribers

Source: Berenberg estimates, company reports, ARCEP

In terms of market concentration, the French HHI based on subscriber numbers, excluding MVNOs, sits at just over 3,100, versus the 2,800 score that implies a sufficiently concentrated market. This is based upon the market share of subscribers. However, once we adjust for MVNOs, then the HHI falls to 2,555, well within the 2,800 threshold. As there is little detail on the breakdown of the MVNO customers in France, we have allocated them among the three legacy MNOs in accordance with there overall market shares.

If we change the parameter to an HHI based on the market share of service revenues, we see a worsening in the metric, as despite having quickly won subscriber share, ILD’s share of service revenues is significantly lagging. It lifts the HHI to 3,261, taking it well above the 2,800 threshold, although a merger between Bouygues and Free would have very little impact – to just 27 points higher at 3,470.

It is harder to factor the HHI for the service revenues, which is obfuscated by the fact that we do not know the roaming revenue received by FTE from ILD (we have assumed €120m in the quarter), and there is no data available about the market share of the MVNOs. We have assumed that it is 9%, below the 10.8% share of subscribers. Again, under this scenario, the competitiveness of the market stands up to scrutiny, suggesting an HHI of 2,722. While this leaves France sitting towards the upper end of the HHI spectrum in Europe, the rapid growth of ILD is likely to significantly lower the HHI over the next two years.

Exhibit 7.6: France service revenues’ HHI, with and without MVNOs

I

Source: Berenberg estimates, company reports, ARCEP

Bouygues, 302


SFR, 1,019

Iliad, 88

HHI = 3,118

Bouygues, 234


SFR, 791

Iliad, 88

MVNO's, 117

HHI = 2,555

Bouygues, 453


SFR, 1,075

Iliad, 26

HHI = 3,218

Bouygues, 371


SFR, 881

Iliad, 26

MVNO's, 81

HHI = 2,722


107

Low prices and big bundles France is a perfect example of how a market could change through the introduction of tough remedies. In the case of France, however, it was driven by the launch of the maverick operator ILD, which acquired the fourth mobile licence. The launch of ILD has heralded a new dawn of big bundles in France, with prices now among the lowest in Europe and the legacy MNOs suffering double-digit declines in ARPU.

The declines are set to continue through 2013 and into 2014 as legacy contracts unwind and subscribers shift to more attractively priced bundles. However, we have seen a significant narrowing in the pricing discrepancy between the legacy MNOs and ILD, which will leave the medium-term evolution for the operators to be driven by subscriber momentum, or of course further price cuts.

MVNOs are not a huge factor in France. They have a combined market share of around 10%, and after the negative initial jerk reaction (like the rest of the market) from the launch of ILD, this has remained fairly stable. The larger MVNO players include Virgin Mobile and NRJ. While a modest component of the market, the importance of the MVNOs in France should not be underestimated as they do bring the HHI under the 2,800 threshold that deems a market sufficiently competitive.

Exhibit 7.73: ARPU per month by operator (EUR)


1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Bouygues 41.3 41.3 41.3 41.3 41.3 41.3 41.3 41.3 41.3 41.5 40.6 38.4 35.9 32.4

Orange France 38.2 35.5 32.3 31.4 31.8 33.1 34.5 34.0 33.5 33.5 32.5 31.8 30.9 27.7

SFR 53.0 41.6 39.3 41.0 36.6 38.9 37.4 37.9 36.7 35.7 34.8 34.2 31.5 28.7

Iliad 16.6

Average 44.2 39.5 37.6 37.9 36.6 37.8 37.7 37.7 37.2 36.9 36.0 34.8 32.8 26.3

20.0

24.3

28.6

33.0

37.3

41.6

45.9

50.2


108

Exhibit 7.74: Details on French mobile distribution



France is predominantly a post-paid market, with around 80% of subscribers taking a post-paid service. This is in line with the share of gross adds, which means that given the lower churn, the proportion of post-paid customers looks set to continue to increase. The figures are a little skewed as ILD have contracts which are essentially rolling one-month contracts and even those on the €2 offer are treated as contract subscribers. With the emergence of low-cost bundles starting at less than €5 per month, we see further support for the increasing penetration of contract customers in the base.


109

Exhibit 7.8: Post-paid share of base by operator



In this segment, we consider the balance of the market using customers and service revenues and then look at how this compares to the spectrum allocation.

Exhibit 7.9: Market split by subscribers, service revenues and spectrum

Source: Company reports; ARCEP

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Bouygues 46.40% 62.00% 64.44% 66.13% 68.42% 71.63% 73.10% 75.22% 76.56% 79.10% 80.60% 83.80%

Orange France 52.99% 55.59% 57.86% 60.62% 61.77% 63.24% 64.80% 67.36% 68.14% 70.50% 71.81% 72.47%

SFR 50.49% 53.05% 57.74% 60.69% 63.26% 64.97% 65.51% 69.11% 72.60% 75.55% 77.18% 80.05%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

Orange SFR Bouygues Free

Subs Serv Revs Spectrum Spectrum (Adj)


110

We have previously highlighted that once the impact of MVNOs is taken into account, the French market is deemed to be suitably competitive, and that we expect this to amplify as ILD grows further. This is also the case for spectrum, although once we adjust the spectrum allocation according to the attractiveness of the different frequencies then it could be argued that the French market is still a little too skewed to the main three players.

Exhibit 7.105: HHI for spectrum allocation and adjusted

Source: Berenberg estimates, company reports

In terms of how the operational performance ties in with the spectrum allocation, Orange is clearly outperforming while Bouygues is under-achieving its spectrum allocation. We would argue that it will be Bouygues that could further destabilise the market as it seeks to use its spectrum position to push cheaper services. We would argue that ILD’s ambition to deliver a market share of 25% from this spectrum position looks difficult to achieve without much higher capex than the company is currently suggesting.

Moreover, in ILD’s case, it is clear that the roaming agreement with Orange will fall away over time in line with the guidance from the regulator, and this will exacerbate the position given that at the moment quality issues may be being masked.

Orange, 969.4

SFR, 864.0

Bouygues,

817.8

Free, 118.2

HHI 2,769

Orange, 989.8

SFR, 968.0

Bouygues,

940.8

Free, 45.6

HHI 2,944


111

Germany: relatively well placed for remedies


There has been much talk and activity related to the potential for O2 and E+ to merge in the German market. The volume of equity market excitement ebbs and flows, and the focus switches between a potential merger and a potential comprehensive network-sharing arrangement. At the very least, it appears that two of the operators in the market might eventually do something together.

Beyond this, the scope for others to consolidate would seem limited given the impact on market concentration and spectrum concentration. The German market is one of the best performing of European mobile markets (and among our focus set of four-player-plus markets), so perhaps the pressure for consolidation is more limited in Germany than in other markets, where service revenues are declining much more steeply.

There may also be a strategic dimension to the lack of progress on merger activity in the German market. Shortly after the breakdown of last year’s merger attempt between O2 and E+, the O2 CEO was quoted on Bloomberg citing the relatively weak strategic position of E+ given its spectrum portfolio, which lacks 800Mhz 4G spectrum. One might ask why O2, if it feels that E+ is likely to struggle as the market shifts towards 4G, would consider helping out E+ by entering into either a merger arrangement or a comprehensive network-sharing arrangement. Why not await de facto consolidation of the market with the eventual exit of E+ as the disadvantage of its weak spectrum portfolio starts to take its toll?


Having said that, the combination of O2 and E+ results in a relatively modest increase in market concentration that still leaves the German market less concentrated, in service revenue terms, than the average of the peer markets in this report.

This advantageous position could be undermined, however, by the impact of an O2/E+ combination on spectrum concentration in the German market. With this combination, Germany would move from being the least concentrated of our peer group markets to be the most concentrated, with its adjusted spectrum HHI increasing from 2,529 to 3,595.

What remedies would be imposed?

Remedies in the event of consolidation would almost certainly include the divestment of significant quantities of 1,800Mhz and possibly some 2,100Mhz spectrum. This could present an opening for a new entrant to enter a market that might actually seem a relatively attractive prospect given the above-average level of prices. Antitrust authorities might also consider enforcing mobile network access agreements to facilitate a new entrant, given the relative lack of comprehensive network-sharing or access arrangements in the German mobile market.

On the other hand, Germany would seem relatively well positioned to avoid remedies to support MVNOs, given its already well developed MVNO sub-sector.


112

Market overview – who is winning, who is losing

An uneasy stability appears to have settled over the German mobile market through late 2012 and early 2013. The price war fears of 2012 have so far proven misplaced. Having reached a service revenue peak in 2005, and then having declined for a few years to 2009, the market now appears to have sustained a broadly stable size for the last few years, with a moderate degree of annual variance explained by the regulators bi-annual approach to MTR reductions.

The long-term retrospective shows a significant change in market shares, with O2 and E+ between them gaining over 14ppt of service revenue share since 2000 (see Exhibit 8.1). TEF’s O2 has been the biggest winner, gaining 11.8ppt of service revenue share since 2000, at the expense of DTE’s T-Mobile, which has lost 7.7ppt of share, and VOD, which has lost 6.6ppt of share. Despite the share gain, both O2 and E+ generate only about half the level of service revenues that T-Mobile and VOD do.

Exhibit 8.1: German mobile market revenues peaked in 2005 but have stabilised in recent years


Exhibit 8.2: German mobile subscriber base growth has slowed


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

O2 586 1,055 1,404 1,842 2,370 2,816 2,809 2,901 2,808 2,860 2,932 2,947 3,152

E-Plus 1,840 1,855 2,074 2,224 2,501 2,461 2,698 2,816 3,005 3,021 3,092 3,099 3,149

Vodafone D2 5,356 7,116 6,635 7,313 7,848 7,925 7,699 6,990 6,820 6,505 6,524 6,408 6,472

T-Mobile 5,719 6,253 6,778 7,377 7,656 7,758 7,435 7,156 7,045 7,008 7,047 6,881 6,779

0

5,000

10,000

15,000

20,000

25,000

Serv

ice R

ev

en

ues

€m

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

O2 2.75 3.66 4.58 5.59 7.40 9.77 11.02 12.47 14.20 15.51 17.05 18.38 19.30

E-Plus 6.66 7.48 7.27 8.21 9.51 10.75 12.65 14.81 17.78 18.99 20.43 22.72 23.40

Vodafone D2 19.39 21.89 22.73 24.67 26.94 29.17 30.62 33.92 36.17 34.63 36.68 37.63 33.89

T-Mobile 19.12 23.08 24.58 26.33 27.47 29.52 31.40 35.95 39.10 39.14 34.69 35.40 36.57

0.00

20.00

40.00

60.00

80.00

100.00

120.00

Su

bsc

rib

ers

(m

ln)


113

In recent quarters, both T-Mobile and VOD appear to have regained the initiative. We think T-Mobile regained service revenue share through Q3 2012-Q1 2013, with VOD also improving share into Q1 2013. E+’s service revenue share ended Q1 2013 50bp below the level of Q1 2012 at 15.5%, while O2 was broadly stable versus Q1 2012 at 15.9%.

German market revenue trends have been aided, relative to their main European peers, by a slower pace of MTR rate reductions, which have tended to take place every two years as opposed to annual reductions elsewhere in Europe. We expect German MTRs to end 2013 at a rate of 1.79c/min following a small reduction from the current 1.85c/min on 1 December 2013. In contrast, most of Germany’s peer markets will be at close to or below 1c/min by end-2013 (eg the UK at 0.79c/min, France at 0.8c/min, Spain at 1.09c/min, and Italy at 0.98c/min).

German mobile market revenues have also benefited from the more robust economic backdrop compared to markets like Spain and Italy, where revenues have been declining in mid-single digits.

Exhibit 8.3: German mobile service revenue share has gone to O2 and E+, albeit T-Mobile and VOD have fought back in recent quarters


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

O2 4.3% 6.5% 8.3% 9.8% 11.6% 13.4% 13.6% 14.6% 14.3% 14.7% 15.0% 15.2% 16.1%

E-Plus 13.6% 11.4% 12.3% 11.9% 12.3% 11.7% 13.1% 14.2% 15.3% 15.6% 15.8% 16.0% 16.1%

Vodafone D2 39.7% 43.7% 39.3% 39.0% 38.5% 37.8% 37.3% 35.2% 34.7% 33.5% 33.3% 33.1% 33.1%

T-Mobile 42.4% 38.4% 40.1% 39.3% 37.6% 37.0% 36.0% 36.0% 35.8% 36.1% 36.0% 35.6% 34.7%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Serv

ice R

ev

en

ue S

hare


114

Exhibit 8.4: All four German network operators have seen service revenues stabilise


Judgement: The German mobile market is one of the better performing of our focus markets, and as such the opportunity cost of not seeing consolidation is relatively low. Other, weaker performing markets have more to lose without seeing in-market consolidation succeed.

Market concentration, MVNO presence and consumer choice

Germany appears to have one of the best developed MVNO segments of the peer group. Public data on MVNOs is hard to come by, but we have collated data from various sources to suggest that MVNOs in Germany account for about a 19% share of subscribers and a 15% share of service revenues. Although network-owned MVNOs (eg T-Mobile’s Congstar, VOD’s o.tel.o, E+’s Base and Yourphone, and O2’s Fonic – see Exhibit 8.5) account for a significant portion of the MVNO segment; independent MVNOs such as Debitel, Lebara, KD8, DRI and Tchibo also have a clear presence in the market.

When considering the relative concentration of the mobile market in Germany, we think it is important to take into account the effect that MVNOs have. Without accounting for MVNOs, the German mobile market HHI sits at 2,884 based on the share of service revenues. When the MVNO share is included, the HHI drops by over 500 points to 2,305. Without accounting for MVNOs, the German mobile market was more fragmented than the peer group average. Including MVNOs in the HHI calculation results in Germany appearing to be the most fragmented of the peer group markets in this report.

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

E-Plus 1,840 1,855 2,074 2,224 2,501 2,461 2,698 2,816 3,005 3,021 3,092 3,099 3,149

O2 586 1,055 1,404 1,842 2,370 2,816 2,809 2,901 2,808 2,860 2,932 2,947 3,152

T-Mobile 5,719 6,253 6,778 7,377 7,656 7,758 7,435 7,156 7,045 7,008 7,047 6,881 6,779

Vodafone D2 5,356 7,116 6,635 7,313 7,848 7,925 7,699 6,990 6,820 6,505 6,524 6,408 6,472

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

Op

erat

or

serv

ice

Rev

enu

es €

m


115

Exhibit 8.5: Germany subscriber HHI with and without MVNOs – including MVNOs substantially reduces the HHI


Exhibit 8.6: Germany service revenues HHI, with and without MVNOs – including MVNOs suggests a substantially more fragmented market


DT, 1,045

Vod, 893

O2D, 285

E-plus, 438 HHI = 2660

DT, 685

Vod, 586

O2D, 187

E-plus, 287

MVNOs, 361

HHI = 2106

DT, 1,130

Vod, 1,289

O2D, 239

E-plus, 226

HHI = 2884

DT, 813

Vod, 927

O2D, 172

E-plus, 162

MVNOs, 231

HHI = 2305


116

Exhibit 8.7: Germany mobile market structure – MVNOs have a significant presence

Source: Berenberg estimates, AGCOM, company reports

Emailed Anthony - awaiting reply on key names

emailed Isabel. Awaiting reply

emailed markus -awaiting reply

emialed Parys representative - Charlotte for answers

O2D

E-Plus branded retail stores, and online

Direct

e.g. Fonic*, KDG,

Tchibo MobilDrillisch, Unitymedia,

est. 10-20

MVNOs

e.g. PhoneHouseMedia Markt

Debitel

Reseller/Retail

O2 branded retail stores and online

Direct

E Plus

e.g. Base*, Yourfone*Aldi Talk

Che Mobil

est. 30-50

MVNOs

e.g. PhoneHouse

Media MarktDebitel

Reseller/

Retail

T-Mobile

Vodafone

e.g. Congstar

LebaraMobilkom/Debitel

est.20-30 MVNOs

e.g. PhoneHouseDebitel

Media Markt, ProMarkt

Reseller/

Retail

T-Mobile branded retail

stores and onlineDirect

e.g. o.tel.o, 1 and 1BildMobilDrillisch

est. 20-30

MVNOs

PhoneHouse

Media MarktDebitel

Reseller/

Retail

Vodafone branded retails stores and online

Direct


C

u

s

t

o

m

e

r

s


117

In terms of concentration in distribution, we think Germany looks like a well-supplied mobile market, with a combination of direct distribution, via MVNOs and via a variety of high street retailers, including well-known electrical goods outlets. Exhibit 8.7 illustrates the German mobile market structure.

Judgement: market concentration, MVNO presence and consumer choice: The German market appears to have the most developed MVNO segment of our focus group, although MVNOs owned by MNOs constitute a significant portion of the MVNO segment. In the event of consolidation in the German market, we think it is relatively well placed to see minimal impacts from remedies aimed at strengthening the MVNO segment.


While the German market appears to be the most fragmented of the peer group based on service revenue HHI, the story is more complex when we look at spectrum concentration. Using our adjusted spectrum approach, which attempts to adjust the HHI calculation to reflect the relatively higher value of spectrum below the 1Ghz frequency band, the German market appears to be among the most fragmented. This seems counter-intuitive as E+ chose not to pay up for 800Mhz spectrum in the 4G auction in 2012. However, E+ does have a lot of 1,800Mhz spectrum (see Exhibit 8.10) which, even using our “adjusted” HHI calculation, equalises the spectrum balance.

The problem arises when we start to consider the prospects for consolidation, specifically the combination of O2 with E+. This combination results in the service revenue HHI increasing by a relatively modest 334 points to 2,639 (see Exhibit 8.9), a figure which leaves the German market still looking more fragmented than the peer group average, assuming no other consolidation among the peer group. If we do assume the most obvious or most-talked-about consolidation among the peer group, then the German market in the event of an O2 and E+ combination moves back to being an outlier on service revenue HHI, with a significant gap between the German HHI (2,639) and its nearest peer, France (2,920).

Exhibit 8.8: Market split by subscribers, service revenues and spectrum; illustrates consolidation scope for customer, but not for spectrum


0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

DT Vod O2D E-plus



118

Exhibit 8.9: O2 and E+ combination results in lowest increase in HHI


Exhibit 8.10: Operator spectrum portfolios show an imbalance of more valuable sub-1Ghz spectrum


The impact of O2 and E+ on spectrum is altogether more challenging. The spectrum HHI, even using the “adjusted” definition of spectrum, increases from 2,529 to 3,595, taking Germany from the least concentrated spectrum market to the most concentrated market.

Spectrum remedies would almost certainly be required to allow consolidation in the German market. Therein lies the risk, that a significant amount of spectrum, most likely in the 1,800Mhz and 2,100Mhz bands, would have to be divested, opening up the possibility of a new entrant in the German market.

HHI

Subscriber

market share

(MVNOs

recognised) Change

Service revenue

market share

(MVNOs

recognised) Change

Current 2106 2305

O2D+E-plus 2570 463 2639 334

DT+E-plus 2993 887 3032 727

DT+O2D 2822 716 3052 747

Vod+E-plus 2926 820 3081 776

Vod+O2D 2768 662 3102 798

DT+Vod 3374 1267 4041 1736

Mhz 800 900 1800 2100 2100 2600 Total Share HHI

DT 20.0 24.8 40.0 19.8 5.0 45.0 154.6 25.5% 648

Vodafone 20.0 24.8 10.0 29.7 5.0 65.0 154.5 25.4% 647

O2 20.0 10.0 34.8 29.7 19.2 45.0 158.7 26.1% 683

KPN 0.0 10.0 54.8 39.6 5.0 30.0 139.4 23.0% 527

Total 60.0 69.6 139.6 118.8 34.2 185.0 607.2 2,506

Spectrum Multiples 1.0 1.0 2.0 2.5 2.5 9.0

Adjusted Mhz 800 900 1800 2100 2100 2600 Total Share HHI

DT 20.0 24.8 20.0 7.9 2.0 5.0 79.7 28.4% 804

Vodafone 20.0 24.8 5.0 11.9 2.0 7.2 70.9 25.2% 636

O2 20.0 10.0 17.4 11.9 7.7 5.0 72.0 25.6% 655

KPN 0.0 10.0 27.4 15.8 2.0 3.3 58.6 20.8% 434

Total (adj) 60.0 69.6 69.8 47.5 13.7 20.6 281.2 2,529


119

Exhibit 8.11: Spectrum concentration and adjusted spectrum concentration – no consolidation; adjusting spectrum HHI to reflect the different value of sub 1Ghz spectrum does not change the HHI much


Exhibit 8.12: Spectrum concentration and adjusted spectrum concentration – with consolidation; the spectrum HHI increases significantly in the event of an O2 and E+ combination


Judgement: spectrum share versus customer share: We think Germany is exposed to significant spectrum remedies in the event of consolidation which could open up the market to a new entrant. However, Germany is better positioned than many peers in that in the event of the most-talked-about combination (O2 and E+) actually occurring, the impact on customer HHI is relatively small, and Germany remains the least concentrated market of our peer group on this measure.


Prices in Germany appear relatively high in the contract market and relatively low in the pre-paid market. In contract SIM-only, looking at the cost of a typical limited-use bundled tariff (aiming for 100 voice MOUs, several hundred SMS and 200-300Mb of data), Germany appears to be the second-most-expensive market, similar to price levels in Spain, and cheaper-than-average prices in Sweden.

Moving up the spend curve to look at big bundle offers, aiming for unlimited voice and SMS and 1Gb of data, Germany looks better, cheaper than Italy Spain and the Netherlands, although still above the peer group average. The picture does not change much when we include a 16Gb iPhone 5 with our big bundle.

DT, 648

Vodafone, 647

O2, 683

KPN, 527

HHI = 2506

DT, 804

Vodafone, 636

O2, 655

KPN, 434

HHI = 2529

DT, 648

Vodafone, 647

O2+KPN, 2,410

HHI = 3706

DT, 804

Vodafone, 636O2+KPN,

2,156

HHI = 3595


120

In the pre-paid segment, Germany appears cheaper than the average, although this reflects quite expensive per minute rates in the Netherlands and the UK. We based our pre-paid comparison on entry-level prices (typically a €5 recharge or top-up) which typically excluded the inclusion of free or bonus minutes and SMS. Voice and SMS rates in Germany were cheaper than in Italy, Spain, the UK and the Netherlands, but more expensive than in Sweden and Austria.

