HSBC - Abundant Scarcity

Global Telecoms, Media & Technology Equity February 2011
Abundant Scarcity
Pricing power is returning to telecoms
The capital intensity of fibre is now improving fixed-line competitive dynamics and pricing conditions Mobile capacity constraints are starting to make their effect felt, both in terms of capex and better tariffs European cable operators, mobile players (Vodafone) and some vendors (Ericsson) are best placed
Disclosures and Disclaimer This report must be read with the disclosures and analyst certifications in the Disclosure appendix, and with the Disclaimer, which forms part of it
Telecoms, Media & Technology Equity 16 February 2011
abc
Summary
A degree of pricing power is (at long last) becoming apparent in the telecoms sector, thanks to scarcity emerging as a factor on both the fixed-line and the mobile sides of the industry. In fixed line, the capital required in the shift from copper-based infrastructure to fibre platforms is re-asserting the importance of scale at the expense of the unbundlers, and resulting in a more benign pricing environment. Meanwhile, in mobile, the intrinsically finite nature of cellular resource has already led operators to begin rationing capacity on price. In view of this new-found abundant scarcity, we believe that the prospects for monetising connectivity services look very encouraging. Alas, we are much less sanguine about the operators ability to monetise the applications that run over this connectivity (such as IPTV or mobile apps). In our opinion, the connectivity opportunity will have to suffice but it is enough.
Abundant scarcity
One way to think about the telecoms sectors complexities is to break it down into its constituent portions. A popular division is frequently made between the fixed-line (wireline) and mobile (wireless) portions of the industry. We would overlay this with a second distinction: between the provision of connectivity (ie raw bandwidth or capacity) and applications (ie the services that exploit the underlying connectivity). Combined, this approach yields a two-by-two grid that acts as a convenient means of analysing the sector. Focusing first on the connectivity layer, our argument is that there is much to play for. The key quality to bear in mind is scarcity. We believe that this vital ingredient has been largely missing in both the fixedline and mobile elements of the sector over the last decade, but is now making a reappearance; as a consequence, we think that telecoms should at long last begin to enjoy a measure of pricing power. This is of tremendous importance: volumes in telecoms traditionally look after themselves; the problem is that price deflation is commonly still more powerful, with the result that revenues decline. But if prices fall less rapidly, the growth in volumes actually has the chance to translate into growth in revenues.
Mobile connectivity
Take the mobile subsector. Although this part of the industry enjoyed heady growth in the late 1990s, since then the relentless deflationary nature of pricing has taken its toll. However, in a series of research
abc
reports from late 2009 through 2010 (The Capacity Crunch, Frequonomics, The Cell Side, SuperFrequonomics), we have argued that scarcity is at long last making an appearance. The physics of mobile telecommunications (specifically Shannons Law) mean that the amount of capacity available is finite. Even new technologies like 4G/LTE, while offering some efficiency gains versus their predecessors, are not sufficient an answer to the explosive demand seen from mobile data capacity. As a result, we have argued that capex is likely to trend upwards, but also that operators will need to ration their scarce capacity by means of price in other words, operators will enjoy a measure of pricing power. Although this view remains very controversial, supporting evidence continues to accumulate. For example, at its interim results, market bellwether Vodafone indicated the need for a modest increase in capex, but at the same time not only raised its revenue guidance but also announced it was on-track to introduce tiered tariffs in all its European markets by the end of calendar Q1 2011. And while, admittedly, plenty of questions remain about the profitability of data services, we would highlight the excellent margins achieved by eMobile (one of the worlds very few data-only wireless operators).
Fixed-line connectivity
Although much of our 2010 thematic research focused on the mobile subsector, a parallel set of points could be made about the fixed-line side of the industry: here too, the power of scarcity is finally making itself felt, even if its origins are very different. To be clear, the scarcity involved is certainly not related to the bandwidth available down superfast pipes: instead, it is associated with the prodigious capex required to deploy such technologies, which limits the number of such competing platforms in any one market. Clearly, in one sense, the scarcity of access infrastructure platforms is nothing new. But, over the last decade, regulators have undermined this barrier to entry by enabling entrants to unbundle the incumbents infrastructure at a price related to the incumbents own unit costs. Not only has this meant competitors have had no need to secure billions in funding to roll out infrastructure of their own, but it also ensured that they could behave, in terms of pricing, as if they had the same scale as the incumbent. With network reach and economies of scale no longer points of differentiation, the competitive battleground moved on to areas like marketing, where nimble entrants have tended to get the better of the incumbents. But now the shift towards next generation access (NGA) platforms providing superfast broadband gives an opportunity to redress the balance, and reassert the importance of scale. In the present-day, copperbased ADSL broadband world, unbundling has required little capital. By contrast, in networks that mix fibre and copper (like FTTN/VDSL), unbundling is intrinsically expensive (as a result of the need to install electronics not at a handful of local exchanges but instead at tens of thousands of street cabinets); while in those networks that consist entirely of fibre, it is currently practically impossible. We therefore
Breaking down the telecoms sector into its component platforms and services, with examples of each category
Applications layer Connectivity layer
Eg IPTV Eg superfast broadband Fixed-line network
Eg mobile apps Eg tiered data plans Mobile network
Source: HSBC
abc
believe that the landscape of competition is set to change, moving away from an unbundled framework and towards being based around wholesale offerings. Crucially, the economics of the latter are radically better than those of the former, so far as incumbents are concerned. Indeed, there are already indications that this transition is underway, as well as that pricing conditions are improving as a result. Naturally, there is the risk that regulators turn tough on the wholesale pricing of fibre-based connectivity. However, to do so would be to jeopardise the roll-out of this infrastructure in less economically attractive areas, thereby creating a digital divide. Our survey of recent regulatory developments points to an environment much improved by comparison with that pertaining to copper infrastructure, though still disappointing in some respects. On the closely related issue of net neutrality, a trans-Atlantic divide is evident, with European bodies adopting a more pragmatic and telecoms-sympathetic stance. We therefore believe that there is, for the first time in recent history, abundant scarcity evident in the telecoms connectivity layer, on both the fixed-line and mobile sides. Unfortunately, though, we think that prospects in the industrys other layer applications are rather less rosy.
Mobile applications
Almost wherever one looks, the telecoms players have failed in their ambitions to become providers of applications. Perhaps the most high profile instance of this has been in the mobile space, where the very term application is now synonymous with Apple and its iPhone. And while Apple does face competition in this area, the most conspicuous source of rivalry is from Googles Android platform, rather than from telecoms companies. The risk here is that, by losing control of the applications layer, the operators will find themselves disintermediated from some of their most traditional activities. For example, consider the rise of social networking sites, and the way that they might very well become the obvious address book intermediary for users looking to contact one of their social circle. Any software like Facebook, which effectively organises an individuals contacts, is in a strong position to help guide how communications between those individuals takes place and could readily direct a call to take place via a VoIP application, like Skype. Efforts by telecoms operators in the mobile space to meet this threat, and bring address book management out of the era of the telephone directory and into the modern age (via platforms like RCS, the Rich Communications Suite) look to be too little too late. The dangers here will be self-evident, although it should be emphasised that even service such as VoIP still rely on underlying connectivity. This remains the preserve of the telecoms operators, and is not territory that we see players in the applications layer as having either the skills or the appetite to pursue. Even modest or oblique encroachments in this area by the vendors or technology companies (such as by setting up as a MVNO or introducing soft SIM functionality) would likely be dilutive and/or counter-productive.
Fixed-line applications
Companies like Skype and Facebook are generally termed over-the-top (OTT) providers, since they offer an application that runs over the top of the connectivity that the telecoms operators deliver. In the mobile subsector, it is already difficult to avoid the conclusion that this is the natural arrangement of things. However, OTT is also becoming increasingly relevant in the fixed-line market, especially with reference to payTV services a topic closely examined in our thematic report, DisContent (September, 2009). Since this publication, ever more parties have entered the fray, with recent initiatives from players
abc
as diverse as Google and Tesco. We suspect that even established media players will struggle in this environment, let alone telecoms operators. Arguably the incumbents, by investing in what are termed content delivery networks (CDNs), are themselves tacitly acknowledging this state of affairs, by looking to provide the hosting services required by OTT providers of video content. In this confusing environment, we would argue that there remain only two fixed poles: firstly, customers want the best content; and, secondly, this content is in need of delivery (with content here referring to any application, whether IPTV, VoIP offerings like Skype, iPhone apps, etc). So, content is king, but it cannot be teleported to the customer: it requires delivery and hence connectivity. We would argue that, at the very least in the developed world, it is time the telecoms operators embraced the realities positive and negative of the markets structure. The industry is well placed to be able to monetise connectivity; it is much less clear what it can really add in the applications layer. Admittedly, there is an argument for a degree of continued activity in this latter space, if only to retain bargaining power with the other market participants and to stimulate the uptake of the connectivity services upon which applications depend. But it is perhaps time that operators embraced the fact that one of the fastest and most efficient means of encouraging the uptake of new forms of connectivity (such as superfast broadband and mobile data packages) is to adopt open platforms, as it is these that best foster innovation. Ironically, the appeal from a profitability perspective of applications like IPTV has long been dubious. Although this service might generate healthy revenues, the bulk of these would need to be passed onto the content owners. As a result, whether or not a telecoms company owns an IPTV revenue stream matters less, in our view, than the fact that this service is likely to promote adoption of superfast broadband connectivity a field where the operator is much more clearly poised to benefit. Moreover, third parties might be less anxious of the need to push net neutrality regulation if they could be confident that prioritisation would not be used to favour the incumbents own retail services in the applications layer.
Winners and losers

Clearly, the emergence of pricing power in telecoms should be a powerful positive right across the sector especially after so long an absence. However, because this positive is concentrated in the connectivity layer, the benefits are by no means evenly distributed. In terms of the fixed-line subsector, we believe that the best-placed operators are those with NGA infrastructure. Incumbents furthest ahead deploying this network will be the first to benefit (Swisscom, KPN, Bezeq and Verizon being good examples), but we think that the greatest benefits will actually accrue to the cable operators (KDG, Liberty Global, Telenet, Virgin Media and ZON; as well as, in a sense, TDC), as their upgrade path via DOCSIS3.0 is relatively inexpensive as well as rapid to deploy. In terms of the mobile industry, we favour larger operators with the necessary scale to out-invest their smaller rivals (and thereby provide superior a quality of service with their data offerings), as well as those equipment manufacturers that will supply them. As a consequence, we prefer networks like Vodafone, Telenor , NTT DoCoMo and KT, plus Ericsson among the vendors. In the emerging markets, the dynamics are somewhat different. There should still be the opportunity to monetise scarcity in terms of connectivity, and hence we would suggest stocks like TIM Brazil. However, additionally, there may still be some scope for operators to capture a portion of the applications layer: we would advocate STC and MTN on this basis. We see the losers on this theme as being those companies that have been slow to invest in their core connectivity capabilities, like TI and TPSA.