Exhibit 8.13: ARPU per month by operator – signs of stabilisation lately – MTR seasonality explains


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

E-Plus 26.0 21.0 25.0 24.0 24.0 21.0 19.0 17.0 15.0 13.0 13.0 12.0 11.0

O2 25.5 29.0 28.6 31.0 31.0 26.5 23.7 19.4 16.9 15.3 14.5 13.8 13.6

T-Mobile 28.7 24.0 24.0 24.0 24.0 23.0 20.0 17.0 15.0 15.0 17.0 16.0 15.0

Vodafone D2 31.5 25.3 26.0 25.9 24.9 22.9 20.9 17.9 16.2 15.8 15.5 15.2 16.5

5.0

10.0

15.0

20.0

25.0

30.0

35.0

AR

PU

€p

cm


121

Exhibit 8.14: Typical mobile tariffs packages in Germany


Exhibit 8.15: Post-paid share of base by operator – Germany has a relatively underdeveloped post-paid segment versus the peer group


Tariff

Contract

period

H/s

upfront

cost Mou SMS

Data

MB WiFi/Cloud

1-time

fees €pcm

Amorti

sed

cost

pcm £

Typical limited SIM-only

Base/E+ Smart 24 0 100 100 100 0 0 15.00 15.00

Simply All-in XL smart 24 0 250 250 500 0 19.95 14.95 15.78

Congstar Smart M 24 0 100 u/l 200 0 0 16.99 16.99

Mobilfunk Tariff Flat 100 24 0 100 u/l 300 0 0 17.91 17.91

O2 Blue Select 24 0 100 u/l 300 0 0 18.99 18.99

1 and 1 Basic 24 0 u/l u/l 500 10Gb cloud 0 19.99 19.99

VOD Red S 24 0 100 3000 200 0 0 26.99 26.99

DT Complete Comfort S 24 0 100 u/l 500 0 0 27.96 27.96

Typical unlimited SIM-only

Drillisch Simply) Flat XM 24 0 u/l u/l 1000 0 19.95 24.95 25.78

Mobilkom (Freenet) Your Flat with SMS & 1Gb options 24 0 u/l u/l 1000 0 0 29.80 29.80

1 and 1 Plus tariff 24 0 u/l u/l 1000 25Gb cloud 0 29.99 29.99

Base/E+ All-in 24 0 u/l u/l 500 5Gb cloud 0 30.00 30.00

Congstar Allnet Flat L 24 0 u/l u/l 1000 0 0 34.99 34.99

Base/E+ Average of Base packages 24 0 u/l u/l 1500 7.5Gb cloud 0 35.00 35.00

O2 Blue All-in M 24 0 u/l u/l 1000 0 0 38.48 38.48

DT Complete Comfort M 24 0 u/l u/l 750 u/l 0 39.96 39.96

Base/E+ All-in plus 24 0 u/l u/l 2000 10Gb cloud 0 40.00 40.00

Vod Red M 24 0 u/l u/l 1000 25Gb Cloud 29.99 44.99 46.24

DT Complete Comfort L 24 0 u/l u/l 1000 u/l 0 55.96 55.96

Typical unlimited SIM-only+16Gb iphn 5

1 and 1 Plus tariff 24 199.99 u/l u/l 1000 25Gb cloud 0 39.99 48.32

Mobilkom (Freenet) Your Flat with SMS + Gb options 24 249.95 u/l u/l 1000 0 0 39.90 50.31

Drillisch Simply 24 635 u/l u/l 1000 0 19.95 26.90 54.19

VOD Red M 24 99.9 u/l u/l 1000 25Gb 29.99 48.99 54.40

Base/E+ All-in 24 89.0 u/l u/l 500 5Gb cloud 0 60.00 63.71

Base/E+ All-in plus 24 89.0 u/l u/l 2000 10Gb cloud 0 63.48 67.19

Congstar Allnet Flat L 24 99 u/l u/l 1000 0 9.99 59.99 64.53

O2 Blue All-in m 24 34.0 u/l u/l 1000 0 0 63.48 64.90

Base/E+ Average of above 24 89.0 u/l u/l 1500 7.5Gb cloud 0 69.99 73.70

DT Complete Comfort M+iphone 5 24 0 u/l u/l 1000 u/l 0 69.95 69.95

0

10

20

30

40

50

60

70

80

90

IT GE UK Avg. NDL Swe ES FR DK

Postpaid customers as % of total, 2009 to Q1/2013


122

Judgement: low prices and big bundles: German SIM-only contract prices generally appear above-average versus our peer group suggesting that Germany, relatively speaking, has less potential than some of the peer group to hope that consolidation can put a floor under prices.


Judgement: pre-paid versus post-paid mix: Except for Italy (which is a pre-paid special case in our view), Germany has the most to gain in the event that consolidation serves to accelerated the mix-shift from pre-paid to higher-spending, lower-churning contract services (see Exhibit 8.15).

Network-sharing

Network-sharing in Germany appears relatively limited (see Exhibit 8.16), with one main arrangement between TEF’s O2 and VOD based solely on a site-sharing agreement. DTE’s T-Mobile does have a backhaul agreement with O2, but this seems more like a commercial agreement with DTE as supplier and O2 as customer rather than a true “sharing” arrangement.

There has been much talk of O2 and E+ entering into a comprehensive network-sharing arrangement as an alternative to a potentially complex full merger. The talk has even considered the prospect of E+ gaining access to O2’s 800Mhz 4G spectrum in return for a payment to O2. The talk surrounding this idea, promoted by comments from both parties, suggests that efficiencies would be material, approaching several billions of euros, and perhaps so material as to render a full merger irrelevant. We are sceptical about the magnitude of synergies from a network-sharing deal – after all, if synergies were so substantial, surely we would have seen far greater progress on network-sharing across the European mobile industry. As it is, network-sharing agreements have rarely progressed beyond site- and tower-sharing.

Exhibit 8.16: Germany network-sharing arrangements are currently limited

Source: Company

With limited network-sharing arrangements in place, this opens up the scope for any consolidation in Germany to be accompanied by network access remedies which would tend to increase efficiencies for smaller operators, providing them with scale benefits to reinvest in lower prices.

Judgement: network-sharing: Germany appears relatively exposed to consolidation remedies that might seek to enforce network access or sharing to underpin weaker players in the market in the event of stronger players consolidating. However, this could be a moot point as the most likely combination in Germany is the two smaller players – we think it unlikely that the regulator

Date Country Operators Sites

Back-

haul

Full

RAN

Spec-

trum

Core

n/w

Aggregate cost saving

identified Comments/rationale

Apr-09 Germany* TEF/VOD Y VOD: 10% of opex Vodafone has sublet towers to O2 on an

agreed price per tower

incorporating c400 sites as per

2009/2010.

Network sharing


123

would demand that a combined O2 and E+ open up their network to T-Mobile and VOD.


As far as we could discover, the German regulatory authorities have not explicitly stated an objection to further network consolidation, or specified a preference for a minimum number of networks. Last year, TEF and KPN progressed O2/E+ merger talks to a fairly advanced stage before calling off such a combination due to credit agency concerns. This suggests that both operators and their parents considered that prospects for a merger to gain regulatory and anti-trust approval were good.


124

Italy: spectrum remedies likely to undermine upside


It certainly looks as if TI and Hutchison are considering having a go at consolidation. If they proceed, then Italy could become an important test case for in-market consolidation given it would involve the challenging prospect of the number one player (by subscribers) combining with the number four player, 3 Italia. If Hutchison is keen, as evidently it is, and TI’s board decides not to pursue a deal, then that could still leave scope for VOD or Wind to try to come to terms with Hutchison. The fact that a deal between TI’s TIM and 3 Italia would likely be overseen by the AGCM, the local Italian anti-trust authority, might also raise expectations of a more lenient approach to remedies. As we show later in this section, though, the characteristics of the Italian mobile market suggest considerable scope for remedies to be imposed.


In reality, the combination of TI and 3 Italia is likely to result in considerable remedies. The same can probably be said of Vodafone Italy combining with 3 Italia. A Wind and 3 combination might appear more likely to succeed given that the combined entity would, with a c31% market share of service revenues, be similar in size to the two incumbents, and would only increase the service revenue HHI by 500 points to 3,116. This would still leave the Italian market above the average HHI of the peer group, but it would not be the most concentrated market – that honour would still reside with the Netherlands.


We see two key remedy risks in the event of a consolidation move in Italy. First, there would likely be a significant risk of spectrum remedies, especially in the event of a TIM and 3 Italia combination. This could see spectrum across the portfolio ring-fenced for a new entrant, or divested to Wind and VOD in order to rebalance the spectrum distribution. Second, given the relatively underdeveloped MVNO segment in Italy, regulators could impose more favourable MVNO terms and conditions on the merged entity, similar to the steps taken in Austria. There also remains a risk that regulators enforce more favourable terms for infrastructure access given the current limited scope of network-sharing in the Italian market.

In summary, then, we think Italian market consolidation, while needed in order to remove excess competitive tension in the market, would face considerable hurdles in the form of remedies to re-balance what would become a relatively concentrated market, both in terms of service revenue market share and spectrum market share.


125


The Italian mobile market has been going through tough times since market service revenue growth peaked out in 2007. Since then, the market has declined at a CAGR of 3.4% pa through to 2012, despite a degree of stability in the base of active SIM cards. The SIM base saw a setback through 2008-2010 as TI went through a massive SIM clean-up programme to remove inactive SIMs from the network (see Exhibits 9.1 to 9.2).

Wind has been the clear winner in the Italian mobile market in the last 10 years, gaining an approximate 17% share of market service revenues to a cumulative 20% share. In the last five years, Wind has taken 5ppt of service revenue share, a slower pace of gain but retaining positive momentum.

Exhibit 9.1: Italian mobile market revenues peaked in 2007, and have declined since


Exhibit 9.2: Italian mobile subscribers by operator


2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

H3G 306 1,808 2,693 2,790 2,673 2,566 2,531 2,361 2,076

Wind 439 1,803 2,403 2,698 2,927 3,176 3,464 3,593 3,695 3,813 3,836 3,669

Vodafone 5,476 6,178 6,956 7,763 8,028 7,848 7,655 7,680 7,694 7,526 7,179 6,442

TIM 7,545 8,198 8,723 9,077 9,161 9,023 8,856 8,615 7,915 7,230 6,537 5,983

0

5,000

10,000

15,000

20,000

25,000

Se

rvic

e R

ev

en

ue

s €

m

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Blutel 1.82

H3G 0.00 0.00 0.25 2.85 5.56 7.08 8.14 8.57 8.84 9.08 9.24 9.53

Wind 7.90 8.73 9.58 11.61 13.70 14.70 15.60 16.90 18.40 19.93 21.01 21.60

Vodafone 17.43 19.00 20.64 22.15 23.69 26.19 29.65 30.04 29.88 30.59 29.96 29.39

TIM 23.95 25.30 26.08 26.26 28.58 32.45 36.33 34.80 30.86 31.02 32.23 32.16

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

100.00

Su

bsc

rib

ers

(m

ln)


126

Hutchison’s 3 Italia, the latecomer to the market, has also gained a 9.5% share of market service revenues since launching in 2004. Wind and 3’s share gain has largely come at the expense of TI, whose revenue shares we estimate has reduced by 23.2ppt since 2001 to the current 32.9%. VOD has also been a share loser, albeit to a smaller degree – we estimate 5.2ppt of share loss since 2001 to the current 35.5%. The arrival of 3 Italia in 2004, coupled with the introduction of number portability from 2002, resulted in a significant step-up in competitive pressures. In recent years, regulatory factors (MTR reductions, roaming rate reductions) have also become an important drag on sector revenue growth, with the weak macroeconomic backdrop adding to the challenge for operators.

Exhibit 9.3: Italian mobile service revenue share has gone to Wind and H3G at the expense of VOD and TIM


Exhibit 9.4: Italy operator mobile service revenues


2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

H3G 0.0% 0.0% 0.0% 1.5% 8.2% 11.8% 12.3% 11.8% 11.7% 12.0% 11.9% 11.4%

Wind 3.3% 11.1% 13.3% 13.6% 13.4% 14.0% 15.2% 15.9% 16.9% 18.1% 19.3% 20.2%

Vodafone 40.7% 38.2% 38.5% 39.1% 36.6% 34.5% 33.6% 34.0% 35.2% 35.7% 36.1% 35.5%

TIM 56.1% 50.7% 48.2% 45.7% 41.8% 39.7% 38.9% 38.2% 36.2% 34.3% 32.8% 32.9%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Se

rvic

e R

ev

en

ue

Sh

are

%

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

TIM 7,545 8,198 8,723 9,077 9,161 9,023 8,856 8,615 7,915 7,230 6,537 5,983

Vodafone 5,476 6,178 6,956 7,763 8,028 7,848 7,655 7,680 7,694 7,526 7,179 6,442

Wind 439 1,803 2,403 2,698 2,927 3,176 3,464 3,593 3,695 3,813 3,836 3,669

H3G 306 1,808 2,693 2,790 2,673 2,566 2,531 2,361 2,076

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

Op

era

tor

Se

rvic

e R

ev

en

ue

s €

mln


127

The market, which has SIM penetration of 150% (indicating that saturation was achieved many years ago), has responded by increasing the competitive intensity of promotions and short-term offers in a market which is dominated by pre-paid customers who churn often and are able to switch quickly between operators to take advantage of the latest cheap offers. The government’s imposition of a tax on monthly mobile phone contracts (€5.13 pcm for consumers) serves to ensure that the market will likely remain a mostly pre-paid market until the tax is removed.

In recent quarters, the pace of decline in market service revenues has accelerated as price competition has intensified and regulatory effects have increased. In Q1 2013, we estimate the Italian market saw service revenues decline by c15% yoy, with MTRs alone accounting for c6ppt of the decline.

While we expect the pace of decline in the Italian market to improve from Q1 (eg few forecast a three-year CAGR of -6% through to 2016), the Italian market ranks as the most challenged, with the worst pace of decline forecast out of the eight markets addressed in this report.

This suggests to us that the Italian market is the market that most needs to see in-market consolidation and market repair. Alternatively, the opportunity cost of not seeing in-market consolidation in the Italian market is the highest of the markets we review.


Italy’s mobile market is the third-most-fragmented of our eight focus markets when MVNOs are taken into account in the HHI calculation (see Exhibit 9.5). According to AGCOM, MVNOs are relatively underdeveloped in the Italian market, with only a 4.5% share of retail subscribers. However, this is enough to reduce the HHI of the market by 184 points to 2,610 (using service revenue share rather than subscriber share), compared to the average of 2,696 for the eight markets in this report, and well below markets like the Netherlands (3,212) and Spain (2,901).

Exhibit 9.5: Italy HHI with and without MVNOs, subscribers

Source: Berenberg estimates, company reports, AGCOM

TI, 1,196

Vod, 991

Wind, 559

three, 106

HHI = 2851

TI, 1,090

Vod, 904

Wind, 510

three, 97 MVNOs,

20

HHI = 2621


128

Exhibit 9.6: Italy service revenues HHI with and without MVNOs

Source: Berenberg, company reports, AGCOM

TI, 993

Vod, 1,209

Wind, 413

three, 180

HHI = 2794

TI, 922

Vod, 1,124

Wind, 383

three, 167 MVNOs,

13

HHI = 2610


129

Exhibit 9.7: Italy mobile market structure – MVNOs have a presence but it is small


Three Italia

Wind branded retail stores and online

Direct

Mobile FastwebBip MobileLycamobile

est 5-10

MVNOs

Petrol stations,

newsagents, street vendors

Reseller/Retail

3 branded retail stores and online

Direct

Wind

A-MobileGreen

NetValue

est 5-10 MVNOs

MediaMarkt/Saturn

EuronicsExpert, Unieuro

Reseller/Retail

TIM

Vodafone

Coop Italia

Tiscali MobileNoverca

est. 5-10MVNOs

Mediaworld, Trony, Euronics, Darty, +supermarkets

Reseller/

Retail

TIM branded retail

stores and onlineDirect

PosteMobileDaily Telecom

Erg Mobile

est. 5-10 MVNOs

Unieuro, Mediaworld, Saturn, Expert

Reseller/Retail

Vodafone branded retail stores and online

Direct


C

u

s

t

o

m

e

r

s


130

Given TI’s comments relating to its ongoing consideration of a combination with 3 Italia, it is worth considering the impact on Italy’s market concentration if such a move were actually to materialise. This would see Italy’s HHI increase significantly to 3,395 (versus 2,610 pre-consolidation). An important factor in any anti-trust remedies would be the degree to which the two combining operators were already substitutes for each other’s services in the market. While operators are keeping the details of which competitors they lose their customers to very close to their chests, we suspect from the market share trends outlined above that 3 Italia has been a significant gainer of market share at the expense of TI and VOD, rather than at the expense of Wind.

The lack of a stronger MVNO sector could be an important factor in determining the likely reaction of anti-trust authorities to any proposed in-market consolidation proposals that materialise. From various sources, we estimate that the Italian market has about 20-40 MVNOs. We know from AGCOM data that they had collectively taken about a 4.5% subscriber share as at end 2012, with one MVNO, Poste Mobile, accounting for about half of that figure, and Fastweb accounting for about 15%. Most industry executives that we spoke to attributed the success of the Poste Mobile MVNO to the broad distribution capability leveraged off the Italian post office’s 26,000 high street outlets. Another newcomer in the MVNO segment is Bip, which has been seeing some success since launching in 2012 with a low-priced voice and data offer (Bip’s tariffing seems to imply that over-the-top “apps” are the future, and SMS is the past).

Given the relatively underdeveloped nature of the MVNO segment in Italy, we believe there is a high risk that consolidation remedies could result in improved conditions to boost the MVNO sector (such as improved terms for network access and support).


Exhibit 9.8: Market split by subscribers, service revenues and spectrum Illustrates spectrum imbalance for 3 Italia


0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

TI Vod Wind Three Italia



131

At first glance, the Italian market looks only slightly more concentrated than the average of our eight focus markets when it comes to looking at spectrum distribution among the operators. However, this does not take into account that 3 Italia has no 800Mhz spectrum, and only 2x5Mhz of 900Mhz spectrum (which it recently re-farmed for use for 3G mobile services). Exhibit 9.8 shows the distribution of customer, service revenue and spectrum share in the Italian market, with a final bar showing our adjusted spectrum share which uses a “factor” to reflect the lower value of spectrum above 1Ghz. On our adjusted figure, 3 Italia would seem to own only 12% of total spectrum in Italy, leaving it at a potential disadvantage to the other three players. The prospects for Italian mobile market consolidation emerge more clearly when we look at the impact on the market’s spectrum HHI of a consolidation move.

Exhibit 9.9: Operator spectrum portfolios are show an imbalance of more valuable sub-1Ghz spectrum


Exhibit 9.10: Spectrum concentration and adjusted spectrum concentration – no consolidation


Mhz 800 900 1800 2100 2600 Total Share HHI

TIM 20.0 44.8 40.0 35.0 30.0 169.8 27.9% 779

Vodafone 20.0 38.4 40.0 35.0 30.0 163.4 26.9% 721

Wind 20.0 35.2 30.0 35.0 40.0 160.2 26.3% 693

3 Italia 0.0 10.0 20.0 35.0 50.0 115.0 18.9% 357

Total 60.0 128.4 130.0 140.0 150.0 608.4 100.0% 2,551

Spectrum Multiple x 1.0 1.0 2.0 2.5 9.0

Adjusted Mhz 800 900 1800 2100 2600 Total Share HHI

TIM 20.0 44.8 20.0 14.0 3.3 102.1 31.3% 981

Vodafone 20.0 38.4 20.0 14.0 3.3 95.7 29.4% 862

Wind 20.0 35.2 15.0 14.0 4.4 88.6 27.2% 739

3 Italia 0.0 10.0 10.0 14.0 5.6 39.6 12.1% 147

Total (adj) 60.0 128.4 65.0 56.0 16.7 326.1 100.0% 2,729

TIM, 779

Vodafone, 721

Wind, 693

3 Italia, 357

HHI = 2551

TIM, 981

Vodafone, 862

Wind, 739

3 Italia, 147

HHI = 2779


132


Source: Berenberg estimates, Company reports

As the focus at present is on the talks between TI and Hutchison regarding a combination with Hutchison’s 3 Italia, we show the impact this combination would have on Italy’s spectrum concentration. On an adjusted basis, ie adjusting 3 Italia’s above-1Ghz spectrum to reflect its lower value relative to TI’s sub-1Ghz spectrum, the spectrum HHI of the market would increase by 710 points to 3,489. While this would still leave Germany as the key outlier on spectrum HHI in a consolidation scenario, Italy would take second place, and in terms of the combined effect of service revenue HHI and spectrum HHI, Italy would occupy the further position in the top right-hand quadrant of our chart. Spectrum remedies would almost certainly result, possibly accompanied by other remedies.

The combination of TI and 3 would end up owning 43.5% of all Italian mobile phone spectrum (using our adjusted measure), and 39% of Italy’s sub-1Ghz spectrum. At the other end of the spectrum (excuse the pun), Wind currently owns only 29% of sub-1Ghz spectrum. One remedy could therefore be to have TI and 3 divest sub-1Ghz spectrum to Wind (and possibly a little to VOD) in order to at least equalise ownership of the most valuable spectrum. Alternatively, the merged entity could be asked to make spectrum available to a new entrant across several spectrum bands.


Italy appears from the outside to be a very competitive market with all four operators seeming to be constantly adjusting their prices via short-term offers and promotions in order to stay in the game. 3 and Wind have tended in the past to be the price disruptors, although every so often VOD and TI have been known to respond aggressively via price. However, comparing the overall level of pricing with the other seven of our focus markets shows that Italy is not the lowest priced market of the eight.

TIM+3, 2,191

Vodafone, 721

Wind, 693

HHI = 3606

TIM+3, 1,888

Vodafone, 862

Wind, 739

HHI = 3489


133

Exhibit 9.12: ARPU per month by operator


Comparing prices within markets is difficult enough. Comparing prices across borders becomes very tricky due to the differing approaches to the market. Nevertheless, we have tried.

Our approach focused on pricing for three main usage/spend patterns; a typical limited use SIM-only offer with around 100 minutes of voice, several hundred SMS and 2-300Mb of data. Secondly, we looked at a typical “unlimited” use bundle offering unlimited any-network voice minutes and SMS, and 1Gb of data. Finally, we looked at this tariff including an iPhone 5 16Gb.

Italy proved difficult due to the broad range of tariffs on offer and the mostly pre-paid nature of offers in the market. As such, we had to be more accommodating when it came to which tariffs we would compare. These are shown in Exhibits 9.13 through 9.15. We used a wide selection of tariffs going beyond the four network operators in order to assess the extent of price discounting.