abc
Global Telecoms: Coverage universe

_______________________Overweight________________________ ___________________ Neutral _____________________ _________________Underweight __________________ Company Currency TP Price Company Currency TP Price Company Currency TP Price Axiata CCS China Tel DiGi.com Indosat NTT DoCoMo PT Telkom KT LGT U+ SmarTone Telstra Tulip Tel Baidu.com* CITIC* Comba T. Sys* eAccess* Netease.com* Tencent* XL Axiata* MYR HKD HKD MYR IDR JPY IDR KRW KRW HKD AUD INR USD HKD HKD JPY USD HKD IDR 5.8 4.6 5.1 29.0 5800 172000 9600 60000 8900 26.0 3.4 235.0 158.0 2.7 11.1 79000 54 240 7700 5.0 5.0 4.4 25.6 5000 151100 7600 41200 5970 24.8 2.9 164.6 129.6 2.6 8.1 51400 44 197 5200 China Mob KDDI M1 NTT SingTel SKT RCOM SKB CHT SoftBank TWM Maxis PCCW StarHub Tel Malaysia Bharti* Idea* J:COM* Perfect World* Shanda* ZTE* Comstar Etisalat Qtel Magyar T. MarocTel Mobinil Safaricom Zain KSA Telkom SA Turkcell Vodacom Rostelecom* VimpelCom* HKD JPY SGD JPY SGD KRW INR KRW TWD JPY TWD MYR HKD SGD MYR INR INR JPY USD USD HKD USD AED QAR HUF MAD EGP KES SAR ZAR USD ZAR USD USD 81.0 525000 2.5 4000 3.3 177000 152.0 6200 91.2 3200 69.0 5.8 4.0 2.9 3.9 370.0 75.0 97000 33.0 43.0 32.0 7.5 12.0 192 575 160 185 5.3 8.5 40.0 20.6 83.0 29.6 18.0 73.1 527000 2.4 3925 3.0 162500 97.2 5010 86.7 2997 66.7 5.4 3.7 2.6 3.9 318.9 63.7 86900 21.4 44.8 30.3 6.7 11.1 171.5 537.0 154.2 133.2 4.4 7.9 34.0 16.0 75.3 33.1 14.4 FET China Uni GTL Infrastructure* MTNL* Tata Com* Tata Tele* TWD HKD INR INR INR INR 39.0 6.7 39.0 59.0 225.0 15.0 42.2 13.5 36.7 42.5 210.1 15.4
Asia
Bezeq Cellcom EE (Mobily) Millicom MTN MTS Oman Tel Partner Sonatel STC TefO2 CZ Tel Egypt Wataniya Orascom T.* Sistema* Aegis Group Atos-Origin Bouygues Capgemini Ericsson Eutelsat Freenet FT Havas Iliad Inmarsat KPN Lagardere Meetic Reed Elsevier Ses Sa Swisscom TDC Telefonica Telenet Telenor Virgin Media Vivendi Vodafone WPP Group ZON C&W* Drillisch* Jazztel* Kabel Deutschland* United Bueiness Media* United Internet* XING* Liberty Global Tim Part Verizon Vivo Part
ILS ILS SAR USD ZAR USD OMR ILS XOF SAR CZK EGP KWD USD USD GBP EUR EUR EUR SEK EUR EUR EUR EUR EUR GBP EUR EUR EUR GBP EUR CHF DKK EUR EUR NOK USD EUR GBP GBP EUR GBP EUR EUR EUR GBP EUR EUR USD BRL USD BRL
11.60 135.0 71.0 110.0 148.0 29.0 1.5 90.0 175000 48.0 450.0 21.0 2.2 6.2 33.0 1.7 43.0 41.0 43.0 95.0 33.0 10.0 21.0 4.9 103.0 8.8 15.0 37.0 22.5 6.6 23.0 470.0 58.0 22.0 34.0 111.0 33.0 28.0 2.3 8.4 4.4 0.8 6.0 4.0 44.0 7.5 14.0 36.0 44.0 7.5 41.0 69.0
10.11 117.5 54.8 92.8 126.0 19.8 1.2 70.6 160000 40.3 400.0 16.0 1.8 3.3 25.0 1.5 41.4 33.7 39.6 82.1 28.8 8.6 16.2 4.2 77.2 7.2 11.8 33.1 16.8 5.9 18.5 430.9 46.3 18.3 31.2 91.5 27.4 20.8 1.8 8.3 3.8 0.5 7.0 3.9 38.1 7.2 12.6 39.3 42.3 6.0 36.4 54.1
TPSA Turk Tel Zain Group*
PLN TRY KWD
15.3 6.8 1.2
16.6 7.1 1.4
CEEMEA
Belgacom BT JC Decaux OTE PagesJaunes Portugal T. Publicis SAP Tele2 TeliaSonera Tel. Austria Informa Group* Forthnet* Telecinco* Versatel*
EUR GBP EUR EUR EUR EUR EUR EUR SEK SEK EUR GBP EUR EUR EUR
30.0 2.0 23.0 6.0 9.0 9.0 40 40 160 58 11.0 4.6 0.8 9.2 5.5
27.0 1.9 24.3 7.6 6.9 8.6 40.9 44.4 148.7 54.5 10.5 4.6 0.6 9.9 5.8
BSkyB DT Elisa ITV Mediaset Mobistar Nokia Fastweb Pearson TI COLT* QSC*
GBP EUR EUR GBP EUR EUR EUR EUR GBP EUR GBP EUR
4.8 9.5 15.0 0.6 4.3 42.0 6.5 18.0 9.3 1.1 0.8 1.4
7.5 9.9 16.8 0.9 4.9 45.6 7.0 18.0 10.6 1.1 1.5 3.0
Europe
AMX AT&T Inc. Tele Norte Telesp Sprint Nextel*
USD USD BRL BRL USD
64.0 30.0 31.0 41.0 5.0
56.8 28.5 26.1 39.5 4.6
Source: HSBC estimates, priced as at 11 February 2011
US & Latam
Net Serv Telmex
BRL USD
19.0 16.0
18.5 17.4
abc
Contents
Fixed-line connectivity Mobile connectivity Fixed-line applications Mobile applications Winners and Losers Company profiles
America Movil (Neutral, TP USD64) AT&T (Neutral, TP USD30) Belgacom (Neutral, TP EUR30) BT Group (Neutral, TP 200p) Bezeq (Overweight, TP ILS11.6) Deutsche Telekom (Underweight, TP EUR9.5) eAccess (Overweight (V), TP 79,000) France Telecom (Overweight, TP EUR21 KPN (OW, TP EUR15) KT Corp (Overweight, TP KRW60,000) Mobily (Overweight, TP SAR71) MTN (Overweight, TP ZAR148) Mobile Telesystems (Overweight, TP USD29)
7 28 54 67 74 97
98 100 102 104 106 108 110 112 114 116 118 120 122
NTT DoCoMo (Overweight, TP JPY172,000) Portugal Telecom (Neutral, TP EUR9) Saudi Telecom Company (Overweight, TP SAR48) Sprint Nextel (Neutral (V), TP USD5) Swisscom (Overweight, TP CHF470) TDC (Overweight, TP DKK58) Tele2 (Neutral, TP SEK160) Telecom Italia (Underweight, TP EUR1.05) Telefonica (Overweight, TP EUR22) Telekom Austria (Neutral, TP EUR11) Telenor (Overweight, TP NOK111) TeliaSonera (Neutral, TP SEK58) Telstra (Overweight, TP AUD3.40) TIM Participaes (Overweight, TP BRL7.50) TPSA (Underweight, TP PLN15.3) Turk Telekom (Underweight, TP TRY6.80) Verizon (Overweight, TP USD41) Vodafone (Overweight, TP 230p)
124 126 128 130 132 134 136 138 140 142 144 146 148 150 152 154 156 158
Disclosure appendix Disclaimer
161 164
abc
NGA already resulting in competitive landscape shifting to wholesale EC regulation more sympathetic to NGA, but concerns remain Net neutrality threat receding in Europe; CDN opportunity emerging
Introduction
There are now the first clear signs emerging that the transition to next generation access (NGA) superfast broadband is shifting the competitive landscape away from unbundling and towards wholesale. This should provide incumbents (as well as their cable rivals) with a considerably better operating environment than has been the case during the last decade, when the highly corrosive regulatory intervention that is the unbundling of the local loop (ULL) dominated the picture. Generally, it is only fair to acknowledge that European regulation is becoming more sympathetic to those deploying fibre upgrades, although some of the detailed terms still fall far short of what the Federal Communications Commission (FCC) offered to supercharge NGA roll outs in the US. However, the situation is reversed on the topic of net neutrality, where it is the Europeans, in our opinion, who have adopted the more rational policy accepting that there are actually benefits to being able to prioritise certain types of data traffic, provided the process is transparent and non-discriminatory. Such a pragmatic approach seems entirely appropriate for a wide variety of reasons. One worth mentioning is the operators increasing interest in entering the content delivery network
(CDN) space, which effectively signals their interest in hosting providers of over-the-top (OTT) services. Hence, far from using their network ownership to discriminate against thirdparty suppliers of services utilising their infrastructure, they are now specifically setting out to sell to such players.
Age of enlightenment
It is obviously ironic to even be discussing scarcity in the context of the tremendous bandwidths that NGA access network upgrades deliver. But the speeds involved do come at a price, particularly for incumbent operators that must generally replace all or part of their existing twisted-copper pair infrastructure with fibre. Historically, regulators have undermined the incumbents traditional barrier to entry (the enormous expense of replicating their access infrastructure) by making it available to third parties at a unit-cost based price. With the infrastructure (and its price) effectively common ground, the terms of competition have migrated elsewhere into areas like marketing and customer service. In such territory, it has all too often been the entrants rather than the incumbents that have held the upper ground. However, the shift towards NGA platforms looks set to redress the balance, and highlight once more the importance of scale in the telecoms industry.
abc
Todays unbundling platforms require relatively little by way of capital commitment, with altnets infrastructure deployment needing to be no more diffuse than encompassing the local exchange. But, in a FTTN/VDSL platform, their equipment must be installed in street cabinets of which there are commonly 20-40 for each local exchange. This implies vastly greater costs for the altnet to the extent that it is very difficult to see it being economically viable. For more information, see our own detailed report on NGA and its consequences, Age of Enlightenment, September, 2008; note that independent reports commissioned by a variety of regulators have reached similar conclusions in both Ireland and the Netherlands. Moreover, in the case of fibre-tothe-premises (FTTP), unbundling is presently practically unfeasible. Hence we conclude that the landscape of competition is likely to fundamentally change, shifting away from the unbundling of copper. Altnets that wish to continue to compete in the world of NGA-delivered superfast broadband will most likely need to purchase the connectivity wholesale from the incumbent. Wholesale pricing is naturally much more attractive to an incumbent than that associated with unbundling (as, with
wholesale, the incumbent is doing all the real work). Naturally, regulators could turn tough on wholesale pricing, but to do this would jeopardise the build out of fibre in less economically attractive regions. Once these projects are complete, the intensification of regulation does therefore represent a real risk, but given the duration of fibre deployments we believe that this is many years away. The process of migrating toward fibre has been a slow and painful one (we originally discussed it in our thematic report Phoenix rising from the ashes, all the way back in September, 2004). Operators have been wary of investing in upgrading their infrastructure, given the very poor reputation that capex in the telecoms sector acquired post the internet bubble or, more specifically, its burst. Moreover, with regulators aggressively pushing interventions like unbundling, management teams have understandably questioned the merits of investing in a platform that they might then be forced to sell at, or even beneath, cost. Then came the credit crunch, and arguably the worst global economic slowdown in three-
Future NGA and DOCSIS coverage plans (as % of total households) ( Incumbent vs cable)
90% 80% 70%
Mid 2011
2015e 2015e
2009
60% 50%
2012e
2012e
2009 2012e
2010e 2010e
2012e
40% 30% 20% 10%
2009 2009
FT plans: Fibre in all 96 homeland and in 3 overseas administrative districts by 2015 Numericable : No future coverage plans available
No cable operator in Italy 2015e
2009
0% Belgium Germany Netherlands Portugal Spain Sw itzerland UK France Italy
Incumbent
Cable (key competitor)
Source: company data, HSBC estimates
abc
quarters of a century. Although the free cash flow yields currently seen in the telecoms sector might give the impression that the industry is particularly fragile (given these valuation levels, the markets would certainly seem to perceive it as such), in fact the sector displayed a quite remarkable degree of resilience during the downturn. Admittedly, revenues were negatively impacted by the troubled macro environment; however, operators were able to offset this pressure by making cost savings with capex being a primary focus in this regard. Inevitably, such cuts had an impact on the pace at which operators invested in NGA upgrades, which are intrinsically expensive. In Europe, even those incumbents that were among the more enthusiastic (like KPN) adopted cautious roll-out schedules. Meanwhile, in the US, although Verizon pushed ahead with its highly ambitious FiOS project, AT&T extended its U-verse programme over a longer period, effectively slowing the pace of deployment. As we argued at the time (in our thematic Dj vu, February, 2009), the extent of the cuts possible in discretionary capex on a short-term view meant that the sectors cash flow generation could remain robust. The key phrase here, though, is short-term. Operators can indeed cut back on
capex over brief timescales a process that is made even easier if their primary rivals are engaged in the same behaviour. But it is dangerous to overly prolong this hiatus and, indeed, undesirable to do so if there is an appealing opportunity in the offing. We believe that NGA upgrades fall into this category. Despite the macro difficulties, these investments are now broadly underway, and some of the benefits that they bring to the competitive landscape are becoming apparent. The most important aspect of NGA builds, so far as we are concerned, are not that they will enable operators to drive retail ARPUs higher (though, of course, this is a distinct possibility), but rather that they encourage a shift away from unbundled-based competition towards that based on wholesale products. The latter are typically priced on terms far more attractive to the incumbent than the former.
Unbundled to wholesale
We have already seen several altnet competitors accepting the logic of this transition. For example, in an event of we would argue seismic significance (but one that went largely unrecognised by the financial press), UK broadband provider Orange (better known for its mobile network) decided to abandon its
and 2010: NGA drives consolidation
German broadband market in 2007
TEF O2 1% Cable 4% Versatel 3% freenet 7% Alice 12%
Vodafone 1%
others 1% DT 44%
Vodafone 14% TEF O2 8% Cable 13%
others 4%
DT 46%
Arcor 13%
UTDI 14%
Versatel 2%
Source: company data, HSBC
UTDI 13%
Source: company data, HSBC
abc
unbundled ADSL offering and move onto BTs wholesale platform. Admittedly, Orange is now a relatively small player in the UK broadband market (though once having been far stronger). Its DSLAM equipment may have been somewhat antiquated, and so may well have required significant investment even to maintain an unbundled ADSL service. The prospect of having to invest capex in an ADSL offering at just the time BT is beginning to deploy fibre was clearly an unattractive one given that any new assets Orange put in place would have a fairly limited lifespan (in the face of the incumbents faster NGA technology). Hence, even if Orange had invested in upgrading its unbundled ADSL platform, it would still have faced the dilemma of whether or not to invest a great deal more in a few years time. Maintaining its presence as an unbundler with ADSL would have been relatively inexpensive, but becoming an unbundler of BTs FTTN/VDSL platform would involve an immense outlay. And if this capital commitment was too large to contemplate, what would be the point in remaining an unbundler in the meantime? In the end, Orange presumably decided that, given it could not afford to unbundle the FTTN/VDSL platform, there was little point in continuing to invest in unbundling ADSL, and the company moved directly into taking a wholesale service.
The example provided by Orange proved to be an early warning of a broader trend. The choice confronting Orange arrived sooner than for most of its peers, owing to the age of some of its existing infrastructure, and therefore its need to make an investment to maintain even its existing, unbundled ADSL capability. However, the same issue would sooner or later confront every unbundler. TalkTalk remains an enthusiastic unbundler of BTs copper at the local exchange, but (although it is conducting a small-scale fibre trial) looks set to wholesale the incumbents NGA product rather than unbundle at the street cabinet level. Indeed, although Ofcom is insisting that BT make available unbundled FTTN/VDSL reference products, the regulator acknowledges that these have not attracted a great deal of attention. TalkTalk has made the point in its strategy presentation that it can still obtain the same, healthy margin reselling BTs wholesale NGA offerings that it achieves today with unbundled ADSL. BTs Openreach division, which has the task of deploying the NGA platform, needs to tread a difficult line in setting its prices. On the one hand, obviously, these must be sufficient to justify the investment made (and the not inconsiderable risk involved); on the other, these must not be so high as to deter take up. On TalkTalks suggested numbers, it can indeed maintain its gross margin, but the really standout aspect is surely rather the fact that its payment to
NGA builds encourage shift from unbundled products to wholesale
Status quo Status quo
NGA builds NGA builds
Copper lines Copper lines Lower cost Lower cost Credit crunch lower capex NGA upgrade Credit crunch lower capex NGA upgrade delayed copper still important delayed copper still important US-FCC: unbundling abandoned on networks US-FCC: unbundling abandoned on networks bringing fibre within 1,000 feet of customer homes bringing fibre within 1,000 feet of customer homes
Replace twisted copper with fibre Replace twisted copper with fibre FTTN/VDSL rollout to street cabinet FTTN/VDSL rollout to street cabinet Higher cost, but superfast broadband Higher cost, but superfast broadband Management wary Management wary Regulatory risk, but LT Regulatory risk, but LT Scale importance Scale importance Encourage shift from unbundled to wholesale Encourage shift from unbundled to wholesale
Source: HSBC
10
abc
BT approximately doubles when comparing wholesale NGA to unbundled copper. The utility of NGA upgrades to the incumbent should therefore be perfectly clear. The incumbents are the only operators with the necessary scale, capital and experience to undertake nationwide fibre upgrades; even unbundling such a platform becomes prohibitively expensive, and this should leave unbundled copper competitors shifting towards a wholesale NGA service, for which they will have to pay (quite properly) considerably more.
pricing power having found itself able to increase the pricing on its services for each of the last three years. In the course of 2010, ZON was able to raise its tariffs in the region of 3-4% across the board. This move was then followed by the incumbent, Portugal Telecom. PT was able to lift its prices because it has itself conducted a substantial NGA network upgrade. ZON expects to be able to increase its prices by about 1% in 2011, despite the austerity measures that have been implemented in Portugal and the 2ppt increase in VAT effective from January 2011. Turning to the UK, cable operator Virgin Media (VMED.O, USD27.38, OW) has been a frontrunner among its European peers in upgrading to DOCSIS3.0, which enabled it to offer a service that is clearly differentiated from that of the incumbent, BT. But BT is now also rolling out a NGA deployment of its own which seems to be persuading present-day unbundlers (like Orange) that a move towards wholesalebased services is inevitable. Note that despite the difficult macro conditions in the UK, Virgin Media announced in February 2011 a set of tariff increases in its TV packages: the price of its L package is to rise by GBP1.25 per month, while that of its XL package is to step up by GBP1.0 per month. In addition, its broadband packages are to cost an extra GBP0.75 per month. Conversely, though, the price of the 30Mbps package has been reduced from GBP20 to GBP18.5 per month. The above examples therefore suggest that investing in network upgrades creates the conditions necessary for the exercise of pricing power. In our view, failure to invest tends inevitably to condemn operators to pure pricebased competition. Arguably, one good example of this is Telecom Italia. On the plus side, the lack of a cable infrastructure in Italy has limited the damage that the companys decision to repeatedly postpone its NGA roll-out plans would otherwise
Incipient fixed-line pricing power

What is important to emphasise here is that the prerequisite for a more orderly pricing environment is the upgrade to NGA infrastructure (to fibre in the case of incumbents, or to DOCSIS3.0 on cable platforms). Unfortunately, because the process of migrating towards fibre platforms is so expensive and time-consuming, progress here has often been frustratingly slow. It is for this reason that the evidence for incipient pricing power emerging on the fixed-line side of the industry is much less clear cut than that seen in the mobile subsector. Nevertheless, we feel that the anecdotal evidence is increasingly convincing. The slow pace of incumbent NGA upgrades also explains why the clearest examples of pricing power in the fixedline market are often to be found among the cable operators, since their upgrade path to NGA status via DOCSIS3.0 technology is relatively swift. Those cable operators already using DOCSIS3.0 in significant portions of their footprint have been able to show a remarkably resilient top line, even during the difficult macro environment of the past few years. ZON Multimedia (ZON.LS, EUR3.78, OW) in Portugal provides an example of a cable operator that has been able to command a certain degree of
11
abc
have caused. However, the resulting lack of service differentiation has exposed TI to the aggressive pricing of altnets exploiting unbundling. For instance, in Q4 2010 Wind/Infostrada, Tele2 and Tiscali started offering double-play broadband products for a two year (sic) promotional price of EUR19.95 per month. TI has opted not to follow this move, but this decision is likely to have hit the performance of its domestic fixed-line business in Q4 2010. The same risk potentially applies to other operators that have elected to postpone NGA capex during the recent recession and so may also apply to the likes of Telefonica de Espana, for example. There is also the hint in some of those (rare) markets where the deployment of NGA could be relatively broad based that the elevated levels of capital involved (a particular step change so far as the incumbents competitors are concerned) will trigger more responsible pricing policies. This would seem to be the case in France, a market that has been the cheapest in Europe in terms of tripleplay services effectively ever since Iliad launched its EUR29.99 per month offer back in 2003. Its competitors were forced to align their pricing to Iliads, with only FT managing to retain some sort of premium to this. (Even this premium has been eroding, following FTs price cuts in summer 2010.) However, in Q4 2010, retail prices were raised by all the major operators, in response to a hike in the VAT rate. What is significant, though, is that the price rises were in excess of what the higher VAT charge alone would have justified, and therefore also represents an increase in the underlying pricing at Iliad and SFR. These operators can legitimately claim that they now incorporate additional features in their offerings, such as unlimited calls to mobile phones or a new set-top box but the fact remains that triple-play
packages now cost more on a headline basis than previously. For example, Iliad's new customers will be paying EUR37.97 per month (for its tripleplay, fully unbundled, unlimited calls to mobile package), compared with EUR29.99 previously. Of the total price increase, only a third can be attributed to the higher VAT rate. Similarly, at SFR, the new triple-play offer now costs EUR36.9 per month compared with EUR29.9 previously. The equivalent bundles would cost EUR39.9 per month at Bouygues (including three hours of calls to mobile, rather than unlimited calls) and EUR37 at Orange (but with only one hour of calls to mobile phones included, and excluding EUR3 for set-top box rental). Note that fibre is currently priced at the same level as ADSL in France, although it has not yet received any mass-market push. Hence, there remains the possibility that the operators will also attach a price premium to the superior bandwidth services that fibre can support. Finally, there is the example of the US. In this market, traditional fixed-line revenues have been under sustained pressure, driven in particular by line losses in the region of 8-10% per annum. An additional headwind arises from the weak economy. Nevertheless, the absence of regulatormandated wholesale or unbundled fibre products, as well as the growing popularity of services that exploit superfast broadband connectivity, do together have the potential to lead to some form of pricing power. Of course, because of the very different bandwidths available over fibre compared with copper, life-for-like price comparisons are somewhat spurious. Nonetheless, it is worth drawing attention to the positive nature of the share of wallet effects as seen at the RBOCs. The ARPU from Verizons FiOS customers in Q4 increased by 4% year-on-year, reaching USD146. Similarly, AT&Ts fixed-line revenues are
12
abc
Ray of light
One concern about any scenario in which the incumbents scale and ability to deploy capital work to its advantage must inevitably be that the regulator will intervene to the detriment of the incumbent. To an extent, this is inevitable. However, incumbents still have some powerful bargaining chips that should ensure they are more equitably treated than in the present, copperdominated world. The first point to make is simply to repeat that it will be very difficult at any conceivable price to make activities like unbundling at the street cabinet very financially attractive, because of the capital outlay involved (not to mention the logistical challenges). Most unbundled players have relatively limited experience with infrastructure, given that they are presently tasked with installing their DSLAMs in perhaps a few hundred local exchanges. They are not generally geared up for deployments that would involve visiting literally thousands of street cabinets. So, even though regulators will undoubtedly define products and pricing on unbundled FTTN/VDSL platforms, this does not imply that they will actually be used. There have been efforts in some countries by governments, notably in Australia, to wrest responsibility for upgrading the infrastructure to fibre away from the incumbent, namely Telstra. However, this has proven hugely controversial, time-consuming and complex. Most governments would likely prefer to see the work done by the
Telco office
FTTP
Telco office
Fibre optic
Copper
Source: HSBC
Incumbents, in our view, should be enthusiastic to build NGA platforms, provided the regulation is supportive. In practice, they likely will look to receive sympathetic treatment, via (for instance) flexibility on pricing, or if it must be dictated tariffs established with the use of generous cost of capital assumptions. Ultimately, once capital has been irrevocably committed, regulators have generally shown a tendency to turn more aggressive. But, in the meantime, while they are looking to see substantial quantities of capital deployed, they have an incentive to remain more generous. And this meantime could represent a considerable period fibre networks can take years to deploy, particularly if there are to be upgrades beyond FTTN/VDSL towards FTTP. Hence the incumbents should possess a bargaining tool that they can make use of for years to come. If regulators were to turn unduly harsh mid-build, this could potentially jeopardise further deployment of fibre. And because operators are likely to begin their roll out in the more affluent, densely populated areas, and only
E nd user building
End user building
benefiting from growth generated by its U-Verse connectivity offerings and related services. Fixedline consumer revenues per household have now been growing for the last 12 quarters in succession. Additionally, the evidence suggests that fibre offerings are helping to mitigate the traditional process of line loss.
local expert (ie the incumbent), using the latters access to capital (rather than potentially impacting an already over-extended state).