Exhibit 9.13: Typical “limited usage” SIM-only monthly costs vary widely


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

H3G 47.2 34.9 34.0 29.3 25.3 23.5 23.6 19.9 17.5

TIM 28.3 27.5 28.6 28.4 29.0 28.3 24.0 20.3 20.8 20.3 18.8 16.9 14.9

Vodafone 28.8 28.9 29.9 30.2 27.7 25.6 21.6 21.5 21.7 20.3 19.4 16.9

Wind 19.0 19.6 22.2 20.8 18.8 19.1 19.2 18.5 17.4 16.6 15.2 13.9

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

AR

PU

€p

cm

Tariff

Contract

period

H/s

upfront

cost

Mou

pcm SMS

Data

MB WiFi

1-time

fees € €pcm

Amortised

cost pcm €

3 Italia All-in Small 12 0 120 120 1000 0 0 4.00 4.00

3 Italia All-in Small 12 0 400 400 1000 0 0 6.00 6.00

BIP Megabip 500 12 0 500 0 2000 0 5 8.00 8.42

Vod Smart 350 LE 12 0 350 350 1000 0 19 9.90 11.48

Poste Smallx2 12 0 500 500 1000 0 0 13.00 13.00

TI TIM Sconta Full Medium 12 0 400 400 1000 0 9 15.00 15.75

TI Tutto Compresso 300 12 0 300 300 1000 0 0 19.00 19.00

Wind All-inclusive Smart 12 0 400 400 2000 0 0 15.00 20.13

TI Tutto Compresso 300 12 0 300 300 1000 0 0 34.13 34.13

Vod Smart 500 12 0 500 500 500 0 0 34.13 34.13


134

Exhibit 9.14: Typical “limited usage” SIM-only monthly cost vary widely


Exhibit 9.15: Typical “limited usage” SIM-only monthly cost vary widely


In the cases of both of our SIM-only target tariffs, we also bent our rules to allow pre-paid tariffs in Italy to be taken into account, since these are by far the prevalent offer in the market. It should be noted that many such “pre-paid” tariffs in the Italian market are effectively pseudo-post-paid in the sense that they can only be acquired with a credit card with automatic monthly debits of a specified recharge amount.

The overall conclusion is that pricing in Italy remains very competitive versus the peer group, although we have to take into account the higher bundled usage in order to see this. In reality, we struggled to find bundles that came close to matching our target usage profiles, particularly in terms of data usage: most offers included a minimum 1Gb of data rather than the 200-300Mb we were looking for. Similarly, when looking at included voice minutes and SMS, bundles were generally larger in Italy than for the peer group.

When we put our Italy pricing data onto comparison charts alongside the peer group (as shown earlier), Italy looks in line to just above the peer group average. This is a valid comparison, since these are real money costs to the consumer. However, it also remains valid to say that Italian consumers receive a lot more value for their spend than consumers in most of the other peer group markets, perhaps with the exception of Sweden.

The high level of competitive intensity, availability of big bundles and good value already available for consumers in the Italian market suggest that the Italian market could benefit from consolidation if remedies are light or if the impact of remedies could be mitigated. Alternatively, without consolidation, we should expect to see the Italian market remaining ultra-competitive.

Tariff

H/s

upfront

cost

Mou

pcm SMS

Data

MB

WiFi/

Cloud

1-time

fees € €pcm

Amortised

cost pcm €

BIP MegaBip without limits 0 u/l 0 2000 0 5 18.00 18.42

TI TIM Sconta Full Large 0 1000 1000 2000 0 0 24.00 24.00

Wind All-inclusive unlimited 0 u/l u/l u/l 0 0 29.00 34.13

3 Italia Top SIM 3000 0 3000 600 20000 0 0 30.00 35.13

Poste Zero Infinity 0 u/l u/l 1000 0 0 34.00 39.13

Vod Unlimited Vodafone 0 u/l u/l 1000 0 0 41.90 41.90

Vod Relax Simple 0 u/l u/l 1000 50Gb cloud 0 44.13 44.13

TI Tutto Compresso Unlimited 0 u/l u/l 2000 10Gb cloud 0 64.13 64.13

Tariff

H/s

upfront

cost

Mou

pcm SMS

Data

MB WiFi/ Cloud

1-time

fees € €pcm

Amortised

cost pcm €

TI TIM Sconta Full Large+iPhone 729 1000 1000 2000 0 0 24.00 54.38

3 Italia Top Smart 3000 570.0 3000 600 20000 0 0 30.00 58.88

Poste Zero Infinity 729 u/l u/l 1000 0 0 34.00 69.51

Vod Unlimited Vodafone 699 u/l u/l 1000 0 0 41.90 71.03

Vod Relax Simple 699 u/l u/l 1000 50Gb cloud 0 44.13 73.26

TI TIM Smartphone Unlimited 379 u/l u/l 2000 0 0 64.13 79.92


135




The Italian market has the lowest proportion of post-paid customers of all of our peer group. This reflects the structural impact of the government’s tax on post-paid services (€5.13pcm for consumers). As with other markets, there is creeping growth in the proportion of post-paid subscribers, although this is the slowest increase since 2009 of all our peer group markets. Given that this state exists largely due to a government tax, we do not see consolidation offering much scope to accelerate the trend towards lower-churning, higher-spending post-paid customers.

Network-sharing

Network-sharing in Italy appears relatively limited (see Exhibit 9.17), with only two arrangements, both involving only site-sharing. This opens up the scope for any consolidation in Italy to be accompanied by network access remedies which would tend to increase efficiencies for smaller operators providing them with scale benefits to reinvest into lower prices.

Exhibit 9.17: Italy network-sharing arrangements – some do exist, but limited to site-sharing

Source: Company

0

10

20

30

40

50

60

70

80

90




Back-

haul

Full

RAN

Spec-

trum

Core

n/w



Nov-08 Italy VOD/TIM Y n/a Site sharing agreement for new

deployment and for consolidation of

existing sites to achieve CAPEX

avoidance for passive infrastructure and

OPEX savings by reducing rental spend.

Jul-09 Italy TIM/3 Y 30% of network costs (3

yrs) or 7-8% of opex

Boost coverage; reduces cost of site

leases, decommissioning of antennaes.

Network sharing


136


As far as we could discover, Italian regulatory authorities have not explicitly stated an objection to further network consolidation, or specified a preference for a minimum number of networks. TI has seemed keen to stress that a combination of TIM with 3 Italia would fall under Italian anti-trust jurisdiction, perhaps implying a more forgiving attitude should be expected than if jurisdiction were to be claimed by Brussels. Even so, national anti-trust authorities are obliged to follow the framework laid down by the EC competition commission when assessing mergers in-market.


137

Dutch mobile has little prospect of consolidation


We see limited prospects for the Dutch market to consolidate over the next two years. Having fallen to three operators, the regulators felt compelled to earmark cheap spectrum for a new entrant at the recent auction, with bid tension in that process coming from two entities that saw merit in launching a network-based operation.

Concerns about the competitive environment were partly behind this decision and it is easy to understand why. Although in terms of spectrum allocation and straight subscriber numbers it appears that there is sufficient competition, that is not the case when the assessment is expanded to include service revenue. Even when MVNOs are included on subscriber numbers, while the HHI falls below 2,800: the Netherlands is still deemed to be the least competitive market that we have researched under this measurement.

In addition, unless the consolidation was to involve the recently issued TEL2B spectrum, there would be very material drops in the perceived competitive intensity of the market. While there has been some press speculation that TEL2B could look to exit the Netherlands using the newly acquired spectrum as a bargaining chip, we suspect that this would be frowned upon by the authorities and we suspect that at best the spectrum would be reissued to another operator, potentially at a lower price.

In terms of the operators’ desire to consolidate, we expect the focus to be on T-Mobile. It is the smallest of the three MNOs, but with ironically the largest amount of spectrum, so it would add further to the risk of spectrum being reallocated. We doubt whether KPN would be able to participate in consolidation given that its market share under any scenario would likely be more than 50% of the subscribers. Finally, VOD would also likely be unable to participate in consolidation and as VOD is currently enjoying the strongest operational growth among the three MNOs, it has little need to become embroiled in a tough regulatory argument over consolidation.


With the HHI based on service revenues well above the ideal threshold of 2,800 and with the subscriber market share only just under the limit, we think regulators would take a dim view of consolidation in the Netherlands. Already the least competitive of the four-player European markets, we doubt that any combination would result in a simple approval. As can be seen in Exhibit 10.1, with the exception of the absorption of TEL2B, all the combinations see a significant lowering in the competitive intensity. Even including TEL2B, the HHI lifts to over 3,250 for any combination with TEL2B and an existing MNO.


138

Exhibit 10.1: HHI for changes in Dutch market structure


All that being said, we suspect the only combination that would be countenanced would be a merger between one of the existing MNOs and TEL2B. Any other combination would imply a very restricted level of concentration in the market and we doubt whether remedies could alleviate the lack of competition. The most obvious candidate on this front would be with T-Mobile, which has struggled in the Netherlands and which is also the smallest MNO.

For TEL2B, while the changes in the HHI on consolidation (and therefore the implied concentration in the market) is modest, we doubt this would hold much water. TEL2B has not yet launched its MNO offering on its own network and is unlikely to do so for some months, which makes the change in the HHI number rather irrelevant for now. Consequently we are not sure the regulators would take a positive view on consolidation.

However, even if it was feasible, we think the remedies would be rigorous.


With the Dutch authorities having recently seen the need for a fourth network operator, which was achieved via the allocation of cheap but high-quality spectrum to a new entrant, we would expect any remedies to be exacting.

We would anticipate further reallocation of spectrum by the regulator, and with T-Mobile already having a disproportionate amount of spectrum, it is possible that this would not necessarily be limited to the recently acquired spectrum. Having already received the cash for this spectrum, the government would be able to focus on the benefits to the consumer and not the public purse.

With the cable operators actively seeking a mobile opportunity and planning moves into the mobile market, there is also a ready-made acquirer. We also think that there could be some remedies in the shape of heightening the opportunity for MVNOs. It is interesting to note that although MVNOs have 3m customers or around 30% of the market (which does help the subscriber-based HHI calculations), the share of service revenue is much more modest and we estimate that this is somewhere between 5% and 10% (towards the lower end).

This suggests that there is insufficient opportunity in what we call the big bundle marketplace and we would expect remedies to be put in place to encourage the development of these big bundles and that this would be best achieved by providing attractive prices and platforms for MVNOs.

HHI

service

revenue mkt

share (inc

MVNO's)

Change

Sub Market

Share (Inc

MVNO's)

Change

Current 3,128 2,744

KPN/T-Mobile 4,915 1,787 4,352 1,609

KPN/Vodafone 5,646 2,519 4,688 1,945

KPN/Tele2 3,380 252 2,957 214

T-Mobile/Vodafone 4,439 1,311 3,761 1,017

T-Mobile/Tele2 3,259 131 2,855 112

Vodafone/Tele2 3,313 185 2,879 135


139

Market overview

The Dutch market is currently in calm waters but there is a potential storm brewing. The storm comes in the guise of TEL2B, which has secured highly attractive 800mHz spectrum which it will use to launch as a full MNO in conjunction with its already well entrenched MVNO. The expected launch date for TEL2B on its own network is some time in 2014, although we would not be surprised to see an earlier launch.

The other dynamic in the Netherlands is the potential for a new MNO targeting the big bundle SIM-only part of the market. In this segment of the market, which has been core to the successful launch of ILD and Telenet Holland is the most expensive in the Europe. In addition, the Netherlands remains one of the most expensive pre-paid markets.

Exhibit 10.2: The Netherlands is expensive for big bundle SIM-only


Against this backdrop, we see significant scope for TEL2B, which has been becoming more aggressive in the market over the past 18 months, in order to shake the market up further, win share and drive down prices.

In terms of current market share by subscribers, KPN remains the largest player by some way, with around 50% of the market, although this has been on the wane with VOD in particular gaining traction. It is also worth highlighting that KPN has by far the largest share of wholesale MVNO customers, with over 2m of the c3m MVNO connections. Notwithstanding this, we understand that TEL2B (which uses the T-Mobile network) is the largest MVNO with just over half a million subscribers.

10.0

20.0

30.0

40.0

50.0

60.0

70.0

FR AU UK DK Avg. Swe GE IT ES NDL

€p

cm

Typical unlimited (voice & SMS) SIM-only big bundles


140

Exhibit 10.3: Dutch mobile subscribers by operator


While KPN’s share of subscribers has been falling, the drop in its service revenue market share is more pronounced (3% over the last three years to 44%). The shift in market share has all been won by VOD, and has likely been driven by the extension OF ITS distribution channels, through acquisition of high street stores. This has caused KPN to invest heavily in distribution over the past couple of years, which has hurt operational performance.

Exhibit 10.4: Dutch service revenue by operator


In terms of the total service revenues in the market, this peaked in 2009, and has since fallen by around 10% to €5,785m. However, it is worth highlighting that data from OPTA (the Dutch regulator) indicates that the network revenues have continued to rise and the entire market decrease has been driven by MTR and roaming cuts. This is an interesting backdrop with regulatory headwinds dropping significantly, although they could potentially be replaced by competitive pressures.

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Vodafone 3.18 3.23 3.28 3.40 3.67 3.98 3.82 4.04 4.54 4.80 4.94 5.26 5.29

T-Mobile 0.77 1.18 1.44 1.99 2.26 2.32 2.55 4.89 5.31 4.59 4.53 4.91 4.72

KPN 4.85 5.23 5.03 5.21 6.08 8.07 8.64 9.39 9.89 10.31 9.81 9.77 9.92

0

5

10

15

20

25

Su

bsc

rib

ers

mil

lio

ns

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

KPN 2,262 2,149 2,368 2,834 2,971 2,972 3,030 2,942 2,722 2,552

T-Mobile 765 948 992 1,076 1,223 1,558 1,497 1,481 1,440 1,366

Vodafone 1,511 1,614 1,664 1,613 1,731 1,865 1,911 1,912 1,900 1,867

0

500

1,000

1,500

2,000

2,500

3,000

3,500

€m


141

Exhibit 10.56: Service revenue total


On the competitive side of the equation, the Netherlands is among the least concentrated markets in Europe when looked at using the HHI. In terms of service revenues under HHI, it is the least competitive market, and gives some indication why the regulators saw it fit to earmark spectrum for new entrants in the recent spectrum auctions.

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Vodafone 1,511 1,614 1,664 1,613 1,731 1,865 1,911 1,912 1,900 1,867

T-Mobile 765 948 992 1,076 1,223 1,558 1,497 1,481 1,440 1,366

KPN 2,262 2,149 2,368 2,834 2,971 2,972 3,030 2,942 2,722 2,552

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

€m


142

Exhibit 10.6 HHI metrics for service revenues across Europe


If we drill down further on the Netherlands, we can look at the HHI metrics on both subscriber numbers and service revenues. The subscriber numbers picture is markedly different from the service revenues measurement. On subscribers, if we look at the MNOs it highlights the apparent concentration of the market with an HHI of 3,727, significantly above the 2,800 threshold that signifies sufficient competition. However, once we take into account the MVNOs (we have split TEL2B out from the overall number), then the competitive intensity on pure customer numbers falls into acceptable levels at 2,744. On this basis, it would appear that with the introduction of MVNOs it is possible to create a sufficiently competitive market place in what is essentially a three-player market.

0 1000 2000 3000 4000

GE

UK

DK

IT

Avg.

FR

Swe

ES

NLD

HHI using MNO+MVNO's

0 1000 2000 3000 4000

UK

IT

DK

GE

Swe

Avg.

ES

FR

NLD

HHI using only MNO's


143

Exhibit 10.7: HHI with and without MVNOs, subscribers


However, when we look at the HHI calculation based on service revenues, the picture takes on a very different profile. While the HHI among the MNOs falls by a couple of hundred points, the HHI it is an unacceptably high 3,583. Including MVNOs, it is 10% higher than when looking at subscribers and is well above the 2,800 threshold at 3,205. This clearly signifies that the MVNO market is dominated by very low-value customers which would likely increase the risk of significant remedies in the Netherlands.

Moreover, it would point to the fact that the regulators were correct in their desire to reintroduce a fourth operator into the market during the last spectrum auction.

Exhibit 10.8: Dutch service revenues’ HHI, with and without MVNOs

Source: Berenberg estimates, company reports, OPTA

High prices market exposed to disruptive pricing The Dutch market remains a highly priced market and has not seen the emergence of big bundles that many other European markets have. Within the market, there has been a clear compression of ARPU per subscriber among the MNOs, with the historical premium enjoyed by KPN being rapidly eroded. Over the last five years, KPN has seen its near 30% higher ARPU halve, and its current trajectory still shows the weakest dynamics. The prospect of this coming under further pressure with the launch of a new entrant is high.

We can add further weight to this assertion if we look at the make-up of the industry in terms of customer mix. As we have shown using the HHI analysis, while the inclusion of MVNOs gives the impression that the Dutch market is fairly competitive in terms of subscriber numbers and lacks concentration, this is not the

KPN, 2,442

T-Mobile, 555

Vodafone, 730

HHI 3,727

KPN, 1,538

T-Mobile, 421

Vodafone, 615

MVNO, 163

Tele2, 7

HHI 2,744

KPN, 2,034 T-Mobile,

522

Vodafone, 1,028

HHI 3,583

KPN, 1,759

T-Mobile, 477

Vodafone, 947

Tele2, 10 MVNO, 13

HHI 3,205


144

case on service revenues. This means that the MVNO market has a significantly lower share of service revenues in the market, and by definition ARPU per subscriber. This is clear from OPTA data, with a very limited portion of the MNO revenues (around 4%) being driven from the MVNO community. This is despite the market being relatively well developed in terms of a shift to post-paid. We therefore believe that the ARPU of the network operators is significantly higher.

Exhibit 10.9: ARPU per month by operator


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

KPN 30.0 30.0 36.0 36.0 30.0 29.0 29.0 26.0 25.8 24.2 24.6 22.4 21.1

T-Mobile 35.0 32.0 33.0 37.0 36.0 37.0 38.0 27.0 24.0 24.0 27.0 26.0 23.0

Vodafone 31.9 36.4 36.4 39.3 37.5 34.5 31.7 35.9 35.3 34.4 31.6 29.6 28.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

€p

er m

on

th


145

Exhibit 10:7: Details on Dutch mobile distribution



As can be seen from Exhibit 10:11, the proportion of post-paid subscribers is over 60% of total subscribers. However, if we look at the OPTA data, then we see some supporting evidence that the actual penetration of post-paid customers of the MNOs is in fact much higher. As we have said, there are around 3m MVNO customers who are included in this breakdown; in addition, the Netherlands has just over 1m M2M (machine-to-machine) SIMs. If we adjust for this, we estimate that the post-paid element of the MNOs base is at least three-quarters. If we add this to the fact that the MNOs generate less than 5% of their revenues from MVNOs, it suggests that they continue to have a stranglehold on the market in the


146

Netherlands, supporting the need for the new entrant and suggesting that further consolidation in the short term would be subject to aggressive remedies that would make a deal unlikely.




KPN has a clear lead in terms of its market share in service revenues and subscribers in comparison to its share of spectrum. Exhibit 10:12 below excludes the TEL2B spectrum from the equation as it has not yet launched as an MVNO. In addition, Ziggo and UPC (the cable operators) won 2,600Mhz spectrum and have been vocal in terms of the plans they have for a disruptive mobile offering. Looking at the spectrum split, it is hard to see how further consolidation could be permitted. It appears that VOD has been most successful in balancing its customers and service revenues with its spectrum. This is also reflected in the fact that VOD has been the main gainer in the market over the last three years.

Exhibit 10.12: Market split by subscribers, service revenues and spectrum


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

KPN 34.39% 33.28% 37.29% 37.52% 35.98% 40.39% 44.61% 45.91% 50.13% 50.33% 55.47% 56.86% 59.00%

T-Mobile 56.77% 47.45% 44.78% 45.37% 46.57% 52.70% 52.39% 42.75% 42.55% 52.40% 58.55% 59.71% 64.66%

Vodafone 30.03% 35.00% 39.50% 43.00% 44.00% 45.60% 53.10% 57.00% 58.50% 60.70% 61.50% 63.40% 66.90%

25.0%

30.0%

35.0%

40.0%

45.0%

50.0%

55.0%

60.0%

65.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

KPN Vodafone T-MobileService Revenues Subscribers Spectrum Spectrum (Adj)


147

If we look in a little more detail at the concentration of spectrum in the Netherlands, it is clear that the spectrum allocation is not overly concentrated. This is true of both actual and adjusted spectrum, with both metrics falling below the 2,800 mark.

Exhibit 10.13: HHI for spectrum allocation and adjusted


However, this is much more down to the even split of the bulk of the spectrum among the three MNOs, with T-Mobile having benefited from its merger with Orange which had dropped the number of operators from four to three.

Exhibit 10.14: Spectrum allocation by operator


We have also looked at an adjusted measure, which weighs the quality of the spectrum with sub-1Ghz spectrum being the most attractive. We have applied a 50% weighting to 1,800Mhz and 1,900Mhz and a 40% weighting to 2,100Mhz spectrum. For 2,00Mhz spectrum, we have applied a division of nine.

Exhibit 10.15: Spectrum allocation by operator, adjusted


While the spectrum allocation looks well diversified and acceptable, it is still hard to devise a link-up of the existing MNOs that would not be the subject of extensive regulatory scrutiny.

KPN, 809.2

Vodafone, 555.0

T-Mobile, 954.2

Tele2, 112.1

Ziggo 4, 42.5

HHI 2,473

KPN, 859.7

Vodafone, 793.5

T-Mobile, 1,038.1

Tele2, 76.3 Ziggo 4, 2.4

HHI 2,770

Mhz 800 900 1800 1900 2100 2600 Total

KPN 20.0 20.0 40.0 5.0 39.6 50.0 174.6

Vodafone 20.0 20.0 40.0 5.4 39.2 20.0 144.6

T-Mobile 0.0 30.0 60.0 24.6 40.0 35.0 189.6

Tele2 20.0 0.0 0.0 0.0 0.0 45.0 65.0

Ziggo 4 0.0 0.0 0.0 0.0 0.0 40.0 40.0

Mhz 800 900 1800 1900 2100 2600 Total

KPN 20.0 20.0 20.0 2.5 15.8 5.6 83.9

Vodafone 20.0 20.0 20.0 2.7 15.7 2.2 80.6

T-Mobile 0.0 30.0 30.0 12.3 16.0 3.9 92.2

Tele2 20.0 0.0 0.0 0.0 0.0 5.0 25.0

Ziggo 4 0.0 0.0 0.0 0.0 0.0 4.4 4.4


148

Network-sharing

In the Netherlands, there is limited network share in potential. There is currently a pilot between VOD and KPN, but it is as yet uncertain what savings can be identified and the extent to which the network-sharing can be developed. Part of the reason for this is the regulatory environment, which largely enables open access to masts. This significantly reduces the benefits of network-sharing and has already seen some of the MNOs dispose of much of the base station sites, given that it is not nearly as much of a competitive advantage due to this.


149

UK: consolidation debate is moot


We think consolidation in the UK market is unlikely simply because existing regulatory attitudes express a preference for four network-based operators to continue to participate in the UK market. Put this important factor aside for a moment, and the UK market appears well positioned to deal with remedies that might arise if consolidation were indeed proposed. The UK has lots of choice, is among the most fragmented markets both in terms of service revenues and spectrum ownership, has a well-developed MVNO sector, and also has extensive existing network-sharing arrangements in place. While remedies could still do damage, a relative judgement that compares the UK to the other markets in our peer group study suggests that the UK is among the best positioned markets to see consolidation while coping with any reasonable remedies. Unfortunately, at least while current regulatory attitudes prevail, it seems unlikely that our theses will be tested.


The combination which would result in the least increase in market concentration would be VOD and 3. However, O2 and 3 would increase market concentration by only slightly more than VOD and 3. As stated above, it seems unlikely that operators would pursue consolidation without a firm sign of a change in regulatory attitudes.