Fibre to the node vs fibre to the premises
FTTN
13
abc
then move onto less financially compelling regions, it becomes particularly important that regulatory decisions do not impair the profitability of earlier build which, in a sense, must crosssubsidise that which follows. In our view, the most enlightened approach (pun intended) is that adopted by the FCC in the US, which decided to abandon unbundling altogether on networks bringing fibre within 1,000 feet of the customer premises. This has encouraged the very rapid deployment of fibre NGA platforms in the US. European regulators have not taken this approach, and consequently have not triggered anything like the same speed of build out. However, regulators like Ofcom have at least conferred pricing flexibility on BT in other words, the incumbent has the freedom to set its own prices for both retail and wholesale NGA products.
argument for its adopting an active role in this process is that regulation in Europe might otherwise become fractured, given the wide variety of approaches national regulators have already adopted towards NGA. We would be tempted to see this as something of an advantage providing an opportunity to test out a variety of techniques, and then evolve regulation across the region towards the most successful variants. The Commission is adamant that the principle of the so-called ladder of investment must be maintained, despite the fact that it has been of only limited success to date. The idea here is to attract players into the market via products that require relatively little capital to exploit (ie wholesale offerings, where the incumbent service is simply resold), and then have them graduate towards taking on a greater degree of responsibility in terms of infrastructure as they gain in terms of size and experience (ie towards unbundled offerings, where the altnet provides the electronics and relies on the incumbent only for passive network components). The same concept was applied to ADSL broadband services, and indeed many altnets began with wholesale and later migrated to unbundling. However, there was no obvious next stage, given the enormous capital involved in access infrastructure. Arguably this is still very much the case. With respect to NGA, the regulatory mandated products cover everything from the basic passive infrastructure (eg trenches and poles) for those wishing to deploy their own fibre connections, up through unbundling (eg of the copper loop between the street cabinet and customer) for those looking to launch unbundled FTTN/VDSL, and culminating in wholesale offerings (ie where the reseller simply resells the incumbents service). But it will be a very tall order, in our view, for an altnet successful with a wholesale product (given the relatively thin margins that a relatively
Commission regulation
It will be interesting to see how this particular arrangement can co-exist with the line coming out of the European Commission. The Commissions earlier documents on the subject of NGA regulation were very worrying from the perspective of those hoping to see a supportive regime put in place to encourage fibre investment. For instance, the Commission has previously suggested that the use of fibre between local exchange and street cabinet was to be considered a conventional upgrade, and therefore not one that should attract any cost of capital premium. This despite the fact that there would be no conceivable to reason to make such an upgrade other than to offer superfast broadband services services that the Commission itself recognises entail taking on board a good deal of risk. The latest documents from the Commission show some relaxation of its attitude, although arguably still leave much to be desired from the perspective of a prospective investor. The Commissions
14
abc
undifferentiated product such as this implies) to make sufficient profit to justify committing the capital required for unbundling, let alone laying fibre of its own. More positively, the Commission does seem clear that part of its role is to encourage investment in network:
The [EC] Recommendation does emphasize the need for NRAs [national regulatory authorities] to reflect investment risk in terms of risk premia and, importantly, in terms of price flexibility
We feel it is a pity, though, that this approach is not more clearly reflected in its comments on FTTP unbundling. Clearly, all forms of NGA are risky, but the degree of capital required for FTTP puts it in a different league to even FTTN. Hence, while we see it as regrettable that European regulators have not followed the FCC in ruling out unbundling even on FTTN, we would concede that this form of regulatory intervention does make more sense applied to FTTN rather than to FTTP. Given the costs of FTTP, it might have been hoped that unbundling would have been ruled out. In fact, with todays technology, unbundling FTTP is not really technically practically feasible, but the Commission still seems determined that it ought to be introduced as soon as it becomes so.
cable companies typically lack the market share and geographic reach of the incumbents, and hence also their scale efficiencies. In our view, if the regulator really wants to encourage infrastructure-based competition, it ought to be pricing unbundled infrastructure at a unit cost price determined not from the incumbent, but rather from its smaller rivals. The smaller the rival selected, the greater the incentive for competitors to commit to investing in infrastructure. (It is also worth pointing out that the cable operators themselves might be tempted to extend their footprints were the potential returns adequately attractive). It is worth highlighting the potential injustice here. An altnet could theoretically simply resell a wholesale superfast broadband service at a price offering it a higher return than a cable operator could generate from its own infrastructure (which would be smaller scale and, thus, higher unit cost than the network of the incumbent that the reseller was wholesaling). This is hardly a circumstance likely to incentivise future investment. Ultimately, though, despite the discouraging nature of the Commissions philosophical approach, a great deal rests simply on the cost of capital that regulators apply to NGA investments. Here the Commission is clear (without prescribing actual figures) that the incumbent is entitled to a premium:
In cases where investment into NGAs depends for its profitability on uncertain factors such as assumptions of significantly higher ARPUs or increased market shares, NRAs should assesss whether the cost of capital reflects the higher risk of investment relative to investment into current networks based on copper.
Cost orientation
Another weakness, in our opinion, of the Commission's approach is the focus (for both unbundling and wholesale, as well as for FTTN and FTTP) on cost-orientated pricing. The problem here is that the regulator is referring to the costs of the incumbent the largest player in the market and therefore that with the lowest unit cost. This tends to penalise those competitors that have gone to the expense and trouble to deploy infrastructure of their own, the most prominent example of which are the cable operators. But
We have long argued that the immense cost of the upgrade to NGA platforms could be in part paid for out of the elimination of the artificial and
15
abc
transient arbitrage that is local loop unbundling. However, one alternative would admittedly be for regulators to choose to permit a greater share of the returns to accrue to the provider of the NGA infrastructure. The problem with this is simply that it is likely to translate into higher retail prices, which is likely to deter uptake thereby potentially damaging the economics of the entire programme. Like Ofcom, the Commission is also prepared to countenance a degree of pricing flexibility. This seems to boil down to permitting the incumbent a limited degree of flexibility in terms of pricing use of its FTTx infrastructure at a discount to those prepared to purchase it either on a longerterm basis or in higher volumes (since both reduce the risk of the NGA investment). However, it does not seem very apparent that the regulator is permitting the type of price differentiation that is commonly seen in other industries. In the scenario where a scarcity of capital on the scale required for superfast broadband platforms results in a diminishing number of infrastructure competitors, it is in theory possible to see the operators being able to introduce a broader range of price differentiation techniques. The most obvious mechanism available is to price differentiate according to the bandwidth provided, selling faster services at a premium. It is possible this could meet with regulatory approval, if only by virtue of the fact that the faster bandwidths possible over a FTTP platform as compared to FTTN/VDSL cost greatly more to provide, because of the need to take the fibre all the way to the customers premises. In a more conventional market, though, other forms of price differentiation would likely appear. For example, operators might sell tiers of data capacity, just as they are starting to do in the mobile side of the industry. The size of the tiers would be considerably greater for fixed-line
broadband than for mobile, but this might nonetheless be a good way of differentiating between casual users of broadband (who would tend to value the service less) and those reliant upon it satisfy all their entertainment needs, thereby consuming a great deal of capacityhungry video (who would tend to value the service more highly). However, it remains to be seen whether the wholesale pricing structures to be imposed by the regulators will support this type of price differentiation. This is despite the fact that it would perform a useful function: enabling operators to charge intensive users a figure more closely representative of the utility they derive from the service, and therefore helping to pay for the deployment of this controversial technology sooner than would be the case had the build out depended exclusively on the price that less committed users would be prepared to pay. The specific issue here is that there is not necessarily a substantial differential between the cost to supply a modest and a large quantity of data over a broadband network (given the essentially fixed-cost structure of the industry). A regulatory-imposed, cost-orientated wholesale pricing regime would therefore tend to undermine any attempt to price differentiate based on volumes, because a reseller could always undercut any materially premium-priced high-usage offering. While such a pricing regime might be justified, were the only goal to eliminate distortions relating to the misalignment of costs and pricing, it would also prevent useful price differentiation aimed at charging customers more closely in relation to their derived utility and thereby securing an improved revenue outlook of just the type that might encourage a faster deployment of NGA infrastructure.
16
abc
Could try harder

Overall, we would argue that the regulatory regime does show some limited signs of being more sympathetic, with the motive being to encourage incumbents to deploy NGA platforms. At the same time, we would continue to emphasise that there are aspects of the transition to NGA that are intrinsically unfavourable to unbundling (ie the expense of visiting the street cabinet). It is evident that regulators themselves do seem nervous that they might be perceived as becoming overly tolerant towards the incumbents, even though the concessions involved (as discussed above) would be solely aimed at encouraging an expensive, risky investment programme (ie NGA) while acknowledging the realities imposed by physics (ie the fact that unbundling is unlikely to be economic at the street cabinet level). It is this anxiety that is perhaps the best way to interpret innovations like the rebranding of Ofcoms Active Line Access (ALA) wholesale product as virtual unbundling. These labels refer to a sophisticated wholesale product that Ofcom intends to impose upon BT one which will enable the wholesaler to exert a good deal more control over its service than would be possible via equivalent ADSL products. This makes perfect
sense, but the decision to rechristen this service virtual unbundling when what is on offer is quite obviously a wholesale product looks decidedly odd. It is almost as if Ofcom felt the desire to retain the unbundling moniker in order to emphasise the fact that it is not relaxing its stance towards BT (a fact that no one would likely dispute in any case). Note that the European Commission has accepted Ofcoms proposal, but indicated that the virtual unbundling product will not be sufficient in itself. The Commission wants Ofcom to secure the unbundling of BTs NGA infrastructure as soon as this is practicable.
Net neutrality
However, there is one further area of activity where European regulators if not their US counterparts seem to be leaning towards a stance that really should be supportive of investment: on the fraught topic of net neutrality. The question here boils down to the degree of freedom that operators should have in how they manage and tariff their networks, and could have wide ranging implications in terms of the structure of the overall market and the degree of incentive operators are provided to invest in NGA platforms (and hence the likely speed of their deployment).