Spectrum divestment would be the most likely remedy given that the combinations suggested above would see a significant increase in spectrum concentration in the UK market. UK spectrum concentration would increase to a similar level to that in Austria after the consolidation there of Orange and Hutchison, a combination which resulted in significant amount of 800MHz spectrum being reserved for a potential new entrant, in order to win EC approval for the merger.

Unlike Austria, we would not expect to see significant remedies to improve conditions for MVNOs given the well-developed MVNO sector and wide consumer choice already in place in the UK market.

Network remedies should not be ruled out, although the UK has well-developed network-sharing arrangements already in place, and in the merger of Orange with T-Mobile to form EE, the EC simply required that 3’s existing network arrangement with T-Mobile be secured.

At the extreme, it is possible to paint a negative remedy outcome in the event of UK mobile market consolidation: for example, spectrum divestment could attract a new entrant (BT?) and the EC could impose network access and/or preferential roaming arrangements as a condition for approving a merger.

As we have stressed throughout this report – our judgements are relative, and while it is possible to paint a negative picture, our overall conclusion is that the UK would be among the better positioned markets to pursue consolidation while facing the least risk from remedies. It is just a shame that prevailing regulatory attitudes seem likely to deter any serious attempts to consolidate the UK market from four networks to three.


150


The UK mobile market peaked in 2008 with service revenues of £19.7bn, and has declined since at an average rate of 2.5% pa through to 2012. Subscribers meanwhile have continued to grow, reaching 77.6m at end-2012, growth averaging 1.5% pa since 2008. ARPU must therefore have declined in the region of 4% pa over the same period, affected by regulatory price cuts, competitive pricing pressure, and a mix-shift towards lower revenue SIM sales (for data devices like iPads, and machine-to-machine).

TEF’s O2 has been the clear share gainer over the long term, increasing its share of service revenues by over 5ppt to 29.3% at the end of 2012. 3 has also increased its share from 8.3% to 10.6% over that period. Over the long term, the share gains of O2 and 3 have come at the expense of VOD and Orange/T-Mobile (now known as EE).

Exhibit 11: UK mobile market revenues peaked in 2008 but have slowly declined since


Exhibit 11.1: The UK mobile subscriber base has continued to expand


2005 2006 2007 2008 2009 2010 2011 2012

3 mobile 1,372 1,853 2,102 2,023 1,866 1,766 1,724 1,892

Vodafone 4,564 4,631 4,869 4,871 4,653 4,761 4,889 4,757

O2 3,958 4,628 5,068 5,414 5,529 5,730 5,361 5,223

OrangeTmobile 6,605 6,815 7,144 7,413 6,646 6,296 6,112 5,952

0

5,000

10,000

15,000

20,000

25,000

Se

rvic

e r

ev

en

ue

s £

m

2005 2006 2007 2008 2009 2010 2011 2012

3 mobile 3.49 3.87 4.13 5.09 5.82 6.75 7.54 9.05

Vodafone 16.33 16.94 18.45 19.17 19.11 19.17 19.33 19.54

O2 15.98 17.63 18.38 20.28 21.30 22.21 22.17 22.86

OrangeTmobile 27.67 27.72 28.46 28.67 26.54 27.21 26.83 26.15

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

Su

bsc

rib

ers

mln


151

However, in recent years, the long-term trend in share has been challenged by the efforts of EE to stabilise its share and a recovery in market share at VOD, affecting O2 rather than 3, which has continued to gain share.

Share trends in recent years can be attributed to a number of factors:

1) the renewal in 2009 by VOD of its Carphone Warehouse distribution agreement – the abandonment of this agreement in 2006 affected VOD’s market share at a crucial time;

2) the end of O2’s iPhone exclusivity in 2009 after two years of being the sole provider of the iPhone in the UK – all operators in the UK now offer the iPhone;

3) the success of EE in accelerating the mix-shift from pre-paid to post-paid customers ahead of the launch of 4G services in late 2012.

Exhibit 11.2: UK mobile service revenue share has gone to O2 and E+, although T-Mobile and VOD have fought back in recent quarters


2005 2006 2007 2008 2009 2010 2011 2012

3 mobile 8.3% 10.3% 11.0% 10.3% 10.0% 9.5% 9.5% 10.6%

OrangeTmobile 40.0% 38.0% 37.2% 37.6% 35.6% 33.9% 33.8% 33.4%

O2 24.0% 25.8% 26.4% 27.5% 29.6% 30.9% 29.6% 29.3%

Vodafone 27.7% 25.8% 25.4% 24.7% 24.9% 25.7% 27.0% 26.7%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Se

rvic

e r

ev

en

ue

sh

are


152

Exhibit 11.3: All four UK network operators have seen service revenues stabilising


In Q1 2013, the UK market reported a decline of about 4% in service revenues, of which we estimate 3ppt was accounted for by termination rate cuts. Excluding termination rate cuts, the UK is among the more resilient mobile markets of our focus markets, which averaged a reported decline of about 8% yoy and an ex-MTR decline of 4% yoy in Q1 2013.

Judgement: We think the UK needs in-market consolidation less than other markets in our peer group study. Current trends, and our expectations for the mid-term outlook, suggest that the opportunity cost to the UK market of not seeing consolidation is relatively low.


The UK is among the more fragmented markets within the peer group, with a service revenue HHI of 2,743 (see Exhibit 11.4), ranking the UK as the most fragmented market. However, we believe that the presence of MVNOs in the market needs to be accounted for in assessing market concentration, and the UK has a relatively well developed MVNO segment which we think controls about 8% share of UK market service revenues.

When the UK market HHI is adjusted for the presence of MVNOs, the market appears even more fragmented, with the service revenue HHI falling by 357 points to 2,386. If we compare this with the HHIs for the rest of the peer group adjusted for MVNO presence in their market, the UK still appears among the most fragmented markets, but loses its top spot to Germany, with the UK dropping to second place.

2005 2006 2007 2008 2009 2010 2011 2012

Vodafone 4,564 4,631 4,869 4,871 4,653 4,761 4,889 4,757

O2 3,958 4,628 5,068 5,414 5,529 5,730 5,361 5,223

OrangeTmobile 6,605 6,815 7,144 7,413 6,646 6,296 6,112 5,952

3 mobile 1,372 1,853 2,102 2,023 1,866 1,766 1,724 1,892

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

Op

era

tor

serv

ice

re

ve

nu

es

£m


153

Exhibit 11.4: UK subscriber HHI with and without MVNOs – including MVNOs substantially reduces the HHI


Exhibit 11.5: UK service revenues HHI with and without MVNOs – including MVNOs suggests a more fragmented market


EE, 1,112

O2, 881

Vod, 620

three, 146

HHI = 2759

EE, 842

O2, 667

Vod, 469

three, 110

MVNOs, 169

HHI = 2257

EE, 1,103

O2, 762

Vod, 729

three, 148

HHI = 2743

EE, 934

O2, 645

Vod, 617

three, 125 MVNOs,

64

HHI = 2386


154

Exhibit 11.6: UK mobile market structure – MVNOs have a significant presence


Three

EE, Orange and T-

Mobile branded retail stores, and online

Direct

e.g. Unify Mobile

ShebangStream Comms.

12

MVNOs

e.g. CPWPhones 4UPC World

Reseller/Retail

Three branded retail stores and online

Direct

EE

e.g. Virgin MobileUnicom

Axis, Catalyst

Est. 20-25

MVNOs

e.g. CPWPhones 4U

Reseller/

Retail

Vodafone

O2

e.g. BT Mobile

LebaraAsda

Est. 10-15 MVNOs

e.g. CPWPhones 4U

Dixons

Reseller/

Retail

Vodafone branded

retail stores and onlineDirect

e.g. TescoLyca

Manx

Est. 10-15

MVNOs

e.g. CPW

Phones 4UTesco

Reseller/

Retail

O2 branded retail stores and online

Direct


C

u

s

t

o

m

e

r

s


155

The UK market also has a well-developed independent retailer channel via outlets such as Carphone Warehouse and Phones 4U (see Exhibit 11.6). In fact, in some high streets, it is possible for the consumer to choose between a wide variety of separately branded mobile service providers (eg VOD, O2, 3, EE, Orange, T-Mobile, Virgin Mobile, Lebara, Lyca, Tesco, Asda), or to choose to buy one of the main network services from a wide selection of retail outlets (Carphone Warehouse, Phones 4U, Currys/PC World). In fact, in some of the UK’s more depressed high streets, it seems that the mobile phone industry, along with a few banks, building societies and charity shops, are responsible for keeping the high street alive.

Judgement: market concentration, MVNO presence and consumer choice: We think the UK already demonstrates a relatively high level of fragmentation and a wide spread of consumer choice, with a well-developed MVNO segment in the market. This suggests that in the event of consolidation, the UK faces a relatively low risk from remedies to strengthen the MVNO segment or increase consumer choice.


The UK market has the second-most-fragmented spectrum distribution of the peer group, when the spectrum HHI calculation is adjusted for the relatively higher value of sub-1Ghz spectrum. We calculate an adjusted spectrum HHI of 2,553, only slightly higher (more concentrated) than Germany, but suggesting a more fragmented spectrum ownership than the average of our peer group at 2,717.

The UK’s relatively more fragmented spectrum position reflects that 1) all four operators own the more-valuable sub-1Ghz spectrum, 2) EE owns a significant amount of 1,800Mhz spectrum, and 3) 3 also now owns a good some 1,800Mhz following the regulatory-enforced divestment by EE of 1,800Mhz spectrum arising out of the merger of Orange and EE.

O2’s apparent under-representation in the spectrum stakes (unadjusted) arises due to its lack of 2,600Mhz spectrum relatively to VOD and EE. However, under our adjusted calculation, O2’s strong position in sub-1Ghz spectrum brings O2’s share of adjusted spectrum back up to 25.4%, much closer to EE’s 29.7% and VOD’s 28.7%.


156

Exhibit 11.7: Market split by subscribers, service revenues and spectrum Illustrates 3’s apparent disadvantageous position

Source: Company reports; Ofcom Note: Spectrum share excludes BT which has 2.2% of adjusted spectrum via its ownership of 2.6Ghz frequencies

Exhibit 11.8: Operator spectrum portfolios show that all four MNOs have at least some sub-1Ghz spectrum. Three still appears relatively disadvantaged, though

Source: Company reports; Ofcom

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

EE O2 Vod three


Mhz 800 900 1800 2100 2600 Total Share HHI

EE 10.0 0.0 90.0 50.0 70.0 220.0 36.4% 1,327

Vodafone 20.0 35.0 12.0 30.0 65.0 162.0 26.8% 719

O2 20.0 35.0 12.0 25.0 0.0 92.0 15.2% 232

H3G 10.0 0.0 30.0 35.0 0.0 75.0 12.4% 154

BT 0.0 0.0 0.0 0.0 55.0 55.0 9.1% 83

Total 60.0 70.0 144.0 140.0 190.0 604.0 2,515

Spectrum Multiples 1 1 2 2.5 9

Adjusted Mhz 800 900 1800 2100 2600 Total Share HHI

EE 10.0 0.0 45.0 20.0 7.8 82.8 29.7% 880

Vodafone 20.0 35.0 6.0 12.0 7.2 80.2 28.7% 826

O2 20.0 35.0 6.0 10.0 0.0 71.0 25.4% 647

H3G 10.0 0.0 15.0 14.0 0.0 39.0 14.0% 195

BT 0.0 0.0 0.0 0.0 6.1 6.1 2.2% 5

Total (adj) 60.0 70.0 72.0 56.0 21.1 279.1 2,553


157



Exhibit 11.10: Spectrum concentration and adjusted spectrum concentration – with consolidation


Exhibit 11.11: VOD and 3 would result in the lowest increase in HHI

Source: Company reports; Ofcom

In the event of consolidation (assuming VOD and 3, or O2 and 3), and comparing the consolidated UK to the peer group markets without consolidation, the UK would move from a relatively fragmented position to an above-average service

EE, 1,327

Vodafone, 719

O2, 232

H3G, 154

BT, 83

HHI = 2515

EE, 880

Vodafone, 826

O2, 647

H3G, 195

BT, 5

HHI = 2553

EE, 1,326.7

Vodafone, 719.4

O2+H3G, 764.5

BT, 82.9

HHI = 2893

EE, 879.6

Vodafone, 826.1

O2+H3G, 1,553.2

BT, 4.8

HHI = 3264

HHI

Subscriber

market share

(MVNOs


Service

revenue

market share

(MVNOs


Current 2257 2386

Vod+3 2712 455 2942 556

O2+3 2799 542 2954 569

EE+three 2866 609 3070 684

O2+Vod 3376 1119 3648 1262

EE+Vod 3515 1257 3904 1518

EE+O2 3756 1499 3938 1553


158

revenue concentration, with a HHI of 2,942 (VOD and 3) against the peer group average of 2,717. The UK would still appear less concentrated than the Netherlands and Austria, and in line with Spain. However, the UK’s spectrum concentration would leap from 2,553 to 3,264, leaving it as the most concentrated of the peer group markets in terms of spectrum.

If instead we compare a consolidated UK market to consolidated peer group markets, the UK actually comes out as the best positioned market when taking into account both service revenue concentration and spectrum concentration. Other markets may appear either in line with the UK, or more fragmented on one of these two parameters (eg Germany remains the most fragmented and France would just pip the UK on service revenue HHI, while Austria, after the Orange and Hutchison merger but before spectrum remedies, has a similar spectrum fragmentation to the UK), but the UK has the best combination of spectrum and service revenues HHI.

Ofcom, the UK regulator has already laid out clear guidelines imposing spectrum caps on UK operators with an intention of protecting the fourth operator (3) during the 4G spectrum auctions and retaining a minimum of four network-based operators in the UK market. As such, our own analysis and conclusions simply confirm what Ofcom is already imposing – in the unlikely event of consolidation, spectrum divestments would be required.

Judgement: spectrum share versus customer share: The UK appears among the most fragmented markets of the peer group, suggesting perhaps a greater amount of flexibility to consolidate relative to the peer group. Spectrum remedies would likely still be required, but the UK stands out as requiring among the lowest amount of spectrum remedies compared to the peer group.

We could also not rule out the prospect that a new entrant, or perhaps BT (which recently acquired 2,600Mhz spectrum in the 4G auctions) might be interested in picking up divestment spectrum with a view to rolling out a network of its own.

Exhibit 11.12: ARPU per month by operator – signs of stabilisation lately


2005 2006 2007 2008 2009 2010 2011 2012

Vodafone 23.8 23.5 22.5 21.5 20.4 21.7 21.4 20.3

O2 22.8 23.0 24.2 23.5 21.7 21.6 19.6 18.6

OrangeTmobile 20.6 21.2 21.8 22.2 20.0 19.7 18.9 19.0

3 mobile 34.5 46.6 43.4 33.6 26.8 22.6 21.9 20.5

Average 25.4 28.6 28.0 25.2 22.2 21.4 20.4 19.6

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Op

era

tor

AR

PU

£p

cm


159

Exhibit 11.13: Typical mobile tariffs packages in the UK



The UK market is a low-priced market with a well-developed range of big bundles for consumers to choose from. Mobile ARPUs have declined from the mid-£25 pcm level several years back, to current levels of just below £20 pcm.

Comparing typical limited-use SIM-only offers and aiming for a usage profile of about 100mpm of voice, 200-300 SMS, and 250-300Mb of data, we found a wide range of offers in the UK (see Exhibit 11.13), which on average left the UK below average price levels for the peer group. We stress again the difficulty of comparing prices across borders due to different pricing strategies and target usage profiles. For example, in the UK, we were typically unable to find tariffs offering limited volumes of SMS – most offers bundles much larger, or unlimited, volumes of SMS.

Looking at typical “unlimited” big bundle offers, and aiming for a usage profile of unlimited any-network voice and SMS, and 1Gb of data, we again found the UK to be very competitive, offering typically below peer-group average packages with a wide selection of prices on offer. At this level, we think the UK appears third in the value stakes, behind France and Austria, but ahead of Denmark and below the average for the peer group.

When we add an iPhone into the typical “unlimited” tariff, the UK continues to look good value, improving its rank to second place as France slips down the list. This might suggest that the UK has slightly higher subsidies for iPhone handsets than peer markets, or that the UK is better able to exploit scale in procurement.

While the UK appears good value for post-paid SIM-Only and iPhone offers, the picture is reversed for pre-paid tariffs, where the UK seems relatively expensive on pre-paid price per minute for voice usage, and per message for SMS. Again, we

Tariff

Contract

period

H/s

upfront

cost Mou SMS Data WiFi

1-time

fees £pcm

Amort

ised

cost

pcm £ FX

Amort

ised

cost

€pcm

Typical limited SIM-only

3 Essential internet 12 0 200 5000 500 0 0 6.90 6.90 1.172 8.08

CPW talkmobile £8 deal 12 0 100 200 1000 0 -9.6 8.00 7.2 1.172 8.43

Tesco Sim Only £7.50 12 0 250 5000 500 0 0 7.50 7.5 1.172 8.79

TMO TMO Sim only 12 0 250 500 500 Tube wifi 0 8.00 8.00 1.172 9.37

CPW Vod RED with 10% cashback 12 0 300 u/l 250 0 -20 10.50 8.83 1.172 10.35

Vod Vodafone Red £10.50 12 0 300 u/l 250 750 0 10.50 10.50 1.172 12.30

O2 O2 Simplicity £11.00 12 0 300 u/l 250 0 0 11.00 11.00 1.172 12.89

Ora Orange Dolphin 12 0 200 u/l 250 0 0 15.50 15.50 1.172 18.16

EE na na na na na na na na na na na na

Typical unlimited SIM-only

3 The One Plan 12 0 2000 5000 2000 0 -25 20.00 17.92 1.172 20.99

CPW 3 plan with 10% discount 12 0 2000 5000 2000 0 -24 20.00 18.00 1.172 21.09

Tesco £20pcm tariff 12 0 3000 5000 2000 0 0 20.00 20.00 1.172 23.43

Vod Vodafone Red £20.50 12 0 u/l u/l 1000 750 0 20.50 20.50 1.172 24.02

O2 O2 Simplicity £21.00 12 0 u/l u/l 1000 0 0 21.00 21.00 1.172 24.60

TMO TMO Full Monty 12 0 u/l u/l u/l Tube wifi 0 26.00 26.00 1.172 30.46

EE EE 1Gb unlimited plan 12 0 u/l u/l 1000 0 0 26.00 26.00 1.172 30.46

Ora Orange na na na na na na na na na na na

Typical unlimited SIM-only+16Gb iphn 5

CPW Vod or Orange via CPW 24 0 u/l u/l 1000 0 0 37.00 37.00 1.172 43.35

3 The One Plan 24 29.00 2000 5000 u/l 0 0 36.00 37.21 1.172 43.59

Vod Vodafone Red 24 59.00 u/l u/l 1000 2000 0 37.00 39.46 1.172 46.23

O2 O2 iphone (most popular tariff) 24 0 u/l u/l 1000 0 0 42.00 42.00 1.172 49.20

EE EE 1Gb unlimited plan 24 29.99 u/l u/l 1000 0 0 41.00 42.25 1.172 49.50

TMO TMO Full Monty 42 24 29.99 u/l u/l u/l Tube wifi 0 42.00 43.25 1.172 50.67

Tesco £20pcm tariff 24 0 3000 5000 2000 0 0 45.00 45.00 1.172 52.72


160

stress the difficulty of comparing tariffs across borders and across widely varying tariff structures, but overall we think our analysis gives a good impression of the broad scope of pricing and relative cost across our peer group markets.

Judgement: low prices and big bundles: If consolidation can result in a firming of price levels and a slowing in the pace of ARPU decline, then the UK market would appear relatively well positioned to benefit from consolidation.


Exhibit 11.14: Post-paid share of base by operator – the UK has a relatively underdeveloped post-paid segment versus the peer group


Exhibit 11.15: UK network-sharing arrangements are currently limited

Source: Company

The UK market appears to have a relatively underdeveloped post-paid segment (see Exhibit 11.14), ranking sixth out of our eight peer group markets in terms of post-paid subscribers as a proportion of the total. However, the mix-shift from pre-paid to post-paid is among the fastest of the peer group, which is encouraging

0

10

20

30

40

50

60

70

80

90



Date Country Operators Sites Back-haul Full RAN Spec-trum Core n/w Aggregate cost saving identified Comments/rationale

Nov-08 UK T-Mob/3 Y Y Y £2bn (10 yrs) in total for both partners Merging 3G networks, 5k sites (out of 18k)

closed. Sharing agreement was retained as

a regulatory requirement for the merger of

Orange and T-Mobile to form EE.

Apr-09 UK* TEF/VOD Y VOD: 10% of opex Joint build;share 2G/3G; economic

situation was a catalyst: • UK: focus on joint build , consolidation

of 2G/3G sites

Feb-10 UK Orange/ T-

Mob

Y Y Y Y Y £154m annual opex savings by year 4

(25% of network opex, 4% of total Opex),

and £100mln pa capex savings (20-25%)

by year 4. £700mln+ capex savings over 10

years.

Resulting from the merger of Orange and T-

Mobile to form Everything Everywhere,

citing opex and capex savings through

joint network operations. Aim to reduce

site count from 28k to 18k+.

Jun-12 UK TEF/VOD Y Y Y n/a 40% increase in coverage to match leader

EE; accelerates 4G coverage.

Network sharing


161

and reflects efforts of operators such as EE to increase the proportion of post-paid subscribers in the mix. At present, the UK ranks above Germany and Italy (which we count as a special case due to taxes on post-paid subscriptions).

Judgement: pre-paid versus post-paid mix: If in-market consolidation can act as a catalyst to accelerate the mix-shift towards more post-paid subscriptions, then the UK appears relatively well positioned to benefit.

Network-sharing

Network-sharing in the UK is relatively well developed, with a number of sharing arrangements in place between the four main MNOs (see Exhibit 11.15). VOD and TEL have an extensive site-sharing arrangement, and agreed a further backhaul and RAN-sharing arrangement in 2012. Orange and T-Mobile are combining networks under the merger which formed EE, while the existing network access arrangement between T-Mobile and 3 was secured as a remedy by the EC in order to grant approval for the EE merger.

Judgement: network-sharing: If consolidation could result in remedies to extend network-sharing or increase network access for smaller players, then the UK would seem relatively well positioned to experience relatively light network remedies in the event of consolidation.


All of the above sounds pretty good for the UK in the event of consolidation being proposed. However, there is one main stumbling block: while the UK has already seen consolidation from five to four MNOs with the Orange and T-Mobile merger, the UK regulator Ofcom has made clear, in its 4G auction consultation, its desire to see a minimum of four network-based operators in the UK.

Jurisdiction in any UK consolidation move would likely fall with the EC (which claimed jurisdiction over the EE merger), and Ofcom’s view would likely be taken into account, albeit would not be binding on the anti-trust authorities. Even so, we would be surprised if operators moved to test Ofcom’s clearly defined boundary by putting proposals to the EC.

If such a step were taken, then operators would also have to take into account Ofcom’s clear guidelines on spectrum caps which would likely result in spectrum divestments as discussed earlier.

Judgement: regulatory attitudes: While the UK appears to be among the best positioned markets to benefit from consolidation and to counter any remedies that may be imposed, a move to consolidation seems unlikely without a change of regulatory attitudes.