Operators can use deep packet inspection (DPI) technology to implement traffic management policies, such as prioritisation according to the requirements of the underlying service (e.g. delay-intolerant voice and video prioritised over email and browsing)
Node
Users Fixed/ mobile Intern et
Deep packet inspection

Header
Policy Engine (PE)

- Prioritize premiu m users - Prioritize b y service need - Block objectionable content
Payload
Source: HSBC
17
abc
We have never found the arguments for net neutrality to be in the least convincing (see Net Neutrality, April 2006). Unfortunately, this has not prevented the concept from gaining widespread acceptance in certain quarters, and with certain regulators especially the FCC in the US. The scope for conflicts of interest in this area will be readily apparent. Operators, faced with the colossally expensive challenge of rolling out superfast broadband networks would like maximum flexibility in terms of managing, and charging for, the use of their infrastructure. This boils down to two key areas: First, telecoms companies would like to manage their resources as efficiently as possible with the use of traffic management techniques. This would involve, for example, prioritising the packets of a time-sensitive video stream (time-sensitive because the viewer will be immediately conscious of any delays or interruptions) over that of an email (where the recipient is likely to be less sensitive to a modest delay); see the accompanying diagram. Second, the operators would like flexibility in charging for this prioritisation. This would involve not only applying a premium for the delivery of prioritised traffic, but also potentially billing not only the recipient but also the party sending the packets. Naturally, the internet players are concerned that their services might be de-prioritised in favour of services sponsored by the telecoms operator (rival video streaming services, for instance); and/or that they would start having to foot more of the bill for the ongoing explosion in data traffic volumes (rather than this being borne by the end-user). The FCC has been largely sympathetic to these worries - one of the reasons that we have become progressively less positive on the US RBOCs as
investments and more upbeat about prospects for the incumbents in Europe, where the regulation on net neutrality is looking more reasonable. It will not have escaped the FCCs attention that the US has produced a set of enormous internet powerhouses, such as Google and Amazon. Such companies can export their services across the globe in a way that domestic telecoms providers cannot. In terms of US industrial policy, therefore, it arguably makes sense to privilege the internet names over their telecoms peers. This is particularly the case given that much of the NGA fibre upgrade has already been completed in the US. In other words, the capital has already been committed, and hence the regulator no longer needs to create a more attractive returns environment in order to stimulate the relevant capex. One flashpoint has been the legal case embroiling the FCC and Comcast, the US cable giant. Comcast had been de-prioritising peer-to-peer traffic on its network from services like BitTorrent. The FCC objected to this, and intervened; in response, Comcast changed its approach but nonetheless introduced a 250GB usage cap on its customers. This move did not please the regulator, which then decided to sanction Comcast. The latter took the matter to court, winning the case on the grounds that the FCC lacked the necessary authority to intervene. The longer-term implications of this tussle, however, remain unclear. One risk is that it incentivises the FCC to secure the necessary authority via the statute book. However, the regulator has also been pursuing a more informal approach, with the aim of keeping regulation of the internet to a minimum (which seems to be something that most sides can at least agree is desirable). Nonetheless, the fact that this dispute entered the courts in the first place is surely somewhat
18
abc
alarming. Comcast acted to ensure that it could preserve a good quality of service on its contended cable network despite the excessive burdens placed on its platform by a small minority of its customers using peer-to-peer applications. Cable networks are intrinsically contended, as all customers on a given street share the same coaxial cable; hence capacity taken by one user is no longer available for another. In such circumstances, individual customers can degrade the service of their neighbours, and Comcasts actions were taken to prevent this. It is surely an added irony that probably the most popular use of the application in question peer-to-peering is video piracy.
offerings, voice, being the clearest example) need to be conveyed with priority over and above data traffic relating to, for example, an overnight routine system backup. Indeed, services that are really demanding in terms of bandwidth such as high-definition video will arguably never be able to mature successfully unless supported by techniques such as prioritisation. It is thus, ironically, actually in the internet players own interests that prioritisation be permitted, because otherwise they will likely find themselves practically constrained in terms of what they are able to offer customers with an adequate quality of service. Although net neutrality advocates often suggest to the contrary, the presence of prioritisation is actually the status quo. For example, to take a case reduction ad absurdum, a corporate taking a leased line is purchasing dedicated capacity between two points, in other words, full prioritisation. Note that traditional switched voice services are also, in effect prioritised, given that the circuit conveying the communications is held open between the callers for the duration of the call, with no other traffic being permitted over that portion of the pipe. It is further worth highlighting that the presence of net neutrality should not be thought of as necessarily helping the little guy. Net neutrality advocates often suggest that prioritisation charges would hamper the development of the next generation of internet entrepreneurs, but it is revealing that the current giants of this space invest prodigious quantities of capital in optimising their response times. For instance, the likes of Google have constructed numerous data centres at optimally located points on the global telecoms network in order to ensure their sites services are as responsive as possible. In other words, even in todays, supposedly net neutral world, the advantage lies with the scale players.
Atlantic divide
In Europe, circumstances are somewhat different to those across the Atlantic. Most of the investment in NGA upgrades has yet to be made, and incumbents are cautious about the economic case for fibre. Given this environment, it makes sense to give telecoms players as much freedom as possible to create business models that will support NGA deployment. Europes politicians have made it perfectly plain that fibre is a priority, out of fear that the continents economy might fall behind; supporting NGA investment will arguably be more important (in their eyes) than ensuring that existing internet giants largely foreign in origin enjoy the benefits of connecting to their end users at the minimum price. Ultimately, we remain of the view that hard-line net neutrality is simply unworkable. Even in the fixed-line network, there will always be bottlenecks where capacity is scarce. It is simply not possible to build a network that at all times and in all places will be free of constraints. Given this reality, it will always be necessary to apply some traffic management policies. Packets belonging to services that are particularly timesensitive to delay (with that most traditional of
19
abc
Admittedly, to agree that there ought to be some form of traffic management does not necessarily imply that a charging structure should be employed. It might be conceptually possible to introduce a hierarchy, where all video enjoyed priority over voice, voice over emails, and emails over backups (for example). But there are obvious problems with this arrangement. For instance, a corporates voice communications might be held to be more important than a schoolchild streaming YouTube videos. Of course, the conventional method employed to allocate scarce resources (at least in capitalist societies) is to use a pricing mechanism. This gives customers the ability to ascribe a value to the timeliness of the delivery of their data. Without such a pricing mechanism acting as a deterrent, the likelihood is that all traffic would soon become flagged as highest priority video, rendering the whole system otiose. There is also the broader point, well made by Ofcom in its recent report Traffic Management and net neutrality (24 June 2009), that it is hardly desirable that regulators determine from the very outset what business model an industry should pursue. Ofcom cites the example of newspapers, which charge not only their readers (ie the purchase price) but also businesses wanting to reach those readers (ie for advertising). In effect, the advertising cross-subsidises the purchase price of the newspaper. To apply a similar template to, say, an IPTV service, the cost might be borne partially by the viewer (via a subscription charge) and partly by the corporate owning the content (via prioritisation fees). This is not so very different from the way in which traditional broadcasters work today, since they already pay fees to those operating the broadcast platform (whether terrestrial or in geosynchronous orbit).
Non-discrimination and transparency

European NRAs seem to be gravitating towards a relatively hands-off approach, while at the same time making it clear that they will be imposing demands in terms of non-discrimination and transparency. On the former issue, ARCEP (the French regulator) has made use of a powerful analogy, comparing the situation to that of a toll road. ARCEP observes that toll roads price differentiate price between cars and trucks, on the grounds that the latter are more demanding than the former (since trucks will cause more wear to the road, and thus require higher maintenance to support). However, toll roads are not able to discriminate between, say, a red truck and a blue truck. Translating this into the world of telecoms, this suggests that a network should be able to price delivery of packets that are part of a high-priority, time-sensitive video stream over and above those that comprise an email. However, it would not be permitted to discriminate between a video packet originating from an incumbent retail divisions IPTV product over one originating with the equivalent service of a rival (assuming, of course, that the two were paying for the same level of video prioritisation). This is, in our opinion, the real risk that regulators are entitled to address, rather than the excitable warnings about the emergence of a new form of censorship that are heard from some quarters. (For instance, certain of the shriller voices on the net neutrality side have even claimed that, were net neutrality ignored, Republican-leaning political blogs would be quicker to access than Democrat-leaning sites; such claims seem far fetched). It will be observed that the debate over nondiscrimination is therefore in essence merely another repeat of the time-worn infrastructure dominance leads to downstream market advantage concern, which has been frequently
20
abc
dealt with in the regulation of incumbents. (A good example being the fact that many incumbents retail arms are obliged to purchase ADSL services from their associated wholesale divisions on precisely the same terms as thirdparty resellers). European regulators like Ofcom do, however, flag the fact that the pricing of prioritisation could in future become a concern to them. For instance, if access to the end consumer was controlled by a relatively small number of parties (quite possible in a NGA world), this might facilitate overcharging for prioritisation which would, in turn, necessitate regulatory action. But, as Ofcom acknowledges, this is certainly not an issue at present. The above said, being permitted by the regulator to charge for prioritisation and actually being able to successfully push through such tariffs are two entirely different things. So, while the regulator might be happy for an incumbent to charge for prioritisation, this is not to say that the latter would actually be able to persuade, say, Google to pay such a fee for streaming HD video. And if an operator was unable to make Google pay, then (on grounds of non-discrimination) it would find it difficult to force others with HD video content to do so. Any broadband provider that refused to provide prioritised connectivity to Google (since the latter would not pay the relevant premium) would obviously risk losing many of its customers who might well churn to any competitor that was prepared to deliver the necessary prioritised bandwidth without the incremental tariff (presumably with the goal of taking market share). This could prove a stiff test, though the concentration of the infrastructure market that we envisage with the evolution towards NGA platforms should help. The capital intensity involved in this transition should concentrate the
market, and with fewer participants (each of which will have committed very substantial capital to its access network), it is easier to envisage how an element of prioritised tariffing might be introduced. It would likely be easiest to charge incrementally for the prioritisation required to support newer services, such as the streaming of HD video. It is also worth highlighting recent developments in the mobile market, where the adoption of tiered pricing plans has been near-uniform despite the temptation that would have presented itself to some networks to decline to follow, and instead to look to capture market share. The other vital requirement is that operators be transparent with their customers about their traffic management. In other words, customers must be made aware of the way in which their traffic may be shaped. Although this objective is simple enough to set down on paper, it could prove a real challenge to implement in practice after all, the topic is a complex one with many variables, and it would be difficult to convey to consumers the day-to-day implications of the differing prioritisation offerings. Nonetheless, according to EC Commissioner, Neelie Kroes, transparent traffic management is non-negotiable. This could become a thorny issue, but the generally well-considered tone of most European regulators on the whole net neutrality subject is surely a positive for telecoms investors. To quote Ofcom:
At the current time we do not think there is compelling evidence of anti-competitive behaviour. Therefore, we do not think that there is a strong rationale for preventing ex ante all forms of traffic management. Indeed, given the potential for network congestion, some forms of traffic management are likely to lead to consumer benefits.
21
abc
Rise of the CDN

Net neutrality is an important topic already, but is set to become still more so as over-the-top (OTT) services develop (where OTT denotes any thirdparty service delivered over the top of a broadband platform). Historically, the telecoms space has been dominated by vertical integration that is to say, those operating the infrastructure have typically also been those selling services utilising that infrastructure (ie voice, and so on). However, regulators have intervened in order to ensure that dominance upstream in the network does not bestow dominance downstream in the services. As the arguments above (hopefully) outline, it is perfectly feasible to introduce priority charging without doing so in a discriminatory fashion that advantages the incumbents own retail services. And, with access to superfast bandwidth consequently equally accessible to all, it becomes all the more important to ask the question whether or not those providing the network connectivity are necessarily best placed to deliver the services that utilise them. After all, to employ an analogy, most road construction companies do not manufacture motor vehicles, or run coach services. In a regulatory environment where operators are able to use techniques like bundling to leverage their network platform advantage (for instance, by offering packages combining telephony, broadband and IPTV), it makes sense to do so (see our thematic report, Quadrophonia, February 2006). But where this type of cross-leverage is inhibited, the best course of action becomes much less apparent. Naturally, there remain some persuasive arguments in favour of telecoms operators persisting with the bundling of products that are outside their traditional sphere of expertise, but
that may nevertheless help them sell connectivity. So we would continue to advocate operators pushing IPTV as a part of their service suite. Persuading customers to take broad bundles also has a proven track record in terms of lowering rates of churn. However, we would caution that it is generally (though with some notable exceptions) difficult to see how the typical telecoms operator at least in developed, western markets is really likely to add value in its capacity as, for example, a media player. Unless the operator is prepared to invest in content of its own (a prospect that would probably terrify the majority of investors), it would simply be reselling others product with likely thin margins. Hence, even if a telecoms company achieved reasonable market share and thereby captured incremental revenues, the additional profitability associated with this activity would likely be relatively minor.