162

Spain: unlikely to see mobile consolidation


We think the Spanish mobile market faces challenges to consolidate in the near term. The market looks more concentrated than peer markets in terms of service revenue market share, and also looks modestly more concentrated in terms of spectrum ownership. Concentration is less of an issue when we look at subscriber share, but this reflects the relative underdevelopment of MVNOs in the Spanish market, where despite taking about 10% subscriber share to date, this only translates to 3% service revenue share for the MVNOs. In the future, market concentration as measured by service revenues may converge with that as measured by subscriber share, and that will result in the Spanish mobile market looking less concentrated that it appears today. However, we think anti-trust authorities are more likely to consider the state of the market at the time of any proposed consolidation, rather than at some distant point in the future.

The market is also relatively highly priced, unless convergence offers are taken into account, and is underdeveloped in terms of infrastructure sharing and access. In short, the market is susceptible to significant remedies in the event of consolidation, although the most obvious pairing of Orange and Yoigo only results in an above-average concentration of service revenue share, while the resulting spectrum concentration would be modestly below the average of peer group consolidation scenarios. A key risk for Spanish consolidation is the prospect that strong MVNO participants such as JAZ and ONO could find themselves gaining access to remedy spectrum.

TLSN recently tried to sell Yoigo – Spain’s fourth-largest mobile network – but had to halt the sale process due to the low amounts offered. This suggests to us that MNOs are aware of the material remedies that could result from acquiring Yoigo, which could provide MVNOs with fixed-line infrastructure, and an opportunity to acquire spectrum licences and develop further their mobile business. In our view, proposed remedies could be sufficient to challenge potential synergies from Yoigo’s consolidation.

Nonetheless, it is worth highlighting that the recent move to convergence offers could change the dynamics of both mobile and wireline markets, and the regulator’s view on “market definition”. With a combined offer to address both markets, we could see M&A between mobile and fixed-line operators in the long run, given that (i) the four main MNOs are not the four biggest broadband players, and (ii) synergies could be more attractive once MVNOs with fixed-line assets gain material mobile market share.


A combination between Yoigo and Orange or Yoigo and VOD could be, in our view, the most acceptable mobile-to-mobile combinations from a regulatory perspective. Any combination of Orange or VOD with Yoigo results in an HHI index of lower than 3,295 (compared to an HHI of over 3500 for a TEF/Yoigo combination) and would leave the combined businesses with a 34% service revenue share as a new number two in the Spanish market. However, it appears that neither Orange nor VOD were willing to pay the price requested by TLSN, given the likelihood of material remedies.


163

In the long term, we could expect some cross-consolidation between wireline and wireless operators. ONO or JAZ could also be interested in Yoigo mobile customers and spectrum assets if they are both successful in growing in their MVNO subscriber base.


In the case of a merger between mobile MNOs, the most likely remedies would include, in our view, the release of spectrum and granting of access to mobile network infrastructure (mobile base stations) and capacity, potentially to one of the already established MVNOs.

This could raise significant risks for MNOs, given the current high pricing regime in the high-value market. Most wireline operators with MVNO contracts generate barely any profit from mobile, but the combination of mobile and fixed-line assets could add another dimension of competition in the market.

Market evolution – convergence the key market trend

The Spanish mobile market was c124% penetrated at end-Q1 2013. The market for mobile connections dropped materially throughout 2012 and faced the first decline in customer growth since TEF’s full privatisation in 1997, reflecting the significant drop in the second-SIM card market and the increasing disconnection of unused SIM cards by operators. While unused SIMs accounted around 2m disconnections, 3.1m disconnections were the result of the weak operating environment.

The service revenue picture (see Exhibit 12.1) portrays better the worsening Spanish mobile telecoms service revenue trend since 2007, a deterioration that was due to the weak macroeconomic backdrop, low consumer confidence, high unemployment and relatively high retail price levels. Spain is experiencing severe economic problems: the country is the worst-performing in the EU on many socioeconomic indicators, with all-time highs recorded in the unemployment and poverty rates.

Consumers’ wallets are under considerable strain and the economic backdrop remains weak. As Exhibit 12.1 illustrates, mobile service revenues declined by 5.5% pa from 2009 to 2011, but the rate of decline increased notably in 2012, accelerating to -11.3% pa. In Q1 2013, we estimate the Spanish market saw service revenues decline by c15% yoy, with MTRs alone accounting for c4ppt of the decline.

The mobile market is dominated by four major MNOs – which combined own 90% of the subscriber base and 97% of mobile service revenues – and about 40 MVNOs. In terms of network operators, TEF is the incumbent and market leader in terms of subscribers and service revenue share, followed by VOD and Orange, then Yoigo, the most recent MNO entrant in the Spanish market, which was established in 2006. In terms of virtual network operators, the market leaders are Euskaltel, JAZ and ONO, which combined account for more than 50% of MVNO service revenues.


164

Exhibit 12.1: Spanish service revenues continue to decline sharply

Source: Berenberg research

Exhibit 12.2: Mobile market lost 5.11m SIMs in 2012, of which 2m were inactive disconnections


TEF has been the main market share loser over the last five years. Service revenues have shrunk market-wide by €4.8bn since 2007 (of which we estimate around 27% of the reduction is due to mobile termination rate cuts), with TEF losing €3.1bn – 65% of the overall decline. VOD is the second-biggest loser in terms of mobile revenues, registering a loss of €2.1bn since 2007. While Orange has more or less maintained its service revenues flat over the same period, Yoigo – the new entrant – has been the key share gainer, and the only operator to grow its service revenues, from €299m in 2009 to €679m in 2012, a performance which reflects its competitive pricing and its ability to attract subscribers – a total of 3.28m over the last five years.

2005 2006 2007 2008 2009 2010 2011 2012

Yoigo 0 0 0 299 467 624 679

Vodafone 5,097 5,798 6,382 6,259 5,715 5,285 4,777 4,202

Telefónica 7,769 8,163 8,586 8,477 7,833 7,299 6,649 5,462

Orange 3,177 3,264 3,299 3,276 3,171 2,972 3,108 3,098

Market growth 9.8% 7.4% 6.0% -1.4% -5.5% -5.8% -5.4% -11.3%

-15.0%

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

20,000

Mo

bil

e se

rvic

e re

ven

ue

(€m

)

2005 2006 2007 2008 2009 2010 2011 2012

Yoigo 0.00 0.02 0.43 0.97 1.51 2.28 3.04 3.71

Vodafone 12.92 14.46 15.81 16.54 16.91 17.48 17.65 15.34

Telefónica 19.89 21.45 22.83 23.60 23.54 24.31 24.17 20.53

Orange 10.30 11.11 11.09 11.37 11.88 11.30 11.66 11.84

Market growth 10.1% 9.1% 6.6% 4.7% 2.6% 2.9% 2.1% -9.0%

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

0.00

10.00

20.00

30.00

40.00

50.00

60.00

Mo

bil

e s

ub

scri

bers

(m

)


165

Exhibit 12.3: TEF and VOD have suffered the most from the weak macro backdrop and competitive environment (€m)


TEF has taken some severe measures to address its domestic business slump and recover leadership in Spain. A key initiative in its bid to regain commercial leadership was the launch of its fixed mobile convergence offers under the Movistar Fusion banner, on 1 October 2012, coupled with reduced handset subsidies. These new fixed/broadband/TV/mobile service triple-/quad-play bundle products offered a 50-70% discount to separate packages when introduced to the market. All the other operators reacted immediately, launching new convergence packages and increasing their cross-selling discounts. Based on Q1 2013 results figures, the Spanish market contains over 3m convergence customers, with TEF leading the way with 1.7m. Clearly, these integrated fixed mobile packages have reshaped the Spanish telecoms market’s structure. Since Q4 2012, convergence has become the key market trend.

In summary, the Spanish market ranks as one of the most challenged. While we expect the pace of service revenue decline to improve, we still forecast that Spain’s mobile market will suffer one of the steepest declines compared to the eight markets we addressed in this report, with a three-year CAGR 2013-16 of -5%. Clearly, the opportunity cost of not seeing in-market consolidation in the Spanish market is one of the highest in the EU.

Market concentration – despite a high service revenue HHI, competition remains intense due to convergence

The national Spanish mobile market (including MVNOs) is very competitive at the subscriber market share level. Despite the entry of Yoigo in 2006, and only taking into account the market share of the MNOs (with MVNO and service provider subscribers consolidated within the MNO network subscribers figure), the market share HHI index equates to 3,045. Nevertheless, the picture changes significantly when we account for the MVNOs subscribers as a separate entity from MNOs: the HHI index drops by almost 500 points. Note that the Federal Communications Commission in its 15th annual competition report (published in June 2011) described a 250-point increase in HHI as an enhancement in market power regardless of the initial HHI – so it would appear that a 500-point drop (including the MVNOs’ share) is a clear indication of the high level of competition in the Spanish mobile market.

3,177 3,264 3,299 3,276 3,171 2,972 3,108 3,098

7,7698,163

8,586 8,4777,833

7,2996,649

5,4625,097

5,7986,382 6,259

5,7155,285

4,7774,202

299 467 624 6790

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

2005 2006 2007 2008 2009 2010 2011 2012

Orange Telefónica Vodafone Yoigo


166

Exhibit 12.4: Spanish MNO subscriber HHI at 3,045

Exhibit 12.5: Spanish MNO/MVNO subscriber HHI at 2,555


Exhibit 12.6: Spanish MNO service revenue HHI at 3,218

Exhibit 12.7: Spanish MNO/MVNO service revenue HHI at 3,018


Convergence offers are reshaping the mobile market share structure, increasing the competition framework and the likelihood of M&A in the long term

Interestingly, the service revenue share HHI analysis suggests a different conclusion to the subscriber share HHI analysis. The service revenue share HHI excluding MVNOs amounts to 3,218; and only falls to 3,018 when MVNOs are included. While MVNOs have a 10% share of the mobile subscriber market, they only account for 3.5% of mobile service revenues. On one hand, this suggests that MVNOs are attracting low-value customers on discounted tariffs. On the other, the gap between the subscriber HHI and service revenue HHI, in our view, could reflect the market’s recent move towards convergence offers in Q4 2012. MVNOs with broadband infrastructure were able to increase their mobile net additions significantly, gaining mobile market share by upselling mobile to their broadband subscribers, but the strong uptake has yet to be reflected in these companies’ respective mobile service revenue figures.

Orange, 583

Telefónica, 1592

Vodafone, 815

Yoigo, 55

Orange, 475

Telefónica, 1270

Vodafone, 662

Yoigo, 45MVNOs,

102

Orange, 649

Telefónica, 1897

Vodafone, 640

Yoigo, 32

Orange, 594

Telefónica, 1806

Vodafone, 573

Yoigo, 32 MVNOs, 12


167

Exhibit 12.8: Orange and VOD could acquire Yoigo


The service revenue HHI picture could thus start to converge with the subscriber HHI in the longer term given that the current trends suggest a further material gain of market share by those MVNOs with wireline assets, such us JAZ and ONO – respectively the fourth- and third-largest fixed broadband operators in Spain – if they take advantage of convergence opportunities to strengthen their mobile business.

If there was consolidation, Yoigo and Orange, or Yoigo and VOD, would have the least impact on market concentration: TLSN was on the verge of selling Yoigo in 2013; however, the company halted the deal due to lower-than-expected bids. It is hard to deny that Spain looks relatively concentrated, and we understand why consolidation of Yoigo has been difficult to conclude. However, if consolidation was to occur among mobile operators, the combination of Yoigo and VOD or Yoigo and Orange seems to be the more acceptable ones based on an HHI analysis.

In Exhibit 12.8, we show the impact of the various operator combination on service revenue HHI. Note that any combination in the Spanish market would fall under EC scrutiny rather than that of the national regulator, given that any combination would have worldwide turnover over €5bn, with each of at least two of the merging parties realising over €250m annual turnover in the EU.

Spectrum share versus customer share As illustrated in Exhibit 12.14, VOD, Orange and Yoigo are left with excess spectrum relative to their market share, while TEF appears to be fully using its spectrum assets.

The spectrum map tells us much about the operators’ network deployment strategies and development plans. Each of the three historical MNOs – TEF, VOD and Orange – recently acquired spectrum in the 800Mhz and the 2,600MHz bandwidths for their LTE strategy, to complement the existing nationwide GSM 900/1,800MHz and UMTS 2,100 MHz networks.

Exhibit 12.9: Swedish spectrum split with spectrum amounts adjusted for coverage

Source: Berenberg

HHI

Subcriber market share (MVNOs

separated out) change

Service revenue market share

(MVNOs separated out) change

Current 2555 3018

TEF+Yoigo 3033 478 3501 483

Orange+Yoigo 2847 293 3295 277

VOD+Yoigo 2900 345 3290 272

VOD+Orange 3677 1122 4184 1166

Subscribers

Service

revenues

Unadj.

Spectrum Subscribers

Service

revenues

Unadj.

Spectrum

Telefonica 39.9% 43.6% 28.5% 1592 1897 815

Vodafone 28.6% 25.3% 31.1% 815 640 966

Orange 24.2% 25.5% 29.0% 583 649 839

Yoigo 7.4% 5.7% 11.4% 55 32 130

Total 3045 3218 2750

Market share HHI


168

Exhibit 12.10: Spanish spectrum split with spectrum amounts adjusted for coverage

Source: Berenberg

Yoigo preferred to develop smart hybrid multiband solutions using a spare 10MHz of free 1,800MHz spectrum to deploy the LTE 1,800MHz networks.

With a 5.7% service revenue market share and a relatively high spectrum capacity, Yoigo still has room to chase more market share. Orange and VOD could also be more aggressive given their spare network capacity.

When we look at the absolute spectrum HHI at 2,750, it is around 300 points and 500 points lower than the subscriber share and service revenue HHI at 3,045 and 3,218 respectively. The absolute spectrum amounts do not take into account the quality and the physical features of the different spectrum bands, hence we calculate an adjusted spectrum holding. According to Ofcom, 1MHz of 800MHz/900MHz spectrum is able to cover 2x the area of 1,800MHz spectrum, 2.5x the area of 2,100MHz spectrum and 9x the area of 2,600MHz spectrum.

Using our adjusted spectrum calculation does not significantly increase the spectrum HHI due to the broadly equal balanced spectrum among the three largest MNOs.

Implications for consolidation

We highlighted earlier that from a regulatory perspective the two deals that would result in the smallest increase in market concentration are the ones between Yoigo and VOD or Yoigo and Orange. The question is: would these consolidations spark any spectrum divestments?

We run different spectrum combinations in Exhibit 12.11, showing the impact on spectrum HHI if Yoigo joined forces with Orange or with VOD. Clearly the spectrum HHI will increase materially if Yoigo were to merge with any of the other operators. In our view, this will drive spectrum divestments in both 2G/3G and 4G.

Exhibit 12.11: Spectrum HHI will increase by around 610 if a merger occurs

Source: Berenberg

Subscribers

Service

revenues

Adj.

Spectrum Subscribers

Service

revenues

Adj.

Spectrum

Telefonica 39.9% 43.6% 31.0% 1592 1897 963

Vodafone 28.6% 25.3% 29.8% 815 640 889

Orange 24.2% 25.5% 28.7% 583 649 823

Yoigo 7.4% 5.7% 10.5% 55 32 109

Total 3045 3218 2785

Market share HHI

HHI

Unadj. Spectrum market

share change

Adj. Spectrum market

share change

Current 2750 2785

TEF+Yoigo 3401 652 3434 649

Orange+Yoigo 3411 661 3385 600

VOD+Yoigo 3459 709 3408 624

VOD+Orange 4550 1800 4496 1711


169

As shown above, the spectrum HHI will move by around 610 points should either of the more likely cases occur (Yoigo/VOD and Yoigo/Orange). As a result, Yoigo’s spectrum in the 900Mhz and 2,100Mhz bandwidths could be auctioned or given to a new entrant, along with some deployment obligations. The potential buyers, in our view, could be those current MVNOs with strong wireline assets (ONO or JAZ), which might acquire in order to leverage their mobile base.

Low prices and big bundles?

Significant competitive tension triggered by convergence The Spanish market, which used to feature relatively high prices, has seen material cuts in mobile pricing since the introduction of Movistar Fusion and convergence offers. The significant discount on convergence packages has weighed heavily on mobile pricing, primarily the high-end bundles that have seen cuts of more than 40%. Movistar (TEF) is now the mobile operator that offers the least choice of mobile tariffs in Spain as the company totally changed its contract tariff portfolio and offers only two mobile contract types: Movistar Cero (a low to mid-usage bundle, voice pay-as-you-go, 500 SMS and 500MB) and Movistar Total (unlimited voice, 1,000 SMS and 1GB).

Today, Movistar is aiming to simplify its offerings and focus on two segments; VOD uses BASE tariffs to address the low- to mid-value market and RED tariffs to address the mid- to high-value market; Orange is targeting all types of customers; and Yoigo is focusing on heavy data users. However, the battleground, in our view, will remain in the convergence bundles arena.

Our analysis in Exhibits 12.13 to 12.15 compares pricing for three profiles in Spain: (i) low usage profile/entry level (100min, 500 SMS and 200MB of data); (ii) unlimited voice SMS and 1Gb data/high end level; (iii) an unlimited bundle with an iPhone 5 16GB.

As is evident from these exhibits, Spain takes a different approach to other European markets in terms of mobile pricing. While most EU markets have moved to all-you-can-eat voice and SMS bundles with subscribers paying for data, Spanish operators bundle high data usage volumes, even in their lower priced bundles, with subscribers instead paying for voice usage. This pricing structure represents a better fit with Spanish consumer demand. Spain has one of the highest penetration rates of OTT applications (such as Whatsapp). This further reduces customers’ need for SMS and has led to increased consumer demand for data. According to Spanish regulator CMT, contract mobile subscribers use around 155 minutes of voice per month while pre-pay customers need no more than 40 minutes on average. In addition, SMS usage in Spain is among the lowest in Europe at around 10 SMS per month.

Entry-level bundles aim is to leverage voice usage The non-inclusion of minute usage at the low-/mid-end user level is a specific feature of Spanish mobile operators and reflects consumer consumption habits. Spain is one of the few markets in which customers are charged a set-up call fee. However, most bundles do not charge customers for minute usage, so the longer the call, the cheaper its cost per minute. By contrast, most MNO bundles include at least 500MB data usage and 500 SMS for a monthly cost of around €10/month (excluding VAT).


170

Exhibit 12.13: Low bundles include high data usage


Exhibit 12.14: “Big” bundles remain relatively costly compared to EU markets


Exhibit 12.15: Most MVNOs do not offer an iPhone 5


In the low-/mid-value market, MVNO Euskaltel appears to have the best-fitting proposition to our low-usage profile. We estimate users’ monthly cost at €11.50 (excluding VAT) for 120 min, 20 SMS and 300MB, while Movistar’s lowest offer is €16.5o (excluding VAT) for 0 min, 500 SMS and 500MB. Yoigo prices its offers differently by proposing a high volume data packages with pay-as-you-go voice and SMS.

All rates are in € Movistar Vodafone Orange Yoigo Euskaltel

SIM-Only, 100mpm, 500 sms, 200 MB

Package name Moistar Cero Base 2 Squirrel 7 La Del UnoFree Flat rate 10

Voice calls (in min) 0 100 0 0 120

Price per extra min

mobile 0 0 0.01 0.01 na

fixed 0 0 0.01 0.01 na

connection charge per call 0.15 0.15 0.15 0.15 na

Data Mb pcm 500 1000 500 1000 300

Price per extra MB na na na na na

SMS 500 1000 1000 0 0

Price per extra SMS 0.15 0.15 0.1 0.08

Subscription terms (months) 18 24 12 0 0

One time charge 0 0 0 0 0

Monthly Flat rate tariff (base, excl VAT) 9 15 10 9 9.9

Monthly Flat rate tariff (adj for usage) 16.5 15 18.5 18.5 11.5

Monthly Flat rate tariff (promo, incl VAT) 20.0 17.0 22.4 22.4 13.9


SIM-Only, unlimited voice, SMS and 500MB

Package name Moistar total RED Dolphin 35La Infinite 30Free Flat Rate 20

Voice calls (in min) unlimited unlimited unlimited unlimited 500

Data Mb pcm 1GB 1.5GB 1.5GB 1GB 1GB


SMS 1000 unlimited 1000 0.1/SMS 0.08/SMS

Subscription terms (months) 18 24 12 0 0

Monthly Flat rate tariff (base) 35 35 35 30 19.9

Monthly Flat rate tariff (promo, incl VAT) 42.4 39.7 38.1 42.4 24.1


SIM-Only, unlimited voice, SMS and 500MB with an Iphone 5 16GB

Package name Moistar total RED Dolphin 35La Infinite 30 na

Voice calls (in min) unlimited unlimited unlimited unlimited na

Data Mb pcm 1GB 1.5GB 1.5GB 1.5GB na


SMS 1000 unlimited 1000 0.1/SMS na

Subscription terms (months) 18 24 24 24 na

Handset cost 597 579 569 600 na

One time charge 597 99 0 669 na

Monthly charge 0 20 25 25 na

Monthly Flat rate tariff (base) 72.5 62.4 66.8 72.6 na

Monthly Flat rate tariff (promo, incl VAT) 72.5 62.4 66.8 72.6 na


171

Our conclusion from this analysis is that MNOs include high data usage in their low-end bundles to satisfy mid-value customers’ needs and to generate high out-of-bundle ARPU from voice, while MVNOs introduce more appropriate offers to target low-value customers with no need for high-value usage.

At the high-end level, competition has been tough over the last year, resulting in a material drop in prices – but prices remain relatively high compared to EU markets At the high-end with an unlimited plan, all MNOs offer approximately the same usage for the same €35 cost (excluding promotions). Note that most operators cut their prices for the high-end market after the launch of convergence offers. However, both VOD and Orange currently give a 25% six-month discount on their unlimited plan tariffs, which make them 10% cheaper than their competitors.

We struggle to find any MVNOs that offer unlimited voice in their high-value bundles. Indeed, most virtual network operators only include 500 minutes in their high-end bundles. We believe MVNOs in Spain prefer to focus on low- to mid-value customers rather than on high-value users. Yoigo uses a different strategy compared with its peers, preferring to position its offers at a lower price point by excluding SMS usage and only providing unlimited voice and data.

Exhibit 12.16: MNO do not subsidise iPhone 5


0.0

5.0

10.0

15.0

20.0

25.0

30.0

Movistar Vodafone Orange Yoigo

Iphone Handset ARPU


172

iPhone subsidies remain low The weak macroeconomic backdrop forced TEF to lower handset subsidies in 2012, and it was immediately followed by VOD and Yoigo. Many Spanish MVNOs do not offer the iPhone 5 as they do not want to commit to certain sales terms with Apple. Exhibit 12.16 shows the ARPU generated from handset sales, which clearly demonstrates that, apart from VOD, most Spanish operators currently do not subsidise handset, as they are offering iPhone 5s at approximately the same price as Apple’s retail shops.