Media content distribution through long-haul vs CDN
Media company using longhaul Media company using CDN
Regional server
Regional server
Source: HSBC
But another powerful justification for operators getting involved in this space is to stimulate the market in general, knowing that they will receive their reward, if not necessarily from selling the service itself, then through providing the underlying connectivity. And, whereas the gross margin involved in reselling others content is often negligible, the gross margin from providing
22
abc
superfast broadband connectivity ought to approach 100%. In other words, operators need not make a colossal success of their own retail IPTV efforts to nonetheless benefit from IPTV services. After all, whoever owns the content still needs a distribution mechanism to get it to the viewer and the most flexible and powerful of these is the telecoms (or cable) pipe. The quicker new applications dependent on superfast bandwidth develop, the quicker and more compelling end customers will find the NGA proposition. In other words, the best way for an operator to spur fibre adoption may be to ensure that a wide range of applications are available over its platform and are certainly not restricted to those provided via its retail arm. Hence, regulatory and commercial interests may converge with the practical business reality that telecoms operators are not identified by consumers as the most natural providers of services beyond telecoms. And there are signs that some operators are beginning to acknowledge this, for instance through their investment in content delivery networks (CDNs). Capex in this field is effectively providing third-party providers of OTT services with the infrastructure they require to ply their trade. Hence we might say that CDNs deployed by telecoms incumbents are, in effect, the infrastructure and connectivity face of OTT services.
thereby saving the content owner on traffic carriage fees. So telecoms operators could be forgiven for regarding CDNs with some suspicion, since they effectively undermine certain revenue streams. But in practice, operators have often been keen to have CDNs attached to their network because the resulting additional traffic enables them to negotiate better interconnection terms with rival telecoms networks. Moreover, if time-sensitive services like video streaming are to achieve the high degree of quality of service required before they are to take off, it is clearly advantageous that the content be as close as reasonably possible to the consumer (as a shorter distance, requiring fewer hops between servers, naturally facilitates quicker delivery and means there is less to go wrong). The CDN arena has been dominated by specialists in this field, often deploying their own proprietary technologies so as to optimise their service. The largest and best known of these companies is Akamai, although it has numerous smaller competitors. These ranks are now being joined by the telecoms operators themselves.
The challenge
The CDN represents an interesting revenue opportunity in its own right so far as the telecoms operators are concerned. But this field is not without its challenges: not only does it already contain a set of entrenched players, but many of the potential clients could be OTT competitors of the incumbents. Nevertheless, CDNs are the type of connectivity service in which telecoms operators ought to have some natural advantages. One notable aspect of this market is that, because it has evolved on an ad hoc basis, there has been little by way of standardisation quite a contrast from the telecoms world, where everything depends on the network effect. Therefore, one
What is a CDN?
The principle behind a CDN is simple enough. Instead of, say, a media company streaming its video content from a single bank of servers in one location to customers wherever they are in the world, the idea is to distribute the content to regional servers, closer to the end user. This obviously significantly cuts down on the quantity of long-haul international data traffic generated,
23
abc
opportunity open to the telecoms operators is effectively to leverage the benefits of standardisation by bringing this to the CDN industry, perhaps by forming an international alliance (although such arrangements admittedly have a poor track record in telecoms). One of the benefits to the client of this approach would be that it could rely upon having its content distributed from multiple locations across the globe while only having to manage one, unified set of technology protocols (even if the various participants were to use different flavours of technology underneath). This would also address an obvious weakness of existing operator CDN capabilities: although incumbent networks have unparalleled depth in their home markets, they typically lack the global reach of the larger CDNs (meaning that content owners could be put to the inconvenience of securing deals with several different companies if they were to cover different regions). But can the telecoms operators persuade content owners to come to their CDNs? Some of the internet giants that have invested heavily in their own hosting facilities are in direct competition with the operators in many market segments; the likes of Google and Apple fall into this category. Another prominent class of CDN customer are those engaged in OTT IPTV services. To cite an example, the BBC uses Akamais platform to host its popular on-demand iPlayer service. Parties such as the BBC, that regard themselves essentially as content producers, and which have no ambitions in the area of connectivity, might well be prepared to consider a telecoms-run CDN, were the price/service level advantageous. However, other OTT players will have conflicts of interest: for instance, BSkyB competes with BT in both television services and ADSL. This rivalry would likely make it difficult for BSkyB to rely upon BTs CDN service.
Nonetheless, it is interesting that BTs Wholesale division has recently established a UK CDN capability (known as Content Connect) with the specific objective of hosting video content suggesting that it sees OTT IPTV of one sort or another as an appealing market segment. Doubtless BT Retail will become a customer, given the latters ambitions for its Vision payTV product. And Orange assuming it moves into the IPTV arena would also seem a likely customer, given that it has already decided to take a BT wholesale product for its retail broadband offering. The other potential differentiation that operator CDNs might offer relates to their ability to leverage their control of the underlying network to provide quality of service (QoS) functionality and guarantees. It is fair to say that content providers at present remain unconvinced about the merits of this proposition. This may largely be as a result of uncertainties over the business model, and in particular the controversial question of which party should pay the premium that a prioritised QoS streaming service would entail. But, at least in those markets where regulators adopt an enlightened approach to net neutrality (ie permit scarce network resource to be rationed on price as looks likely in Europe), it would appear probable that telecoms CDNs could monetise this capability. This should particularly be the case as video becomes ever higher definition, so imposing a greater burden in terms of the sheer volume of data packets requiring prompt delivery. Furthermore, it may become harder for existing CDNs to continue to provide their present level of quality of service. In the past, CDNs have often been able to install their servers at key points on the telecoms operators networks. The operators have consented to this because having the data cached locally cuts down on the cost of transiting it in from elsewhere. But the telecoms players
24
abc
might now start charging properly for this privilege; if this leads to third party CDNs hosting their content at more distant locations, it could impair the QoS provided. In turn, this would increase the appeal of operator-controlled CDNs, given their ability to provide QoS guarantees. We therefore conclude that CDNs should provide some potential upside for the telecoms operators. Based on the companys own estimates, Akamai delivers around 15-30% of all web traffic. In 2011, Bloomberg consensus anticipates revenues of USD1.2bn and EBITDA of USD544m; as of 11 February, the company had a market capitalisation of USD7.5bn. If we make the simplistic assumption that all web traffic is handled by CDNs (which is not as unrealistic as it might sound given the particularly heavy bandwidth that video consumes), then we can gross up to an estimated market size of about USD4-8bn. Of course, given the likely increasing importance of this service, these numbers are doubtless set to grow strongly. Nonetheless, the size of the market while material is still relatively modest by comparison with that for connectivity in the access layer, which NGA upgrades address. Hence the importance of CDN for telecoms incumbents is arguably as much in their recognition of the need for CDNs to drive OTT services in order to spur NGA penetration rates as in their intrinsic profitability.
systems? Recently, a set of upgrades to copperbased DSL technologies have led to the suggestion that fibre investment may be premature. Were this the case, it would obviously call into question the presence of scarcity in fixedline telecoms, and thus the sustainability of any pricing power that the incumbents might have begun to enjoy here.
Phantom of the copper-a

Although fibre clearly represents the future of telecoms access networks, as an infrastructure it is plainly both expensive and time consuming to deploy. (Indeed, this is one of the reasons we like the technology: since it is intrinsically expensive even to unbundle, let alone to replicate and the capital involved should therefore also elicit more supportive regulation). However, the above said, a number of vendors notably including Alcatel Lucent, ECI Telecom and Ericsson are actively pushing two innovations that could potentially squeeze more life out of copper-based DSL systems: namely vectoring and what is (rather cryptically) termed phantom mode. Alcatel Lucent announced in 2010 the results of a laboratory test in which two bonded VDSL2 copper pairs delivered 300Mbps over 400 metres and 100Mbps over 1km. By comparison, a single VDSL2 copper pair line would have a maximum bandwidth of around 100Mbps. These enhanced speeds were, in part, the result of implementing dynamic spectrum management level 3 (DSM L3) commonly known as vectoring. This exploits similar techniques to those used in noise cancelling headphones to mitigate far-end cross-talk (FEXT), which acts as a substantial impairment to the bandwidth and range that can be provided over bonded DSL pairs.
Fixed-line technology
The view that a degree of pricing power should return to the fixed-line telecoms subsector is clearly predicated on the idea that an element of scarcity is returning to the sector, given the tremendous expense of the fibre upgrades that are now underway. What, though, if alternative technologies emerge that would permit operators to upgrade to superfast broadband speeds without the capital commitment demanded by fibre-based
25
abc
A further boost to the available bandwidth is generated through the creation of a third, virtual pair. DSL pairs consist of a ground wire and a signal wire. When two pairs are bonded together, one of the ground wires is effectively surplus, and can be converted into a signal wire, a technique known as phantom mode. Again, vectoring is applied to reduce the negative effects of cross-talk from the phantom mode circuit. Bonded together, these techniques together can in the laboratory at least create a 300Mbps pipe. Naturally, these speeds sound exciting. Moreover, superficially, these technologies might seem to be a real positive for the industry, since they could materially reduce its capex requirements. However, for the reasons outlined above, it turns out that the amount of capital required for fibre upgrades (affordable to incumbents but inaccessible to most altnets) is in fact an important positive for the sector. But, leaving such points aside, the new technologies are in fact less practical than they might at first appear. Firstly Alcatel Lucents test relied upon high-quality 0.6mm-gauge copper. While this might be standard issue for new copper pairs installed in the last twenty years or so, the vast majority of the worlds installed copper plant consists of poorer quality, thinner, less conductive 0.4mm or 0.5mm lines.
Secondly, to take advantage of the new technologies, a home requires not one but two copper pairs. This is unusual in most markets: lines were doubled up in the narrowband internet era, when households would have a second line dedicated to their internet connection, but this was rendered redundant by the shift to broadband. It is also worth noting two additional handicaps. Firstly, new customer premises equipment (CPE) is required, as well as upgraded line cards (incorporating dedicated chips capable of coping with the greater signal processing demands of realtime noise cancellation). Secondly, in order to derive the full benefit from vectoring, an operator really needs to control all of the pairs connected to the DSLAM; clearly, this is typically not the case in markets where local loop unbundling has gained traction. Hence, our conclusion remains that fibrebased systems (whether FTTN/VDLS or FTTP) are required to deliver superfast broadband services, and given the capital that these roll-outs require, we continue to be of the view that they will reassert the importance of capital and scale within the industry.