Conclusion: there is room for a disruptive operator in the high-end market, given the relatively high pricing point From our analysis, we conclude that despite the move to convergence offers, unlimited bundles are still relatively expensive versus the peer group in this report. In our view, this would mean any new entrant to the Spanish market could easily disrupt the market and launch cheaper unlimited bundles with handset subsidies. MVNOs could play the role of disruptor if they decided they wanted to broaden their customer universe, and were prepared to bear the upfront margin pressure of such a move.

Mobile customers can save more with convergence While mobile tariffs may appear relatively expensive, the mobile bundle included in convergence packages is on average 30% cheaper than the standalone mobile offer. Nevertheless, falling consumer spending and increasing unemployment have been weighing on households’ telecoms services budgets, and will continue to do so until at least 2015. Spanish citizens remain very sensitive to price reductions and cost-savings, more so than their counterparts elsewhere in the EU. Clearly, the significant discount on convergence offers will attract Spanish households and have an impact on the telecoms market structure and its dynamics.

In our view, the move to convergence offers could put more pressure on regulators to re-define “the market” and re-designate operators with significant market power. Any market concentration analysis from the regulator would have to take into account competitive pricing points in the overall market, not just in either the fixed or mobile markets. In our view, this could facilitate in-market consolidation in the long term. The move to convergence will also strengthen the MVNOs’ positioning in the mobile market, as it is easier for wireline operators to upsell mobile products to their broadband customers. JAZ and ONO are the two MVNOs that gained most mobile market share since the introduction of Movistar Fusion in October 2012.

The conclusion is that standalone mobile pricing in Spain remains relatively expensive versus European peers, and mobile bundles appear underdeveloped. Given this pricing backdrop, consolidation would seem to have lees prospects of putting a floor under prices as MVNOs, the main market disruptors, would be unlikely to be consolidated, and could even find their positions strengthened through access to remedy spectrum and infrastructure.

A well-developed MVNO sector

At the end of 2010, MVNOs accounted for 2.3m mobile subscribers or 4.2% of the market. By Q1 2013, this figure had increased to 5m. There are around 47 MVNOs in Spain, which have a combined 10% subscriber market share but only a 3% service revenue share. Four MVNOs – Euskaltel, ONO, JAZ and Lebara Mobil – account for more than 50% of mobile virtual network market. These MVNOs operate on one of the three main mobile networks in Spain – TEF’s,


173

Orange’s and VOD’s. Yoigo does not have a tie-up with an MVNO in Spain given its lower spectrum capacity.

Euskaltel operates in Spain’s Basque region under the VOD network, and offers wireline and wireless telephone services. It is the top broadband operator in the autonomous Basque region with a 44% broadband local market share and 27% in mobile.

ONO is the largest Spanish cable operator, with 1.4mln broadband subscribers. Since the introduction of Movistar Fusion, ONO has boosted its mobile net additions by 557,000 (Q1 2013).

JAZ is the fourth-largest broadband operator in terms of subscriber share. The company has 1.35m broadband customers and expects to reach 2.4m mobile subscribers by 2017, from c550,000 at end-Q1 2013. Jazztel has added 57% of its mobile base over the last two quarters following the launch of its bundle offers.

Lebara is the only big MVNO that does not own a fixed line infrastructure. It is a pure MVNO that targets international communities and migrant workers by offering customised offers.

In summary, Spain has a reasonably well-developed MVNO sector – more so than markets like Austria, Sweden and Italy, but less so than Germany, France, the UK and the Netherlands. However, MVNOs’ service revenue share remains low, hence consolidation could well result in pro-MVNO remedies such as more favourable contract terms, the chance to acquire spectrum, or improved infrastructure access (mobile base stations), to assist further or strengthen the MVNO segment.


The distribution network in Spain relies mainly on points-of-sale (POS), including MNOs’ own shops, franchises, specialised shops and retailer networks. For instance, 81% of Orange’s customer acquisitions were made through direct and indirect sales. The remaining 19% stem from remote sale channels and its own online sales portal. Orange’s retail distribution network consists of 2,971 POS, of which 400 are retail direct channel. TEF has 2,800 retail stores (direct channels) and approximately 620 stores (indirect sales), and according to the company, sales activity at its online channel is not material. The majority of Yoigo’s sales come from POS; the company has over 1,000 branded stores, which are run on a franchise basis.


174

Exhibit 12.17: MVNO/distribution summary in Spain

Source: Berenberg


At the end of 2009, post-paid subscribers accounted for 61.3% of the Spanish market and 84.6% of service revenue. By end-Q1 2013, these figures had increased to 70.3% and 90% respectively. TEF is leading the market in terms of contract customer mix; 75% of its mobile base are post-paid, followed by Orange with 69.7%, VOD with 67.8% and Yoigo 54.9%.


175

Exhibit 12.18: Spain network-sharing arrangements

Source: Company

Interestingly, Orange has improved its customer mix by 13.7ppt over the last three years, while Yoigo only improved its mix by 7ppt. Yoigo could thus be viewed by the other three MNOs as an interesting target in terms of giving them scope to improve churn, as a merger would give them an opportunity to migrate Yoigo’s pre-paid customers to post-paid. In our view, Yoigo is competing mainly against MVNOs in the pre-paid market, as most MNOs are not present at the low end of the market. Moreover, Yoigo’s strategy is to deliver high-speed mobile broadband to customers, while MVNOs focus more on voice and SMS usage.

In conclusion, if one potential benefit of consolidation is the opportunity to accelerate the migration of pre-paid subscribers to lower-churning, higher spending post-paid tariffs, the Spain would seem relatively less well positioned than peer group markets, simply due to the already advanced state of Spain’s post-paid market.

Network-sharing

In urban areas, network-sharing in Spain appears relatively limited. VOD has a network-sharing agreement on passive mobile infrastructure with TEF. In rural areas, network-sharing is relatively more developed, VOD does 2G and 3G RAN sharing with Orange in areas with less than 25,000 inhabitants, the network-sharing enables both companies to improve their coverage by 25% while reducing the projected total number of sites needed for the deployment of both networks by almost 40% in the areas that benefit from the agreement.

We do not believe the current state of network-sharing will be a hurdle to consolidation as network-sharing remains largely passive. However, anti-trust authorities could seek to use infrastructure access as a remedy in the event of consolidation, facilitating a new entrant’s ability to rapidly gain coverage in the early years of remedy-spectrum ownership.


We do not think the Spanish regulatory authorities have explicitly expressed their view on network consolidation, or specified a preference for a minimum number of networks.


Back-

haul

Full

RAN

Spec-

trum

Core

n/w



Oct-07 Spain Vod/Orange Y Y Y Orange: €200m (5 yrs) or

6% of opex

3G sharing in less dense areas; reduces sites needed by 40%. Agreement was extended

to 2G in June 2012.

Apr-09 Spain TEF/VOD Y VOD: 10% of opex Joint build;share 2G/3G; economic situation was a catalyst:

• Spain: extend 2007 site sharing agreement to power/cabinets/mast.

Network sharing


176

Swedish consolidation too complex to attain


Despite conjecture in the market, following comments made by 3’s CEO Peder Ramel, we do not believe consolidation is likely. We see three major headwinds to consolidation in Sweden: the current JV structures within the market, the low level of MVNO activity and the risks posed to market structure should a new entrant be able to acquire spectrum cheaply. Looking at the structure of the market, TLSN is well placed given that it is the largest operator and still growing, while the TEL/ TEL2B Net4Mobility JV also offers both players the bulk of network synergies that could be sought through consolidation.

3 is arguably well placed, given its strong spectrum position in Sweden, but this is potentially a double-edged sword, given that any consolidation scenario could result in spectrum divestments. While 3 would clearly have the most to gain from consolidation, we doubt any of the other operators would pursue it. Another reason is that any remedies imposed could be a deal-breaker. We would expect an MVNO remedy given the low market share, while the threat of spectrum being auctioned to a new entrant could be material. We view Swedish consolidation as a tail event, not something investors should expect.

Which combinations are most likely to gain approval?

A combination between TEL and 3 is in our view the most likely to be acceptable from a regulatory perspective, given that it results in the smallest increase in HHI (+553) on service revenues. It would leave the combined business with a 33.5% service revenue share as a new number two in the Swedish market. However, the JV structure makes this scenario highly unlikely, in our view.

With TEL2B and TEL entrenched within the JV, an alternative consolidation scenario could be a tie up between TLSN and 3. However, we struggle to see why this would be allowed, given it would drive a 1,221 increase in service revenue HHI to 4,041 and would result in the combined entity having a 56% service revenue share. This in our view would be an unacceptable market share level.

Could imposed remedies be a deal-breaker?

In any consolidation scenario, some form of imposed MVNO access is likely, given that the MVNO market share was only 3.2% in June 2012 (per Swedish regulator PTS). When we look at a potential deal between the fourth player, 3, and the third player, TEL, the adjusted spectrum HHI increases by 809 versus 553 in service revenue HHI, leaving the combined business with 40% of the adjusted spectrum in the market versus a 27% service revenue share. Were the excess spectrum to be auctioned or given to a new operator, this could provide a threat to the logic of any deal – especially with high entry level pricing in Sweden. If this was accompanied with a tower share or a roaming obligation, it could really have teeth. But given that TEL is already operating via the Net4Mobility JV, we see this deal as highly unlikely.

A deal between TLSN and 3 would face many of the same regulatory issues and more, given that the combined entity would have a 56% service revenue share. In our view, TLSN would have to divest low-end brand Halebop, and only TEL would likely be allowed to buy it. The question then would be: why would TLSN want to pay a premium for 3 and then have to sell Halebop at a discount?


177

Exhibit 13.1: Market structure and potential remedies make consolidation in Sweden highly unlikely

Source: Berenberg

Market overview

Overall, the Swedish mobile market has been a rare example of growth among the heavy declines in other European markets – and this is despite it being a four-player market.

• With a population of just under 9.5m people, Sweden has over 14m mobile subscriptions.

• Based on the regulatory data from June 2012, the market was segmented with 10.5m private subscriptions and 3.2m business subscriptions.

• Of the private subscriptions, there were 1.4m mobile broadband-only subscriptions, implying just under 9.2m call subscriptions. Thus, even taking account of some multiple SIMs on the pre-paid side, we can say that almost everyone has a call subscription of some type.

• What stands out in the Swedish market is the number of mobile broadband subscriptions (ie dongles, iPads). In June 2012, this stood at 1.4m in the consumer segment – equating to 15% of the population. In addition, there were 670,000 mobile broadband subscriptions in the business segment.

Potential deal Market issues to consider Potential remedies Conclusions Deal likelihood (1-10)

Telenor + Three

1) Telenor already shares network in

JV with Tele2

2) It also shares spectrum - so an

agreement between Telenor and Three

would also require an agreement

between Telenor and Tele2

3) It could be costly to leave the JV

relationship

4) Telenor already has network

synergies through its JV with Tele2.

1) We do not believe that Three could

simply be brought into the

Net4Mobility JV, mainly because it

would leave two main networks in

Sweden - which is unfavourable from

a regulatory perspective

2) The combination would be left with

40% of the spectrum and 27% of

service revenues. So spectrum

divestments are a reasonable remedy

3) This could be accompanied with a

remedy to force roaming or tower

sharing, given the large Swedish

geography

4) An MVNO remedy would also be

likely given low MVNO share

DEAL IS UNLIKELY: mainly due to it

being unlikely that the regulator

would accept a joint Telenor+Three

/Tele2 JV. So any deal would be

contingent on the break up of the JV

which would be costly and unwind

network synergies.

3

Telia + Three

1) From a market perspective there is

little to get in the way of this deal

2) Three has too much spectrum

relative to its market share while Telia

has too little

3) So from a market perspective this

would be positive.

1) With a combined 56% service

revenue share this deal would

definitely attract regulatory scrutiny

and synergies.

2) However given the combined

business would have 54% of the

spectrum in the market, a divestment

is a likely remedy.

3) With Net4Mobility holding most of

the rest of the spectrum, this would

leave a network duopoly. So spectrum

divestments would most likely be

targeted towards a new entrant

4) An MVNO obiligation would be

likely

5) In addition, Telia could be obliged

to divest its Halebop brand to make

the deal happen

6) Given this competes closely with

Tele2, Telenor is the only likely buyer

DEAL IS UNLIKELY: From a

regulatory perspective, given a 56%

service revenue share, it is possible

the deal would simply not be

allowed. If it was, Telia would likely

need to divest Halebop. So for the

deal to happen, Telia will need to pay

a premium for Three, and arguably

sell Halebop at a discount. We see

this as unlikely. Were Three a

distressed seller, this may make sense

to acquire the spectrum, but we still

see this as unlikely.

2


178

• Data has been one of the key growth drivers in Sweden. In June 2012, 64% of mobile voice subscriptions were taking a data package, while 31% of subscriptions were taking a data package with over 1GB of data. Our interactions with the companies in the sector suggest that data take-up has continued to grow at a rapid pace.

In Exhibit 13.2 below, we show that since end-2006:

• TLSN has lost 4ppt of subscriber market share, with a share of 40% at end-2012;

• TEL2B has lost 2ppt of market share, with about 30% of the market at end-2012;

• TEL has lost 1ppt of market share and gone down to a 16.5% market share;

• 3 has been the big market share gainer, moving from a 5% to a 10% market share, while MVNOs have also gained, although they still remain at the periphery, with 3-4% of the market.

In Exhibits 13.3 and 13.4, we show service revenues split by operator. As can be seen, the main beneficiaries of growth in Sweden have been TLSN and 3, with growth at TEL and TEL2B being less compelling. 2012 service revenues were 45% higher than where they were in 2006, equivalent to a 6.4% CAGR. However, growth has become ever more challenging, slowing to 3.3% in 2012 from 6.6% in 2011. 3 continues to have the best growth rate at 11% in 2012, while TEL’s 2012 service revenues were flat.

Exhibit 13.2: Strong market subscriber growth has been driven by data (subscribers quoted in thousands)

Source: Berenberg, PTS Note we adjust numbers to align with PTS regulatory data, which adjusts for company reporting and standardises (for example it adjusts out M2M, which is included in TLSN subscriber data)

2006 2007 2008 2009 2010 2011 2012

Other 144 101 142 212 305 415 568

3 480 567 773 975 1,155 1,299 1,454

Tele2 3,103 3,217 3,464 3,760 3,985 4,099 4,161

Telenor 1,672 1,851 1,895 1,974 2,056 2,210 2,324

TeliaSonera 4,208 4,381 4,618 4,829 5,191 5,371 5,562

-

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000


179

Overall, we can conclude that the Swedish market has benefited from various positive tailwinds, although these are now becoming harder to monetise.

1. Between 2008 and 2012, growth in mobile broadband subscriptions provided the market with a big boost. In 2011, about half of service revenue growth came from mobile broadband subscriptions. However, most contracts now include a set amount of data, with the ability to spread that data allowance over your mobile broadband SIM, so further monetisation of mobile broadband looks challenging from here.

2. Voice stability: For some time, voice revenues were stable in Sweden, as a result of the high levels of line loss and increasing usage on the mobile network. However, bundles have now moved to a size where monetising increasing usage is more challenging, while we are seeing some cannibalisation at the high end. Consequently, voice revenues are no longer stable in the Swedish market.

3. Data growth: Overall, data growth on smartphones continues to drive the market, with the average smartphone user consuming over 1GB of data. However with smartphone penetration in Sweden now over 80%, this growth becomes harder to achieve. The next leg of growth, however, is coming from 4G, which is boosting usage and in turn boosting upsell.

In conclusion, the Swedish market has stood out as a very rational, well developed market, also aided by the network JV structure driving a more rational pricing environment. However, the strong growth story is starting to come to an end, and with the TEL/TEL2B JV showing such network strength, the dynamics of market share growth could also change, particularly with regards to the business segment.

In our view, therefore, the dynamics of growth are such that we expect a more stable environment rather than a high-growth environment. This does risk seeing competition ratchet up, if there are any obvious losers. However, Sweden still stands out as a good market relative to the rest of Europe. Exhibit 13.3: Data growth has driven market service revenue growth (data is in SEKm)

Source: Berenberg

2006 2007 2008 2009 2010 2011 2012

3 1,415 2,458 2,174 2,829 4,125 4,828 5,364

Tele2 5,907 6,632 7,609 7,471 7,768 8,226 8,517

Telenor 5,662 6,173 6,218 6,199 6,462 6,734 6,743

TeliaSonera 10,991 10,996 12,324 13,074 13,859 14,692 15,053

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000


180

Exhibit 13.4: TLSN and 3 have exhibited the strongest service revenue growth (data is in SEKm)

Source: Berenberg

Market concentration

We use both market share and HHI to analyse market concentration in the Swedish market, and we do this by looking at both subscriber share and service revenue share.

• Subscriber share (Exhibits 13.5 and 13.6): If we just look at the market share of the MNOs (MVNO and service provider subscribers are consolidated within the MNO network subscribers), we find that TLSN has a 40% share, TEL2B 30%, TEL 19% and 3 a 10.5% share. This equates to an HHI of 3,006 – broadly in line with the average for other four-player markets. However, the right way to do this in our view is to split out MVNOs from the market. This has a less material impact in Sweden, given that MVNOs have only a 4% market share in Sweden, with Lyca mobile being the only MVNO of note. This leaves Lyca with a market share of just over 2% – and drives a lower HHI of 2,822.

• Service revenue market share (Exhibits 13.7 and 13.8): We repeat the above analysis but for service revenues. Including MVNOs, TLSN has a 41% market share, followed by TEL2B at 23%, TEL at 19% and 3 at 15%. It is on this metric that the HHI is lowest at 2,820.

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

2006 2007 2008 2009 2010 2011 2012

TeliaSonera Telenor Tele2 3


181

Exhibit 13.5: Sweden MNO subscriber HHI at 3,006

Exhibit 13.6: Sweden MNO/MVNO subscriber HHI at 2,822


Exhibit 13.7: Sweden MNO service revenue HHI at 2,913

Exhibit 13.8: Sweden MNO/MVNO service revenue HHI at 2,820


How concentration evolves under various consolidation scenarios, as well as the implications for competition is evidently a key EC metric for analysing what remedies to put in place. It is worth mentioning that most combinations in the Swedish market should qualify as a “concentration” with a “community dimension” – meaning it falls under EC scrutiny rather than that of the national regulator (as shown by the criteria laid out in Exhibit 13.9). However TEL2B is a grey area given that ex-Russia, its revenues fall below the €5bn and above the €2.5bn threshold. Given that a merger in Sweden would be a with one entity only (likely 3), this would in our view mean that it escapes EC scrutiny and would instead be looked at by the local regulator KKV.

Exhibit 13.9: EC criteria for community dimension shows all consolidation scenarios should fall under the EC mandate, given the acquirer’s position

Source: Berenberg (i) combined worldwide turnover is more than €5bn and each of at least two of the merging parties realised more than €250m in turnover in the EU, or (ii) combined worldwide turnover is more than €2.5bn; their combined turnover is more than €100m in each of at least three member states; in each of those three member states, the turnover of each of at least two of the merging parties is more than €25m; the community-wide turnover of each of at least two of the merging parties is more than €100m (iii) unless each of the merging parties obtains more than two-thirds of its EU turnover in one member state.

TeliaSonera, 1,625

Telenor, 361

Tele2, 909

Three, 111

TeliaSonera, 1,563

Telenor, 273

Tele2, 875

Three, 107 Lyca mobile, 5

TeliaSonera, 1,745

Telenor, 389

Tele2, 559

Three, 222

TeliaSonera, 1,708

Telenor, 350

Tele2, 543


Worldwide

turnover (€) EU turnover (€)

Swedish turnover

(€)

Turnover

>€100m in 3

EC states (i) (ii) (iii)

TeliaSonera 12,248 8,654 2,020 YES YES YES NO

Tele2 (ex-Russia) 3,589 2,765 1,168 YES NO NO NO

Telenor 13,366 3,507 1,101 YES YES YES NO

Hutchison (telco only for EU) 23,801 5,778 740 YES YES YES NO


182

When we look at the various consolidation options, we conclude that 3 is the asset that would most likely have to form part of the consolidation, were it to happen.

• TLSN and TEL, or TLSN and TEL2B: We do not view either of these two scenarios as credible. TLSN and TEL2B would result in a service revenue market share of 65% and subscriber market share of close to 70%, driving the subscriber HHI above 5,000. TLSN and TEL drives a service revenue market share of 60% with HHI close to 4,400.

• TLSN and 3 (Exhibits 13.10 and 13.11): This would result in a subscriber market share of 50% (with MVNOs separated out) and a service revenue market share of 56%. However with TEL and TEL2B operating a JV, this could effectively leave Sweden with only two networks. In reality, we do not think this is a situation that would be acceptable given that it would leave Sweden with an effective duopoly of networks.

• TEL2B and TEL (Exhibits 13.12 and 13.13): This combination would create a new market leader, with a 46% subscriber market share and 42% revenue market share. To a large extent, the two companies operate at differing ends of the market, and logistically, the fact they are operating through the same JV would make this deal easier to achieve, although with smaller synergies. So given that network synergies have already been achieved, both companies may prefer to avoid a lengthy approval process on any transaction. In addition, the subscriber HHI would go from 2,800 to 3,800 – a rather material increase. In addition, such a deal would likely come with the requirement to divest spectrum to 3, thus offsetting any non-network synergies the deal may create.

• TEL2B and 3 (Exhibits 13.14 and 13.15): A combined market share of 40% at subscribers and 38% at service revenues would create a very strong number 2, leaving Telenor as the number 3 with 19% service revenue share. However, we see two major stumbling blocks to this deal. Firstly, from a regulatory standpoint, it clearly matters which segment of the market a company competes at, and the price war of 2012, when TEL2B responded to 3’s pricing, clearly showed that TEL2B and 3 are very much competing for the same customers. This would likely prove highly problematic to any deal. Secondly, TEL2B is in a JV with TEL, and were it to bring 3 into that JV, Sweden would effectively be left with two networks. This would leave the JV with a lot of spectrum, which would either have to be divested to TLSN or a new entrant or would lead to a deal being blocked because having two networks only would be considered undesirable. For these reasons, we would see this deal as unlikely.

• TEL and 3 (Exhibits 13.16 and 13.17): Arguably, this is the most likely of any potential deal, and it drives the smallest impact on HHI. It would leave TEL as the third operator in terms of subscriber share at 27% but would give it service revenue share of 34%. So it still breaches the EC’s 25% threshold to qualify for review, but given it is a combination between number three and four in the market, it is clearly the most acceptable of the potential outcomes. However, the JV issues that we highlight for a TEL2B and 3 combination are just the same here, and again, this forms the key reason why we view any kind of combination as unlikely. Why should TEL risk forgoing the benefits of its JV with TEL2B to go and pay a premium for 3’s business?