Techniques such as vectoring and phantom mode have produced greater speeds from bonded VDSL2 over copper in lab tests
Ground wire Ground wire
Vectoring mitigates FEXT between signal wires
Ground wire
Ground wire
Signal wire Phantom cir cuit
Signal wire
Signal wire
Source: HSBC
26
Vectoring mitigates FEXT from Phantom
Signal wire
Signal wire
abc
NGA roll-out status Country Belgacom Technology VDSL Coverage achieved by the incumbent 76% population coverage by end-Q3 10 Ramping up the coverage, aim to reach 4m HH by Dec 2010, adding 70k HH/week VDSL coverage 25% of German homes Incumbent's coverage target 80% by mid-2011 Cable network status Cable vs incumbent Broadband market shares 61% incumbent and altnets, 39% for Cable
TNET has upgraded 100% of Battle of equals its network with DOCSIS3.0 50% UK coverage , but100% VMED at an advantage of its network is DOCSIS3.0 in the medium term UnityMedia and KabelBW have upgraded 95% of footprint, Kabel Deutschland 30%; by March 2012 65% of all German homes will have DOCSIS3.0 Numericable covers 9m homes out of 26m, no clear indication of DOCSIS3.0 coverge, we assume c2.7m homes UPC (Liberty Global) is nearly fully DOCSIS3.0 upgraded and Ziggo is fully DOCSIS3.0 DOCSIS3.0 100% upgraded completed with 3.2m homes passed
British Telecom Deutsche Telekom
VDSL/FTTH VDSL/FTTH
10m (c40%) premises by 2012 (o/w 2.5m FTTH) Plans to increase VDSL coverage to 31% HH (c12.4m HH) by 2012, 10% FTTH coverage (c4m HH) by 2012
21% for Virgin Media Vs 28% for incumbent Cable at advantage even DT 45%, Cable 13%, rest altnets in 2012 1/3 of German homes covered by DOCSIS3.0 but no telco NGA infrastructure Cable at a disasdvantage (weak financials + FT Ilad and SFR inveting in FTTH) Battle of equals Cable is 6% of the BB market
France Telecom
FTTH
c650k HH passed by FTTH at Fibre in all 96 homeland and in end Q3 10 3 overseas administrative districts by 2015 KPN is rolling out FTTH through its JV Reggefiber and has already passed 658k homes. Has 80 % IPTV coverage with VDSL/ADSL+ 1m homes passed by Dec 2010 Targeting 1000-1300k homes passed by FTTH by 2012e and has ambition of 30-60% coverage in Netherlands Reach 1.6m households with FTTH
KPN
VDSL/FTTH
KPN 41%, all other cables 40% (as of Q4)
Portugal Telecom
FTTH
Swisscom Telecom Italia Telefonica
VDSL/FTTH VDSL VDSL
Telekom Austria
VDSL/ FTTEx/ FTTC/FTTH
DOCSIS available to 75% of home passed by Cablecom (Liberty Global) Upgrading Milan and to start Target to reach 10-12% HHs by No cable in Italy Rome, no significant progress 2012. And 50% HHs by 2018 yet. ONO covers 7m homes (44% Awating for regulatory clarity. Previously guiding to spend of total), out of which 4.7m No material progress as such EUR 1bn in the period 20082012. Update expected in April upgraded to DOCSIS3.0 (as on Nov 10) 2011 (Investors day) 42% of HHs have access to Giganet Access (combination of UPC Austria (LGI) has FTTEx, FTTC, FTTB, FTTH) for DOCSIS3.0 covering 40% of Giganet network through comination of FTTEx (38%), 50% of Austrian Households by Austrian homes 2011 FTTB,FTTC & FTTH (4%) Fiber to 430k households (10% of total households) Fiber to 50% of households by 2014 ComHam covers 40% of Swedish homes and has mostly upgraded to DOCSIS3.0
78% VDSL coverage
FTTH coverage of 1m (33%) households by 2015
PT: 46% ZON: 33% Battle of equals (ZON covers more households (as of Q3 10) with DOCSIS3.0, but PT's FTTH is superior to DOCSIS3.0) Battle of equals Cablecom 18% vs incumbent 55% TI 55.7% (as of Q3 10) Battle of equals TEF: 54%, all cable 18.5% (Q2 10 CMT data) 30% TKA, 16% cable, 35% mobile bb only, rest altnets Telia 35%, ComHem 19%, Bredbandsbolaget 15%, rest altnets
TeliaSonera VDSL/FTTH
Battle of equals, TKA covers cable footprint with VDSL or FTTH; TKA only incumbent guiding for net line growth Battle of equals but with Telia clear commitment to invest in NGA
Source: Companies, HSBC
27
abc
Mobile connectivity
Pricing power evident in widespread shift to tiered data tariffs Certain operators now prepared to indicate capex is set to rise eMobile shows data-only services can generate healthy profits
Introduction
The idea that mobile data capacity is intrinsically scarce and that this will result in pricing power for the operators has proven highly contentious. For a variety of entirely understandable reasons, operators are nervous about broaching the subject. Even the suggestion that a companys network is under strain and might require remedial investment might scare off investors and customers alike. But, that said, there is now an increasing body of evidence in support of the arguments originally outlined in our thematic reports The Capacity Crunch (8 December, 2009) and Frequonomics (March, 2010). For those unfamiliar with the technological arguments underpinning our view on the intrinsic scarcity of mobile data capacity, we have summarised our thesis in the final part of the present section. A few operators such as AT&T and America Movil now have the confidence to guide for higher capex, and, in the meantime, there has been a profusion of networks introducing tiered data plans to ration their scarce bandwidth resource. We believe that tiered tariffs will enable operators to properly monetise the data opportunity. The potential profitability of this area remains subject to intense scepticism, although we would highlight the fact that the track record of eMobile in Japan a mobile operator that only provides data services is extremely encouraging (moving
from service launch to operating profit break-even in just under three years).
SuperFrequonomics
In our view, the mobile telecoms market stands at an important juncture. In the past, capacity has been relatively abundant, thanks in no small part to the fact that each new generation of technology has enabled operators to squeeze significantly greater volumes of traffic over a given portion of radio spectrum. In tandem with steady increases in the quantity of radio frequencies devoted to mobile services, and rapid growth in the number of base stations deployed by the operators, this has enabled the industry to move from niche status to supplying the communications needs of billions globally. However, existing technologies are already highly efficient, and, indeed, are approaching the physical constraint that is the Shannon limit. This caps the quantity of data that can be conveyed over a Hertz of radio frequency in a second. Given the interference levels that we can expect in most environments, the Shannon Limit stands at around 2 bits per second per Hertz. Current technology already manages around 1, implying that a doubling in efficiency is the theoretical maximum; in practice, the next generation of 4G/LTE equipment is likely to be around 30% more efficient than todays systems.
28
abc
Shannon Limit sets a ceiling on mobile spectrum efficiency
6
Shannon li mit
5
Achieva ble rate (bps/Hz)
Shannon li mit with 3dB offset OFDMA CDMA (HSPA)
4 3 2 1 0 -15 -10 -5 0
Required SN R (dB)
L ower interference e nvironmen ts e .g. isolate d hotspots, f emtocell s
Inaccessible Region
10
15
Typica l loaded mobile n etw ork (outdoors)
20
Source: Alcatel Lucent, HSBC
Of course, operators can also purchase more spectrum (although this can be very expensive) and build more base stations (also costly). We would expect them to do both, and with vigour. We also anticipate use of alternative strategies everything from MIMO (multiple input multiple output antennae) to base station caching to wireline offload (of which more later). Yet, despite all of this, we still think that capacity will be intrinsically scarce and hence require rationing by means of price. Given the unpopular and contentious nature of
Capex/sales for various operators with wireless operations
18% 16% 14% 12% 10% 8% 6% 4% 2% 0%
Vo dafo ne AM X AT&T Wireless Sprint Nextel Wireless Verizo n Wireless
capex in telecoms, we would expect operators to delay as long as possible before investing more. Certainly, they will need to demonstrate the attractiveness of the opportunity before they will be able to confidently justify the additional expenditure. Hence, the first evidence that we are on the right track with this thesis is unlikely to emerge in the capex line. Instead, we would argue that investors should look to operators approach to tariffs as a leading indicator. Because, by adopting the appropriate pricing strategy, and rationing their scarce capacity resource, operators should be able to manage their capex line while demonstrating to the market the appeal of data services. Once this has been accomplished, they can subsequently announce their intention to invest more without the same risk of pillory.
The AT&T narrative

AT&T is arguably the best example of a corporate establishing this narrative. On the companys Q3 2009 conference call in October, management were adamant that the network was up to the job, despite the numerous complaints from iPhone customers that the AT&T platform was
2009 Source: Company reports and HSBCe
2010
2011e
29
abc
struggling. Then, at the Q4 2009 results conference call in January 2010, the company unveiled a significant network investment programme. But with service revenue growth in the mobile division of nearly 10%, there were few complaints. And, in fact, capex growth in the wireless division for the full year 2010 eventually came in at around 50%, considerably higher than implied in the original guidance (a USD2bn or c30% increase) Investors, though, can plainly see that the capital is being invested in what is a very attractive business. AT&T was also among the first operators to shift to tiered pricing structures for its iPhone4 smartphones. Such plans have the advantage of linking consumption with revenues, encouraging customers to consume capacity responsibly (rather than utilising their connectivity for peer-to-peer activities, for example) and, very likely, to ultimately spend more to obtain additional bandwidth (in the same way that many customers have migrated over a period of time to larger minute plans commanding a higher ARPU).
contemplating the otherwise risky introduction of caps to its data plans can move with greater confidence because it understands that its rivals are in a very similar situation, facing similar challenges. In other words, any operator leading the way by introducing tiered pricing plans will likely find its chief competitors following, rather than attempting to seize market share by exploiting the lead operators introduction of a more complex and inflationary tariff. So it was perhaps not so surprising that, at its H1 2010/11 results presentation, Vodafone management indicated that in almost all of its European markets, its introduction of tiered data plans had been followed rather than undermined by its chief competitors. We would infer that the one exception that Vodafone alluded to was the Dutch market. And even here, according to KPNs Q4 2010 results presentation, the market has now moved in the direction of tiers. This is not to suggest that the pricing moves have been entirely in one direction. Nevertheless, the degree of uniformity witnessed is remarkable by telecoms standards. But what of the dissenting voices? As discussed in our SuperFrequonomics report (September 2010), some of these are relatively small operators that are perhaps not greatly significant in terms of the overall market (such as Virgin Mobile USA, part of Sprint Nextel). Others result from local market circumstances (for instance, South Korean operators have an unusual abundance of capacity due to years of government-sponsored, industrial policy-driven, over-investment in network). Still others have indicated that their current all-youcan-eat offers are unsustainable in the longer-term (for example, Telenor Sweden). However, one prominent European operator is keen on flat-rate data tariffs, namely Three. The view at Three is that the operator has sufficient capacity not to need to ration it tightly
Remarkable uniformity
Operators right across the developed world have followed suit in a remarkably short period of time (while most in the emerging markets had refrained from offering unlimited data plans in the first place). In a fractious, highly competitive market like mobile, where operators are quick to pounce on any opportunity to take market share, a shift displaying such uniformity demands a coherent explanation. The control experiment provided by the voice market (where unit prices continue to decline rapidly) indicates that, whatever is taking place, it is unique and specific to the data arena rather than being a function of the broader competitive environment, for instance. In our minds, the only plausible driver of this rapid transition is the emergence of capacity constraints among the operators. An operator
30
abc
by means of price. What is particularly interesting about this is that its network sharing partners in the UK (originally T-Mobile, but now also Orange, as a result of the ongoing merger between these two) seem to take a very different view, despite the fact that all effectively operate off the same infrastructure. So it is notable that, despite Everything Everywhere (the name of the combined Orange/T-Mobile UK entity) highlighting at its inaugural presentation its advantage in terms of numbers of base stations over its rivals O2 and Vodafone, it nonetheless has the tightest caps among its peers restricting the usage of its iPhone4 customers. Who is right? Is capacity abundant, implying that Three UK can afford to maintain its generous offers on mobile data? Or is Orange/T-Mobile UKs characterisation more accurate, implying that Three will in time be forced to more tightly limit the bandwidth its customers consume? At this stage it is too early to tell directly, although it must have been one fear at France Telecom and Deutsche Telekom that merging their UK mobile assets would effectively permit Three (which had established a prior network sharing arrangement with T-Mobile UK) to piggyback on the capacity advantage that the merger was in part designed to create. Everything Everywhere has therefore been at pains to emphasise that, if Three generates the requirement for additional network capacity, then it must bear the relevant costs. The real challenge for Three will be what an economist would term adverse selection. What type of customer would consider churning from their existing network to Three because the former introduced tiered pricing structures, thereby effectively capping usage? The answer, plainly, is that the customers most prone to jump ship are those who consume the greatest bandwidth. Given the capex required to support such customers, the
network losing them may consider itself to have enjoyed the better side of the bargain.