183

Exhibit 13.10: Sweden MNO/MVNO subscriber HHI assuming TLSN and 3 at 3,639

Exhibit 13.11: Sweden MNO/MVNO service revenue HHI assuming TLSN and 3 at 4,041


Exhibit 13.12: Sweden MNO/MVNO subscriber HHI assuming TEL2B and TEL at 3,799

Exhibit 13.13: Sweden MNO/MVNO service revenue HHI assuming TEL2B and TEL at 3,692





TeliaSonera, 2,487

Telenor, 273

Tele2, 875

Three, - Lyca mobile, 5

TeliaSonera, 3,147

Telenor, 350

Tele2, 543 Three, - Lyca mobile,

1

TeliaSonera, 1,563

Telenor, 2,125


TeliaSonera, 1,708

Telenor, 1,765


TeliaSonera, 1,563

Telenor, 273

Tele2, 1,593

Lyca mobile, 5

TeliaSonera, 1,708

Telenor, 350

Tele2, 1,449

Lyca mobile, 1


184




We should also not discount 3 being an acquirer, although that would do little to address any of the above issues. TEL is unlikely to be a seller, and if 3 wanted to acquire TEL2B’s Swedish business, as well as the competition issues, it would also require TEL2B and TEL to agree on a split of the spectrum within the JV, making it a complex and in our view insurmountable transaction.

In Exhibit 13.18 below, we summarise the impact on HHI.

Exhibit 13.18: A combination of TEL and 3 has the smallest impact on HHI

Source: Berenberg

TeliaSonera, 1,563

Telenor, 721

Tele2, 875

Lyca mobile, 5

TeliaSonera, 1,708

Telenor, 1,121

Tele2, 543 Lyca mobile,

1

HHI

Subscriber market

share (MVNOs


Service revenue

market share

(MVNOs


Current 2,822 2,820

Telia + Three 3,639 817 4,041 1,221

Tele2 + Three 3,433 611 3,508 688

Telenor + Three 3,164 342 3,373 553

Telenor + Tele2 3,799 977 3,692 872

Telia + Tele2 5,160 2,338 4,746 1,926

Telia + Telenor 4,128 1,306 4,367 1,547


185

Low prices and big bundles?

In the last three years, we have seen a step change in pricing in Sweden. For some time, we would have characterised Swedish pricing as being opaque, as some tariffs offer simply an entry for a price per minute, other tariffs offer in bundles, and some offer a pot structure (customers pay a certain amount and each minute or SMS that is deducted from the entry fee).

However, today we are in the realms of all-you-can-eat voice and SMS bundles, with subscribers paying for data. This trend was really driven by TEL2B, as it chose to price aggressively to try and defend against pre-paid subscriber losses. Now, most of the other operators have followed, as operators are choosing to monetise on data rather than voice and SMS, when it comes to upsell.

In Exhibit 13.19 we show the pricing for three user profiles in Sweden:

1) a user wanting 100 minutes, 500 SMS and 200MB of data;

2) a user wanting unlimited voice/SMS and 1GB of data;

3) the same as 2) but with an iPhone 5 16GB.

The results are particularly interesting, especially when we look at the low end of the market.

1) In the low-end user segment, while it is possible to replicate the user profile, it tends to be far more expensive than a bundle, mainly due to the high levels of SMS. This means that the cheapest option available to the consumer is typically the low-end bundle on offer – and this may not exactly match the profile being looked at. So in some cases, we have tried to match the profile as closely as we could.

The low-end tariffs on TEL, TEL2B and TLSN are all priced at close to the SEK200 level – equivalent to €23. These packages offer more data than is needed, and also more voice. However, they are still cheaper than trying to replicate this usage profile with a lower-end tariff.

The cheapest option at this end of the market is with the low-end TEL2B, TLSN and 3 brands, Comviq, Halebop and Hallon. This again still offers more usage than this user profile would need, but is the cheapest way of achieving it. Yet still this costs c€17 with Comviq and €11.5 with Hallon, which is a new brand.

The conclusion from this analysis is that while the bundles are clearly large, they do not offer a cheap entry point, thus forcing upsell at the low end of the market. In our view, this would mean any market concentration analysis would be very focused on ensuring that competitive pricing points could still be sought at the low end of the market. This would mean that Comviq and Hallon would need to remain separate to one another.


186

Exhibit 13.19: Pricing analysis of the Swedish market (prices in SEK)

Source: Berenberg

2) Given Sweden is a high data usage market, and further along the data

demand curve than most of its European peers, most of the unlimited voice/SMS offers come with more than 1GB of data. In some cases, we can replicate an equivalent (TEL, for example), while in others we have to accept the higher usage.

In this segment, the pricing points tend to range from SEK200-300 per month, with most of the variance driven by the amount of data on offer. The cheapest match of contract available is with Comviq (a low-end TEL2B brand) and this costs SEK195/month (€23). With TEL, customers can almost replicate the package as well, and this costs SEK232/month (€27 – we apportion the cost of an incremental 1GB of data to match the requirements as a guide). With TEL2B, if customers want unlimited voice/SMS, the cost is SEK339 (€40), but with 3GB of data.

100 mins, 500SMS, 200MB Telenor Telia Tele2 Tre Halebop Comviq Hallon Lyca

Brand name XS Small Low Mini+surf base Hello Fixed price Hallon Bra

Voice calls (in min) Unlimited 300 300 100 200 Unlimited 60 calls 300

Other services - - -

Price per extra min 0.69 0.49 0.19 0.19

Starting price 0.99 0.99 0.99 0.99

Data 500Mb 500MB 1GB 1GB 500Mb 0.2GB 0.5GB 0.3GB

Price per extra GB na na na na 0 0

SMS Unlimited 300 300 500 5000 Unlimited 5000 300

Price per extra SMS 0.49 0.19

Subscription Fee 0

Subscription terms 24months 24 months 24 months 24 months 24 months 24 months 24 months 24 months

Monthly Flat rate tariff 199 199 189 243 134 145 99 149

blended ppm

Unlimited voice / SMS + 1GB Telenor Telia Tele2 Tre Halebop Comviq Hallon Lyca

Created with add

on

Brand name XS L Medium Vanner A lot Fixed price Fixed price Bra

Voice calls (in min) Unlimited 1500 Unlimited 3000 1200 Unlimited Unlimited Unlimited

Other services - - - -

Price per extra min 0.69 na 0.59

Starting price 0.99 na 0.99

Data 1GB 3GB 3GB 3GB 6GB 1GB 5GB 3GB

Price per extra GB na na na na 0 0 0

SMS Unlimited 1500 Unlimited 3000 SMS & MMS 12000 Unlimited Unlimited Unlimited

Price per extra SMS 0.69 0.69 0.49

Subscription Fee

Subscription terms 24months 24 months 24 months 24 months 24 months 24 months 24 months 24 months

Monthly Flat rate tariff 232.5 349 339 300 369 195 249 200

Unlimited voice / SMS + 1GB +iphone Telenor Telia Tele2 Tre Halebop Comviq

Created with add

on

Brand name XS L Medium Vanner A lot Fixed price

Voice calls (in min) Unlimited 1500 Unlimited 3000 1200 Unlimited

Other services - - - -

Price per extra min 0.69 na 0.59

Starting price 0.99 na 0.99

Data 1GB 3GB 3GB 5GB 6GB 3GB

Price per extra GB na na na na 0 na

SMS Unlimited 1500 Unlimited 3000 SMS & MMS 12000 Unlimited

Price per extra SMS 0.69 0.69 0.49

Subscription Fee

Subscription terms 24months 24months 24months 24months 24months 24months

Monthly Flat rate tariff 482.5 499 489 499 499 445


187

3) We repeat the analysis in 2) to reach the same equivalent prices for the same contracts with an iPhone. This shows that actually there is very little different in package costs, but the amount of data on offer varies quite a lot. Comviq stands out as the cheapest, with an equivalent €52, while all the other tariffs have an equivalent price of €56-58, typically coming with about 3GB of data. This drives a wide variation in implied subsidies, with an iPhone5 16GB as cheap as €400, but offset by a higher SIM-only tariff. The iPhone is a key component to most sales in Sweden.

As well as the above analysis, the Swedish mobile market has become increasingly nuanced. Firstly, TLSN is now promoting its family packages, which allows users to add an unlimited voice/SMS SIM, but share the data package of the group. We show a summary of the pricing in Exhibit 13.21. In addition, TEL is offering a variation of this, where customers can add another SIM to their package for a tablet for only SEK29, which will then share the data of the total package.

In conclusion, the Swedish market, once viewed as cheap across Europe, no longer screens as having particularly low prices – especially at the entry level for low data users. Coupled with a low MVNO presence, this could provide a risk of a remedy for greater MVNO access. However, given the high and increasing data usage on average in Sweden (data users consumer over 1GB of data on average), this is likely to be a less material portion of the market – especially from a value perspective.

Exhibit 13.20: Implied cost of iPhone 5 16GB within contract (€)

Source: Berenberg

Exhibit 13.21: TLSN family packages

Source: Berenberg

0

100

200

300

400

500

600

700

800

TDC Telenor Telia 3 Onfone Telmore CBB Oister

Prices (SEK) 1 phone 2 phone 3 phone 4 phone

2GB 399 548 697 846

4GB 499 648 797 946

6GB 599 748 897 1046

8GB 699 848 997 1146

10GB 799 948 1097 1246

Prices (SEK) 1 phone+tablet 2 phone+tablet 3 phone+tablet 4 phone+tablet

2GB 428 577 726 875

4GB 528 677 826 975

6GB 628 777 926 1075

8GB 728 877 1026 1175

10GB 828 977 1126 1275


188

MVNOs’ presence in the market

The MVNO market in Sweden has remained small, and based on regulatory data, it had a c3.2% market share back in June 2012. Of this, 1.7% of the market share was from Lyca mobile, which is by far the largest MVNO in Sweden, operating on the TEL network. All other MVNOs are too small to note (see Exhibit 13.22).


Distribution is evolving along different paths for different companies in Sweden. Online remains an important growth component in Sweden, as evidenced by the greater push of SIM only tariffs on the Comviq and Hallon brands. However, the importance of handsets in the Swedish market remains high, as evidenced by the strong subsidies offered.

• TEL2B: As well as rolling out 50 out of a planned 110 stores in Q1, TEL2B has taken an interesting approach to distribution through its low-end online brand Comviq. As well as being online, Comviq also has vending machines in nearly 100 seven-eleven stores, which allow the consumer to buy a phone and SIM at a lower cost to TEL2B. TEL2B is trying to move the focus of sales away from the multibrand operators such as The Phone House, as this has become a more expensive sales channel.

• TLSN: TLSN has arguably contributed to increasing competition in Sweden. The company started selling its lower-end Halebop brand through its own TLSN stores, which was a step-change. TEL2B has done this with Comviq in the past, but is doing so less now.

• 3: The online platform remains important, but according to the company, its importance is actually declining. As TEL2B is focusing more on its own stores, 3 is using this as an opportunity to increase its sales through the multibrand stores such as The Phone House. This remains the main distribution focus for the company currently.

• TEL: Focus is on store presence with a focus on attracting higher value subscribers. Essentially TEL’s strategy is similar to TEL2B’s in that it is trying to migrate additions from the multibrand stores to its own sales channels.


At the end of 2009, post-paid accounted for 54.5% of market subscribers. By Q1 2013, this has increased to 68.8%. We estimate that post-paid accounted for 90% of service revenues at Q1 2013, compared with closer to 65% in 2009.

Rapid post-paid growth in Sweden has been driven by the strong uptake in smartphones, with some commentators estimating smartphone penetration in Sweden at over 80%. With bundles now moving to the all-you-can-eat model, we expect the current run rate of 1-2ppt per year post-paid share growth to sustain in the medium term.


189

Exhibit 13.22: MVNO/distribution summary in Sweden

Source: Berenberg


In Exhibit 13.23, we show how spectrum is split between operators and the JVs through which they operate. When we look at the spectrum HHI, at 2,617 (and 2,717 for the less than 1GHz spectrum) it is close to the subscriber share and service revenue HHI at c2,800. However, when we look at the market shares, we see that 3 is left with excess spectrum relative to its market share, while TLSN is left with too little.

3

70 storesDirect

Comhem

Fogg5-6

MVNOs

The Phone HouseMobilizera

Reseller/

Retail

68 3 stores in SwedenDirect

Telenor

Lyca mobile (the largestMVNO in Sweden)

Connexion

2MVNOs

Mobilizera

Phone FamilyDialect (for SME)

Reseller/

Retail

TeliaSonera

Tele2

Only SP, of which Alltele is the largest with c.90k

subs

No MVNOs

The Phone HouseElgiganten

Reseller/Retail

87 TeliaSonera stores in

SwedenDirect

TDC MVNO 37k subs. Wireless Maingate is M2M MVNO. Ludo &

Joors are SPs

3-4

MVNOs

The Phone HouseElgigaten

Reseller/Retail

Rolled out 45 of a

planned 110 own stores in Q1

Direct


C

u

s

t

o

m

e

r

s


190

Exhibit 13.23: Swedish spectrum split

Source: Berenberg

Exhibit 13.24: Total spectrum share

Source: Berenberg * assumes JVs split spectrum equally amongst owners

Exhibit 13.25: Swedish spectrum split for <1GHz spectrum bands


However, this is based on absolute spectrum amounts, which fails to take into account the differing quality of different spectrum bands. So we calculate an adjusted spectrum holding based on 1MHz of 800MHz/900MHz spectrum being able to cover2:

• 2x the area of 1800MHz spectrum;

• 2.5x the area of 2100MHz spectrum; and

• 9x the area of 2600MHz spectrum.

So if we adjust the spectrum holdings for this as we do in Exhibit 13.26, we can see that 3’s share of spectrum holding is double the level of its current market share at 10.3%. This should in theory leave 3 well placed to continue taking market share, with TLSN looking exposed given its higher subscriber share versus spectrum share.

2 Based on data provided by OFCOM

FDD Spectrum TOTAL Share TOTAL <2GHz Share TOTAL <1GHz Share

Telia 150 28.3% 110 40.7% 40 30.8%

Svenska 40 7.5% 0 0.0% 0 0.0%

Net4Mobility 182 34.3% 102 37.8% 32 24.6%

Telenor 50 9.4% 10 3.7% 10 7.7%

Tele2 18 3.4% 18 6.7% 18 13.8%

3 90 17.0% 30 11.1% 30 23.1%

TOTAL 530 100.0% 270 100.0% 130 100.0%

TOTAL Spectrum TOTAL Share HHI Subscriber market share

Telia 170 32.1% 1029 39.5%

Telenor 141 26.6% 708 16.5%

Tele2 129 24.3% 592 29.6%

3 90 17.0% 288 10.3%

Other 4.0%

TOTAL 530 100.0% 2617 100.0%

TOTAL Spectrum <1GHz TOTAL Share HHI Subscriber market share

Telia 40 33.3% 1111 39.5%

Telenor 16 13.3% 178 16.5%

Tele2 34 28.3% 803 29.6%

3 30 25.0% 625 10.3%

Other 4.0%

TOTAL 120 100.0% 2717 100.0%


191

Exhibit 13.26: Swedish spectrum split with spectrum amounts adjusted for coverage


Implications for consolidation

We have highlighted that we did not see any obvious M&A deals in Sweden, mainly due to the complication of the network JV structure. However, the one deal that would most likely be acceptable from a regulatory perspective is one between number three operator TEL and number four operator 3. As we showed in Exhibit 13.18, this would leave the subscriber HHI at 3,164, and the service revenue HHI at 3,373.

TEL and 3: The question that follows is what impact would it have on spectrum HHI and would this drive any spectrum divestments? To carry out this analysis, we assume that the deal would drive a split in two of the spectrum that TEL holds through its Net4Mobility JV with TEL2B. The result of this is shown in Exhibit 13.27 for unadjusted spectrum and Exhibit 13.28 for adjusted spectrum.

• Subscriber HHI: This increases by 342 to 3,164 as a result of a TEL and 3 tie-up.

• Service revenue HHI: This increases by 553 to 3,373.

Exhibit 13.27: Swedish spectrum split – assumes TEL acquires 3 Sweden


Exhibit 13.28: Swedish spectrum split with spectrum amounts adjusted for coverage – assumes TEL acquires 3 Sweden


TOTAL adjusted Spectrum TOTAL Share HHI Subscriber market share

Telia 87 34.5% 1189 39.5%

Telenor 54 21.3% 453 16.5%

Tele2 64 25.2% 636 29.6%

3 48 19.0% 362 10.3%

Other 4.0%

TOTAL 254 100.0% 2640 100.0%

TOTAL Spectrum TOTAL Share HHI Subscriber market share

Telia 170 32.1% 1029 39.5%

Telenor 231 43.6% 1900 26.9%

Tele2 129 24.3% 592 29.6%

Other 4.0%

TOTAL 530 100.0% 3521 100.0%

TOTAL adjusted Spectrum TOTAL Share HHI Subscriber market share

Telia 87 34.5% 1189 39.5%

Telenor 102 40.3% 1624 26.9%

Tele2 64 25.2% 636 29.6%

Other 4.0%

TOTAL 254 100.0% 3449 100.0%


192

• Spectrum HHI: This increases by 904 to 3,521. This is obviously a far greater increase than seen in the subscriber or service revenue HHI, and would leave the combined business with 44% of the spectrum and only a 27% market share.

• Adjusted spectrum HHI: This increases by +809 to 3,449. This still leaves the combined business with 40% of the spectrum and only 27% market share.

As a result of the higher move in spectrum HHI, it is reasonable to expect a combined TEL and 3 to need to divest spectrum, with both TEL2B and TLSN arguably well placed to capitalise. Alternatively, such a combination could result in spectrum being auctioned, or even given to a new entrant, with an obligation on towers being made. With pricing in Sweden arguably not low, this could pose a threat to any consolidation. Coupled with the issues of network share, this continues to support our view that consolidation in the Swedish market is unlikely.

TLSN and 3: We replicate the analysis above for TLSN and 3. From a market perspective, this is an easier deal, given that TLSN does not have JV issues adding any complications. In addition, TLSN’s market position is such that it has more subscriber and service revenue share than it does spectrum, while 3 is in the opposite position. The result of this is shown in Exhibit 13.29 for unadjusted spectrum and Exhibit 13.30 for adjusted spectrum.

• Subscriber HHI: This increases by 817 to 3,639 as a result of a TLSN and 3 tie-up.

• Service revenue HHI: This increases by 1,221 to 4,041.

• Spectrum HHI: This increases by 1,089 to 3,707. It would leave subscriber share and spectrum share closely aligned at close to 50%.

• Adjusted spectrum HHI: This increases by 1,312 to 3,951.

Exhibit 13.29: Swedish spectrum split – assumes TLSN acquires 3 Sweden


Exhibit 13.30: Swedish spectrum split with spectrum amounts adjusted for coverage – assumes TLSN acquires 3 Sweden


Although more appealing from a market perspective, this deal would likely come with tougher regulatory hurdles. The high combined service revenue share would likely force subscriber divestments as well as spectrum divestments. Spectrum

TOTAL Spectrum TOTAL Share HHI

Subscriber

market share

Telia 260 49.1% 2407 49.9%

Telenor 141 26.6% 708 16.5%

Tele2 129 24.3% 592 29.6%

Other 4.0%

TOTAL 530 100.0% 3707 100.0%

TOTAL adjusted

Spectrum TOTAL Share HHI

Subscriber

market share

Telia 136 53.5% 2863 49.9%

Telenor 54 21.3% 453 16.5%

Tele2 64 25.2% 636 29.6%

Other 4.0%

TOTAL 254 100.0% 3951 100.0%


193

divestments would have to be to a new entrant – otherwise the Swedish market would be left as a network duopoly. The most obvious subscriber divestment, would be TLSN’s Halebop brand, but given how closely this competes with Comviq (TEL2B), it is likely either a new entrant or TEL would be the only potential buyers. So this deal is effectively reliant on TLSN paying a premium for 3, and potentially selling Halebop at a discount. We struggle to see why it would do this. While we cannot conclusively rule this out, we think it would only be likely if 3 was a distressed seller.

Network-sharing

Sweden as a market has led network-sharing initiatives, with active infrastructure sharing driving a dense and more cost effective network build-out. There are three separate sharing agreements at present in Sweden.

• 3GIS: This is the 3G sharing arrangement between 3 and TEL in Sweden. This differs from the other sharing arrangements in Sweden in that it shares the 3G infrastructure outside the major cities, and it does not include the sharing of spectrum.

• Svenska: This is the 3G (2.1GHz) sharing arrangement between TLSN and TEL2B.

• Net4Mobility: This is the active sharing of 2G and 4G network between TEL and TEL2B. Net4Mobility bid for 800MHz and 1800MHz spectrum independently of TEL and TEL2B.

PTS highlighted its support for network-sharing as a means for achieving high levels of 3G coverage of a very large geographical area. Sweden is a sparse country with only 20 inhabitants/km2 versus Denmark at 128 inhabitants/km2. This meant, as PTS highlighted at the time, that the cost of rolling out 3G coverage to the rural areas in Sweden would be large, relative to a market such as Denmark (as shown in Exhibit 13.32 by the number of sites required).

Exhibit 13.31: Network-sharing in Sweden

Source: PTS


194

Exhibit 13.32: 3G rollout requirements estimated by PTS

Source: PTS

Exhibit 13.33: Capex per subscriber (SEK)

Source: PTS

As a result, a much higher capex per subscriber was required to achieve rural coverage, thus making the relative economics of a network-sharing agreement in a country with a demographic like Sweden versus Denmark more favourable.

TEL2B and TEL have, however, taken the economics of network-sharing to the next level with the extensive sharing of all 2G and 4G infrastructure across the whole country.

Number of sites

0

2 000

4 000

6 000

8 000

10 000

12 000

Sverige Denmark

Urban Rural

Capex per sub

124 89

5 041

466526

152

0

1 000

2 000

3 000

4 000

5 000

6 000

Sverige Denmark

Urban Rural Total


195

Please note that the use of this research report is subject to the conditions and restrictions set forth in the “General investment-related disclosures” and the “Legal disclaimer” at the end of this document.

For analyst certification and remarks regarding foreign investors and country-specific disclosures, please refer to the respective paragraph at the end of this document.

Disclosures in respect of section 34b of the German Securities Trading Act (Wertpapierhandelsgesetz – WpHG)

Company Disclosures Telefónica SA no disclosures Belgacom SA no disclosures Telenor ASA no disclosures Cable & Wireless Communications plc no disclosures BT Group plc no disclosures Deutsche Telekom AG 5 Drillisch AG 3 Elisa Oyj no disclosures freenet AG 5 France Télécom SA 5 Iliad SA no disclosures Jazztel plc no disclosures Kabel Deutschland Holding AG no disclosures KPN NV no disclosures Millicom International Cellular SA no disclosures Mobistar SA no disclosures Portugal Telecom SGPS SA no disclosures QSC AG 3 Swisscom AG no disclosures TalkTalk Telecom Group plc no disclosures TDC A/S no disclosures Tele2 AB no disclosures Telecom Italia SpA no disclosures Telecom Italia SpA (sav.) no disclosures Telekom Austria Group no disclosures Telenet Group Holding NV no disclosures TeliaSonera AB no disclosures United Internet AG 3, 5 Vodafone Group plc no disclosures Zon Multimedia SGPS SA no disclosures (1) Joh. Berenberg, Gossler & Co. KG (hereinafter referred to as “the Bank”) and/or its affiliate(s) was Lead

Manager or Co-Lead Manager over the previous 12 months of a public offering of this company. (2) The Bank acts as Designated Sponsor for this company. (3) Over the previous 12 months, the Bank and/or its affiliate(s) has effected an agreement with this company

for investment banking services or received compensation or a promise to pay from this company for investment banking services.

(4) The Bank and/or its affiliate(s) holds 5% or more of the share capital of this company. (5) The Bank holds a trading position in shares of this company.


196

Historical price target and rating changes for Telefónica SA in the last 12 months (full coverage)

Date Price target - EUR Rating Initiation of coverage

26 July 12 9.70 Hold 25 January 10

26 September 12 10.40 Sell

04 March 13 9.40 Sell

18 June 13 9.10 Sell

Historical price target and rating changes for Belgacom SA in the last 12 months (full coverage)


06 March 13 17.00 Sell 25 January 10

18 June 13 16.00 Sell

Historical price target and rating changes for Telenor ASA in the last 12 months (full coverage)

Date Price target - NOK Rating Initiation of coverage

25 June 12 117.00 Hold 05 October 10

10 July 12 119.00 Hold

27 July 12 122.50 Hold

26 September 12 137.00 Buy

29 October 12 141.00 Buy

07 January 13 138.00 Buy

26 February 13 140.00 Buy

08 April 13 141.50 Buy

03 May 13 137.00 Hold

Historical price target and rating changes for Cable & Wireless Communications plc in the last 12 months (full coverage)

Date Price target - GBP Rating Initiation of coverage

25 June 12 0.32 Hold 26 March 10

26 September 12 0.34 Hold

06 December 12 0.38 Hold

14 January 13 0.40 Hold

06 March 13 0.42 Hold

18 June 13 0.40 Hold

Historical price target and rating changes for BT Group plc in the last 12 months (full coverage)


26 September 12 2.60 Buy 25 January 10


15 May 13 3.75 Buy

Historical price target and rating changes for Deutsche Telekom AG in the last 12 months (full coverage)


26 September 12 10.50 Hold 25 January 10



197

Historical price target and rating changes for Drillisch AG in the last 12 months (full coverage)


07 September 12 11.00 Buy 07 September 12

16 November 12 12.00 Buy

23 January 13 12.50 Buy

06 March 13 13.30 Buy

27 March 13 15.00 Buy

02 May 13 15.00 Hold

23 May 13 14.00 Hold

Historical price target and rating changes for Elisa Oyj in the last 12 months (full coverage)


26 September 12 17.00 Sell 23 September 10

12 November 12 17.00 Hold

31 January 13 16.00 Sell

06 March 13 14.50 Sell

15 April 13 14.50 Hold


18 June 13 12.85 Sell

Historical price target and rating changes for freenet AG in the last 12 months (full coverage)


31 July 12 13.50 Buy 28 May 10



10 December 12 15.50 Buy

01 March 13 16.10 Buy

02 May 13 18.00 Hold

Historical price target and rating changes for France Télécom SA in the last 12 months (full coverage)



18 February 13 8.00 Hold

Historical price target and rating changes for Iliad SA in the last 12 months (full coverage)


25 June 12 104.00 Sell 26 September 11


06 March 13 128.00 Hold

18 June 13 150.00 Hold

Historical price target and rating changes for Jazztel plc in the last 12 months (full coverage)


30 November 12 6.55 Buy 30 November 12


11 April 13 7.00 Buy


198

Historical price target and rating changes for Kabel Deutschland Holding AG in the last 12 months (full coverage)


26 September 12 40.00 Sell 13 October 10


18 June 13 85.00 Hold

Historical price target and rating changes for KPN NV in the last 12 months (full coverage)


26 September 12 6.50 Sell 25 January 10

30 October 12 5.00 Sell

28 December 12 3.50 Sell

07 February 13 2.20 Sell

25 April 13 1.62 Sell


Historical price target and rating changes for Millicom International Cellular SA in the last 12 months (full coverage)

Date Price target - SEK Rating Initiation of coverage

10 July 12 730.00 Hold 26 September 11

08 August 12 660.00 Hold



20 December 12 620.00 Hold


08 March 13 545.00 Hold

09 April 13 550.00 Hold

18 June 13 540.00 Hold

Historical price target and rating changes for Mobistar SA in the last 12 months (full coverage)






18 June 13 16.00 Sell

Historical price target and rating changes for Portugal Telecom SGPS SA in the last 12 months (full coverage)


25 June 12 4.60 Buy 25 January 10



06 March 13 4.50 Buy

28 May 13 4.45 Hold

Historical price target and rating changes for QSC AG in the last 12 months (full coverage)


26 September 12 2.10 Hold 11 March 04


09 May 13 3.00 Buy


199

Historical price target and rating changes for Swisscom AG in the last 12 months (full coverage)

Date Price target - CHF Rating Initiation of coverage

26 September 12 390.00 Hold 02 June 11

06 March 13 385.00 Hold

18 June 13 408.00 Hold

Historical price target and rating changes for TalkTalk Telecom Group plc in the last 12 months (full coverage)

Date Price target - GBp Rating Initiation of coverage

26 September 12 180.00 Hold 26 September 12


06 March 13 290.00 Buy

03 April 13 270.00 Hold

10 May 13 160.00 Sell

Historical price target and rating changes for TDC A/S in the last 12 months (full coverage)

Date Price target - DKK Rating Initiation of coverage

13 August 12 50.00 Buy 03 December 10


06 March 13 46.50 Buy


24 May 13 42.00 Sell

Historical price target and rating changes for Tele2 AB in the last 12 months (full coverage)


25 June 12 148.00 Buy 12 October 10




28 March 13 105.00 Buy

26 April 13 104.00 Buy

Historical price target and rating changes for Telecom Italia SpA in the last 12 months (full coverage)






Historical price target and rating changes for Telecom Italia SpA (sav.) in the last 12 months (full coverage)







200

Historical price target and rating changes for Telekom Austria Group in the last 12 months (full coverage)


24 July 12 7.20 Hold 02 September 11

22 August 12 6.20 Sell


29 October 12 5.00 Sell


18 June 13 4.20 Sell

Historical price target and rating changes for Telenet Group Holding NV in the last 12 months (full coverage)


25 June 12 27.00 Sell 24 November 10

13 August 12 27.00 Hold




Historical price target and rating changes for TeliaSonera AB in the last 12 months (full coverage)


02 August 12 47.00 Sell 16 September 10


12 October 12 51.00 Hold





31 May 13 42.00 Hold

Historical price target and rating changes for United Internet AG in the last 12 months (full coverage)


11 September 12 17.00 Hold 23 October 03




03 June 13 26.00 Buy

Historical price target and rating changes for Vodafone Group plc in the last 12 months (full coverage)




Historical price target and rating changes for Zon Multimedia SGPS SA in the last 12 months (full coverage)


20 June 12 2.80 Buy 26 September 11





201

Berenberg distribution of ratings and in proportion to investment banking services

Buy 41.28 % 46.88 % Sell 19.96 % 12.50 % Hold 38.76 % 40.63 %

Valuation basis/rating key

The recommendations for companies analysed by the Bank’s equity research department are either made on an absolute basis (“absolute rating system”) or relative to the sector (“relative rating system“), which is clearly stated in the financial analysis. For both absolute and relative rating system, the three-step rating key “Buy”, “Hold” and “Sell” is applied. For a detailed explanation of our rating system, please refer to our website at

http://www.berenberg.de/research.html?&L=1

NB: During periods of high market, sector or stock volatility, or in special situations, the rating system criteria as described on our website may be breached temporarily.

Competent supervisory authority

Bundesanstalt für Finanzdienstleistungsaufsicht -BaFin- (Federal Financial Supervisory Authority), Graurheindorfer Straße 108, 53117 Bonn and Marie-Curie-Str. 24-28, 60439 Frankfurt am Main, Germany.

General investment-related disclosures Joh. Berenberg, Gossler & Co. KG (hereinafter referred to as „the Bank“) has made every effort to carefully research all information contained in this financial analysis. The information on which the financial analysis is based has been obtained from sources which we believe to be reliable such as, for example, Thomson Reuters, Bloomberg and the relevant specialised press as well as the company which is the subject of this financial analysis. Only that part of the research note is made available to the issuer (who is the subject of this analysis) which is necessary to properly reconcile with the facts. Should this result in considerable changes a reference is made in the research note.

Opinions expressed in this financial analysis are our current opinions as of the issuing date indicated on this document. The companies analysed by the Bank are divided into two groups: those under “full coverage” (regular updates provided); and those under “screening coverage” (updates provided as and when required at irregular intervals).

The functional job title of the person/s responsible for the recommendations contained in this report is “Equity Research Analyst” unless otherwise stated on the cover.

The following internet link provides further remarks on our financial analyses: http://www.berenberg.de/research.html?&L=1&no_cache=1

Legal disclaimer This document has been prepared by Joh. Berenberg, Gossler & Co. KG (hereinafter referred to as „the Bank“). This document does not claim completeness regarding all the information on the stocks, stock markets or developments referred to in it. On no account should the document be regarded as a substitute for the recipient procuring information for himself/herself or exercising his/her own judgements. The document has been produced for information purposes for institutional clients or market professionals. Private customers, into whose possession this document comes, should discuss possible investment decisions with their customer service officer as differing views and opinions may exist with regard to the stocks referred to in this document.

This document is not a solicitation or an offer to buy or sell the mentioned stock.

http://www.berenberg.de/research.html?&L=1

http://www.berenberg.de/research.html?&L=1&no_cache=1


202

The document may include certain descriptions, statements, estimates, and conclusions underlining potential market and company development. These reflect assumptions, which may turn out to be incorrect. The Bank and/or its employees accept no liability whatsoever for any direct or consequential loss or damages of any kind arising out of the use of this document or any part of its content.

The Bank and/or its employees may hold, buy or sell positions in any securities mentioned in this document, derivatives thereon or related financial products. The Bank and/or its employees may underwrite issues for any securities mentioned in this document, derivatives thereon or related financial products or seek to perform capital market or underwriting services.

Analyst certification I, Paul Marsch, hereby certify that all of the views expressed in this report accurately reflect my personal views about any and all of the subject securities or issuers discussed herein.

In addition, I hereby certify that no part of my compensation was, is, or will be, directly or indirectly related to the specific recommendations or views expressed in this research report, nor is it tied to any specific investment banking transaction performed by the Bank or its affiliates.

I, Stuart Gordon, hereby certify that all of the views expressed in this report accurately reflect my personal views about any and all of the subject securities or issuers discussed herein.


I, Barry Zeitoune, hereby certify that all of the views expressed in this report accurately reflect my personal views about any and all of the subject securities or issuers discussed herein.


I, Usman Ghazi, hereby certify that all of the views expressed in this report accurately reflect my personal views about any and all of the subject securities or issuers discussed herein.


I, Wassil El Hebil, hereby certify that all of the views expressed in this report accurately reflect my personal views about any and all of the subject securities or issuers discussed herein.


Remarks regarding foreign investors The preparation of this document is subject to regulation by German law. The distribution of this document in other jurisdictions may be restricted by law, and persons into whose possession this document comes should inform themselves about, and observe, any such restrictions.

United Kingdom This document is meant exclusively for institutional investors and market professionals, but not for private customers. It is not for distribution to or the use of private investors or private customers.


203

United States of America This document has been prepared exclusively by the Bank. Although Berenberg Capital Markets LLC, an affiliate of the Bank and registered US broker-dealer, distributes this document to certain customers, Berenberg Capital Markets LLC does not provide input into its contents, nor does this document constitute research of Berenberg Capital Markets LLC. In addition, this document is meant exclusively for institutional investors and market professionals, but not for private customers. It is not for distribution to or the use of private investors or private customers.

This document is classified as objective for the purposes of FINRA rules. Please contact Berenberg Capital Markets LLC (+1 617.292.8200), if you require additional information.

Third-party research disclosures

Company Disclosures Telefónica SA no disclosures Belgacom SA no disclosures Telenor ASA no disclosures Cable & Wireless Communications plc no disclosures BT Group plc no disclosures Deutsche Telekom AG no disclosures Drillisch AG no disclosures Elisa Oyj no disclosures freenet AG no disclosures France Télécom SA no disclosures Iliad SA no disclosures Jazztel plc no disclosures Kabel Deutschland Holding AG no disclosures KPN NV no disclosures Millicom International Cellular SA no disclosures Mobistar SA no disclosures Portugal Telecom SGPS SA no disclosures QSC AG no disclosures Swisscom AG no disclosures TalkTalk Telecom Group plc no disclosures TDC A/S no disclosures Tele2 AB no disclosures Telecom Italia SpA no disclosures Telecom Italia SpA (sav.) no disclosures Telekom Austria Group no disclosures Telenet Group Holding NV no disclosures TeliaSonera AB no disclosures United Internet AG no disclosures Vodafone Group plc no disclosures Zon Multimedia SGPS SA no disclosures (1) Berenberg Capital Markets LLC owned 1% or more of the outstanding shares of any class of the subject

company by the end of the prior month.* (2) Over the previous 12 months, Berenberg Capital Markets LLC has managed or co-managed any public

offering for the subject company.* (3) Berenberg Capital Markets LLC is making a market in the subject securities at the time of the report. (4) Berenberg Capital Markets LLC received compensation for investment banking services in the past 12 months,

or expects to receive such compensation in the next 3 months.* (5) There is another potential conflict of interest of the analyst or Berenberg Capital Markets LLC, of which the

analyst knows or has reason to know at the time of publication of this research report.

* For disclosures regarding affiliates of Berenberg Capital Markets LLC please refer to the ‘Disclosures in respect of section 34b of the German Securities Trading Act (Wertpapierhandelsgesetz – WpHG)’ section above.


204

Copyright The Bank reserves all the rights in this document. No part of the document or its content may be rewritten, copied, photocopied or duplicated in any form by any means or redistributed without the Bank’s prior written consent.

© May 2013 Joh. Berenberg, Gossler & Co. KG

205

Contacts: Investment Banking

Equity Research E-mail: [email protected]; Internet www.berenberg.de

BANKS ECONOMICS MID-CAP GENERAL

Nick Anderson +44 (0) 20 3207 7838 Dr. Holger Schmieding +44 (0) 20 3207 7889 Gunnar Cohrs +44 (0) 20 3207 7894

James Chappell +44 (0) 20 3207 7844 Dr. Christian Schulz +44 (0) 20 3207 7878 Bjoern Lippe +44 (0) 20 3207 7845

Andrew Lowe +44 (0) 20 3465 2743 Robert Wood +44 (0) 20 3207 7822 Anna Patrice +44 (0) 20 3207 7863

Eoin Mullany +44 (0) 20 3207 7854 Stanislaus von Thurn und Taxis +44 (0) 20 3465 2631

Eleni Papoula +44 (0) 20 3465 2741 FOOD MANUFACTURING

Michelle Wilson +44 (0) 20 3465 2663 Fintan Ryan +44 (0) 20 3465 2748 REAL ESTATE

Andrew Steele +44 (0) 20 3207 7926 Kai Klose +44 (0) 20 3207 7888

BEVERAGES James Targett +44 (0) 20 3207 7873 Estelle Weingrod +44 (0) 20 3207 7931

Philip Morrisey +44 (0) 20 3207 7892

Josh Puddle +44 (0) 20 3207 7881 GENERAL RETAIL & LUXURY GOODS TECHNOLOGY

Bassel Choughari +44 (0) 20 3465 2675 Adnaan Ahmad +44 (0) 20 3207 7851

BUSINESS SERVICES John Guy +44 (0) 20 3465 2674 Sebastian Grabert +44 (0) 20 3207 7834

William Foggon +44 (0) 20 3207 7882 Daud Khan +44 (0) 20 3465 2638

Simon Mezzanotte +44 (0) 20 3207 7917 HEALTHCARE Ali Khwaja +44 (0) 20 3207 7852

Arash Roshan Zamir +44 (0) 20 3465 2636 Scott Bardo +44 (0) 20 3207 7869 Tammy Qiu +44 (0) 20 3465 2673

Konrad Zomer +44 (0) 20 3207 7920 Alistair Campbell +44 (0) 20 3207 7876

Charles Cooper +44 (0) 20 3465 2637 TELECOMMUNICATIONS

CAPITAL GOODS Louise Hinds +44 (0) 20 3465 2747 Wassil El Hebil +44 (0) 20 3207 7862

Frederik Bitter +44 (0) 20 3207 7916 Adrian Howd +44 (0) 20 3207 7874 Usman Ghazi +44 (0) 20 3207 7824

Benjamin Glaeser +44 (0) 20 3207 7918 Tom Jones +44 (0) 20 3207 7877 Stuart Gordon +44 (0) 20 3207 7858

William Mackie +44 (0) 20 3207 7837 Laura Janssens +44 (0) 20 3465 2639

Margaret Paxton +44 (0) 20 3207 7934 HOUSEHOLD & PERSONAL CARE Paul Marsch +44 (0) 20 3207 7857

Alexander Virgo +44 (0) 20 3207 7856 Jade Barkett +44 (0) 20 3207 7937 Barry Zeitoune +44 (0) 20 3207 7859

Felix Wienen +44 (0) 20 3207 7915 Seth Peterson +44 (0) 20 3207 7891

TOBACCO

CHEMICALS INSURANCE Erik Bloomquist +44 (0) 20 3207 7870

Asad Farid +44 (0) 20 3207 7932 Tom Carstairs +44 (0) 20 3207 7823 Kate Kalashnikova +44 (0) 20 3465 2665

John Philipp Klein +44 (0) 20 3207 7930 Peter Eliot +44 (0) 20 3207 7880

Jaideep Pandya +44 (0) 20 3207 7890 Kai Mueller +44 (0) 20 3465 2681 UTILITIES

Matthew Preston +44 (0) 20 3207 7913 Robert Chantry +44 (0) 20 3207 7861

CONSTRUCTION Sami Taipalus +44 (0) 20 3207 7866 Andrew Fisher +44 (0) 20 3207 7937

Chris Moore +44 (0) 20 3465 2737 Oliver Salvesen +44 (0) 20 3207 7818

Robert Muir +44 (0) 20 3207 7860 MEDIA Lawson Steele +44 (0) 20 3207 7887

Michael Watts +44 (0) 20 3207 7928 Robert Berg +44 (0) 20 3465 2680

Emma Coulby +44 (0) 20 3207 7821

DIVERSIFIED FINANCIALS Laura Janssens +44 (0) 20 3465 2639

Pras Jeyanandhan +44 (0) 20 3207 7899 Sarah Simon +44 (0) 20 3207 7830

SalesSpecialist Sales Sales E-mail: [email protected]; Internet www.berenberg.de

BANKS LONDON BENELUX

Iro Papadopoulou +44 (0) 20 3207 7924 John von Berenberg-Consbruch +44 (0) 20 3207 7805 Miel Bakker (London) +44 (0) 20 3207 7808

Matt Chawner +44 (0) 20 3207 7847 Susette Mantzel (Hamburg) +49 (0) 40 350 60 694

CONSUMER Toby Flaux +44 (0) 20 3465 2745 Alexander Wace (London) +44 (0) 20 3465 2670

Rupert Trotter +44 (0) 20 3207 7815 Sean Heath +44 (0) 20 3465 2742

David Hogg +44 (0) 20 3465 2628 SCANDINAVIA

INSURANCE Zubin Hubner +44 (0) 20 3207 7885 Ronald Bernette (London) +44 (0) 20 3207 7828

Trevor Moss +44 (0) 20 3207 7893 Ben Hutton +44 (0) 20 3207 7804 Marco Weiss (Hamburg) +49 (0) 40 350 60 719

James Matthews +44 (0) 20 3207 7807

HEALTHCARE David Mortlock +44 (0) 20 3207 7850

Frazer Hall +44 (0) 20 3207 7875 Peter Nichols +44 (0) 20 3207 7810 Sales Trading

Richard Payman +44 (0) 20 3207 7825 HAMBURG

TECHNOLOGY George Smibert +44 (0) 20 3207 7911 Paul Dontenwill +49 (0) 40 350 60 563

Jean Beaubois +44 (0) 20 3207 7835 Anita Surana +44 (0) 20 3207 7855 Christian Endras +49 (0) 40 350 60 359

Julia Thannheiser +44 (0) 20 3465 2676 Gregor Labahn +49 (0) 40 350 60 571

UTILITIES Paul Walker +44 (0) 20 3465 2632 Chris McKeand +49 (0) 40 350 60 798

Benita Barretto +44 (0) 20 3207 7829 Fin Schaffer +49 (0) 40 350 60 596

FRANKFURT Lars Schwartau +49 (0) 40 350 60 450

INDUSTRIALS Michael Brauburger +49 (0) 69 91 30 90 741 Marvin Schweden +49 (0) 40 350 60 576

Chris Armstrong +44 (0) 20 3207 7809 Nina Buechs +49 (0) 69 91 30 90 735 Tim Storm +49 (0) 40 350 60 415

Kaj Alftan +44 (0) 20 3207 7879 André Grosskurth +49 (0) 69 91 30 90 734 Philipp Wiechmann +49 (0) 40 350 60 346

Boris Koegel +49 (0) 69 91 30 90 740

CRM Joerg Wenzel +49 (0) 69 91 30 90 743 LONDON

LONDON Stewart Cook +44 (0) 20 3465 2752

Greg Swallow +44 (0) 20 3207 7833 PARIS Simon Messman +44 (0) 20 3465 2754

Laura Cooper +44 (0) 20 3207 7806 Christophe Choquart +33 (0) 1 5844 9508 Stephen O'Donohoe +44 (0) 20 3465 2753

Dalila Farigoule +33 (0) 1 5844 9510

CORPORATE ACCESS Clémence La Clavière-Peyraud +33 (0) 1 5844 9521 PARIS

LONDON Olivier Thibert +33 (0) 1 5844 9512 Sylvain Granjoux +33 (0) 1 5844 9509

Patricia Nehring +44 (0) 20 3207 7811

ZURICH

EVENTS Stephan Hofer +41 (0) 44 283 2029 Sovereign Wealth Funds

LONDON Carsten Kinder +41 (0) 44 283 2024 LONDON

Natalie Meech +44 (0) 20 3207 7831 Gianni Lavigna +41 (0) 44 283 2038 Max von Doetinchem +44 (0) 20 3207 7826

Charlotte Kilby +44 (0) 20 3207 7832 Benjamin Stillfried +41 (0) 44 283 2033

Charlotte Reeves +44 (0) 20 3465 2671

Hannah Whitehead +44 (0) 20 3207 7922

US Sales E-mail: [email protected]

BERENBERG CAPITAL MARKETS LLC

Member FINRA & SIPC Andrew Holder +1 (617) 292 8222 Kelleigh Faldi +1 (617) 292 8288

Colin Andrade +1 (617) 292 8230 Kieran O'Sullivan +1 (617) 292 8292

Cathal Carroll +1 (646) 445 7206 Emily Mouret +1 (646) 445 7204

Burr Clark +1 (617) 292 8282 Jonathan Saxon +1 (646) 445 7202

Julie Doherty +1 (617) 292 8228

telecommunications - startseite | berenberg · european telecommunications telecommunications 3...

Documents