Network failure
Furthermore, recent events in Australia at another of Threes networks do raise questions about operators ability to meet the rapidly intensifying demands of todays mobile customers. Threes Australian business has merged with that of Vodafone, and competes against Telstra, the incumbent, and Optus, part of SingTel. In an effort to gain market share, Vodafone/Hutchison3 positioned itself with a set of generous plans encouraging heavy voice and data usage. Unfortunately, even the operators combined network does not seem to have been fully up to the challenge of meeting the resulting demand. Customer dissatisfaction has culminated in the appearance of a local website dedicated to charting the networks failings, and even to the potential launch of a class action lawsuit, with disgruntled customers looking to sue over the allegedly poor nature of the service. In response, the CEO has issued a public apology, and pointed to the companys AUD550m network upgrade programme. With respect to Vodafone specifically, it should be stressed at this point that the groups Australian activities are not a substantial proportion of the company (and thus do not derail our investment thesis); as well as that Vodafone has typically considerably outinvested its rivals, and so should actually enjoy a competitive advantage versus the competition in most of its markets of operation. Nevertheless, this example is useful in terms of highlighting the problems that can rapidly emerge when pricing plans encourage usage that the underlying network is unable to support. To summarise, in view of the type of experience seen in markets like Australia (as well as the previously mentioned example of Telenor Sweden), it remains our view that unlimited plans
31
abc
are counter productive. If network service quality is impaired, the resulting damage to an operators brand could prove enduring. The risks therefore involved with flat-rate pricing perhaps go a long way towards explaining its increasing rarity. Note, though, that while operators would (in our opinion) be well advised to avoid selling data tariffs on the basis of a flat rate, all-you-can-eat approach, nevertheless this concept remains highly attractive from a marketing perspective. Hence operators marketing may well continue to suggest that their plans provide a flat-rate approach, even while they are in actuality merely selling a finite chunk of data. So, for example, certain of E-Pluss data plans in Germany are described by the operator as flat-rate but in fact have caps varying between 50MB and 5GB.
option of upgrading to a bigger plan. Such solutions reinforce the inflationary potential of tiered data plans. Vodafone will extend the service to other Western European markets this quarter, starting with Germany and the Netherlands. A further sign of Vodafones determination to ration data capacity comes with the news that allowances in Italy are being reduced from 2GB/month to 1GB/month, although this move has so far not been followed by the main competitor, Telecom Italia.
Everything Everywhere (UK)
Recent pricing trends positive

As discussed above, starting from Q2 2010, tiered data pricing has been spreading rapidly across the industry. In the months following the publication of SuperFrequonomics report (September 2010), there have been many further announcements confirming this shift as a consolidated trend. A selection of the most interesting examples are detailed below.
Vodafone (Europe)
The new joint venture between Orange and TMobile in the UK unveiled its strategy in September last year. The new entity will have the densest network in the country and enjoy the largest spectrum allocation. Despite this, it has decided to opt for tiered data plans: it is currently offering data in similarly sized blocks as Vodafone and O2 (500MB or 1GB per month). Hence, all large operators in the UK have now introduced what appear to be sensible usage caps. However, Hutchison 3G has instead decided to head in the opposite direction and revert to a flatrate all-you-can-eat approach for the 16GB version of the iPhone4 (for the 32GB model 500MB and 1GB per month plans are offered in addition to all-you-can-eat). This is likely to encourage very heavy users (eg those partial to peer-to-peer download activity) to churn towards Three. Arguably, these are the least desirable type of customers: heavy users with no willingness to pay. It is also worth introducing our usual market share analysis at this point. We would describe Three as an infant player, as it has not yet achieved a percentage market share in the teens (so as to qualify as a teenage operator). Just like their teenage rivals, infant networks certainly have the
Vodafone announced in its fiscal Q3 2010/11 (calendar Q4 2010) results release that the company had completed the migration to tiered data plans in eight western European markets. The remainder will follow in Q4 2010/11, so that by the end of March 2011 all the European operations should have moved to tiered data tariffs. Taking tiering one step further is the practice of smart notification, which has been tested with success in the UK over the last quarter. As users approach their allocations, they are contacted and offered either a block of additional data or the
32
abc
incentive to disrupt the market; however, unlike teenage operators, they typically lack sufficient weight to really rock the boat. When teenagers introduce aggressive price plans, the adults (companies with 20%+ market share) generally must counter; but infants generally the lack the brand, network and high-street penetration to force a response.
KPN (Netherlands)
Management has now delivered on the undertaking made at the Q2 2010 results, namely that KPN would migrate to tiered data plans in the Netherlands by the end of 2010. All iPhone plans come with a data bundle of either 500MB or 1GB per month. The 1GB per month data plans cost between EUR45 and EUR110, depending on the chosen allocation of voice minutes (from 250 to 1,000 per month) and texts (500 to 2,000 per month). Meanwhile, KPNs German operation E-Plus announced in October a new range of smartphones to be sold in conjunction with a flatfee data package. Though, at first sight this would appear to be the very opposite of a tiered tariff, the plans in fact seem sensibly designed. Initial caps are as low as 200MB, and beyond this point customers then have the option of either accepting a reduced speed or topping up. It is particularly encouraging to witness a teenage operator like E-Plus displaying such a rational stance with tiered pricing structures. Firstly as a teenager E-Plus has real scope to
damage the overall German market. If it were to introduce aggressive pricing arrangements even if these were of the all-you-can-eat variety - then the two leading operators (T-Mobile and Vodafone) might be forced to follow suit. It has certainly been the case in the past that E-Plus has adopted aggressive pricing policies in order to take share, and this has been one of the major drivers behind the heavy price pressure seen in Germany over the past half decade. It is also worth mentioning that O2, the other teenager in the market, has also been circumspect with its data tariffs. Its iPhone plan has a monthly data usage cap of 300MB per month.
Sweden
Sweden is currently one of the most attractive mobile markets in Europe, with all the major operators here reporting high single-digit service revenue growth in Q4 2010. On its results call, Tele2 management expressed the view that the market remains only at the bottom of the data adoption curve. Tele2 was sufficiently confident about the strength of mobile data demand to be able to guide for high single-digit service revenue growth to continue in 2011. In our view, the key drivers here are rational pricing in data, together with relatively benign price competition in voice (and note that MTRs are already relatively low in Sweden). With the exception of Telenor, all the major operators have introduced tiered data pricing. TeliaSoneras offering starts with a 2GB block of capacity for SEK99 (EUR12.7) per month; while its largest
KPN Base Germany: New flat rate data tariffs promotions are actually capped and throttled Internet Flat S Offer for Internet Flat Internet Flat L Internet Flat XL
Data Volume speed throttled to GPRS (56kbps) at Monthly fee (EUR)

Source: KPN Base Germany
...people who only people who access the smartphone users people who need mobile occasionally browse the web internet with a smartphone regularly access the internet internet access for their laptop unlimited unlimited unlimited unlimited 50MB 200MB 1GB 5GB 5.00 10.00 15.00 20.00
33
abc
package is a 4G offer of 30GB for SEK599 (EUR77) per month. Currently TeliaSonera imposes speed restrictions of 120kbps once a user exhausts their data cap. Telenor, on the other hand, has continued to sell unlimited data plans, although it does restrain consumption by capping the speed it provides once users exceed their allotted usage. Its cheapest plan is priced at SEK199 (EUR26) per month, while its high-end offering weighs in at a hefty SEK549 (EUR71). The relatively expensive nature of these plans, in combination with the use of speed throttling, suggest that the risk to Telenor Sweden from selling flat-rate plans is, at the moment, modest. Nonetheless, it was interesting on the companys Q4 2010 results conference call to hear management express the view that unlimited plans are not sustainable in the long term.
Telecom Italia
data usage. So while TI has not yet followed Vodafone Italys move to reduce the cap on its data tariffs from 2GB to 1GB per month, we would not be surprised if an announcement in this direction was made in due course.
SKT and KT (South Korea)
The South Korean mobile market was showing real promise last year, with KTs introduction of tiered data pricing plans. However, in July 2010, SK Telecom (SKT) announced the introduction of flat-rate data tariffs for customers spending in excess of KRW55,000 (cUSD50) per month, as well as free VoIP minutes over its 3G network presumably in a bid to take back market share. Initially, KT insisted that it would not retreat, but subsequently, in September, the company capitulated, indicating that it would match SKTs flat-rate offers. While SKTs move and KTs decision to follow were disappointing, there are some important mitigating factors to consider. First, note that both operators flat-rate offerings come with constraints: usage is restricted in peak hours, as well as on a dynamic basis in busy cell sites thus limiting the potential for debilitating congestion. Second, although this is hardly the popular perception of the market in Korea, smartphone penetration is currently relatively low, at around 15% of total accounts as at the end of 2010. But, by the end of 2011, we project this figure could reach as high as 45%. Hence the operators probably will be able in the very short-term to absorb the additional volumes that their new tariff plans will doubtless stimulate. The acid test will only come as usage grows: will the Korean operators then be able to retain their present stance? This pressure is likely to fall disproportionately on SKT, as it lacks KTs extensive WiFi presence (and thus the ability to offload traffic onto the fixed-line network). As at September 2010, an impressive 67% of KTs total
With regards to data pricing, Telecom Italia (TI) remains an exception among the large European operators, in that it continues to offer what are essentially flat-rate plans (albeit with a fair-usage cap of 2GB per month). This state of affairs is probably explained by the very specific situation in which TI finds itself. Following an illconceived price increase introduced in September 2009, the domestic mobile operation has been underperforming its peers. In addition, its 2010 tariff rebalancing in favour of flat-rate plans (in voice and data) has caused both revenue cannibalisation and ARPU dilution. We expect the mobile business will continue to show significant weakness in Q4 2010 (results to be reported on 24 February 2011), and most likely also throughout H1 2011. However, we do not think that Telecom Italia is in a different position to its European peers with respect to the intrinsic scarcity of mobile capacity, and the pressure that will result from increasing
34
abc
mobile data traffic (2,500TB per month) was being offloaded onto WiFi. Thirdly, it is important to recognise that the price points associated with these tariffs represent a clear step-up as compared to overall ARPU levels. For example, SKTs All-in-One flat-rate tariffs begin at KRW55,000, a considerable premium over its network ARPU of KRW36,000. Arguably, the situation is analogous to the Japanese mobile market back in 2003, when KDDI first introduced its flat-rate data plans, and initially achieved very strong growth. The risk is really that the Korean operators like their Japanese counterparts subsequently find themselves unable to move away from flat-rate pricing, because of the sheer number of customers for whom this has become the norm. However, this does not constitute a near-term risk.
Verizon Wireless (US)
of the then current and immediately following billing cycle. Verizon Wireless has further clarified that data throttling will be applied in those areas where a users data consumption is impacting surrounding users. This clearly indicates that the company is very much mindful of capacity and quality of service issues, and so we suspect that its approach is ultimately unlikely to be very different from that of AT&T. We see its initial tariffs as a tactical move to maximise the appeal of its variant of the iPhone and poach customers on the AT&T network who are frustrated with the quality of its network. In due course, we would anticipate Verizon Wireless transitioning to tiered plans. Smaller operators Sprint Nextel and T-Mobile USA for now are also sticking with unlimited plans. However, Sprint commented on its Q3 2010 results conference call that it was continuing to evaluate tiered pricing plans.
Latin America
AT&Ts original introduction of tiered data plans has arguably been the highest profile to date. This move was necessitated by the immense loads that the iPhone (which was exclusive to AT&T until very recently) imposed on the companys network. However, the biggest new development in the US is very clearly that Verizon Wireless has now finally also been able to launch a version of the iPhone compatible with its 3G platform. At least in the initial stages, Verizon Wireless is not being aggressive as AT&T in charging its customers for data usage on a tiered basis. The minimum requirement for its iPhone customers is that they should take up an unlimited data package of USD29.99 or higher. However, in order to combat the potential for network congestion, Verizon Wireless has stated that it may throttle data rates (ie slow the bandwidth) to those customers falling within the top 5% of users. For such customers, Verizon Wireless has further warned that it may reduce data throughput speeds periodically for the remainder
The situation in Latin America is rather different from that in the US, given that most operators here never sold unlimited data packages in the first place. Admittedly, the Brazilian mobile networks did initially offer unlimited access to email and Facebook-type applications, but distinguished such usage from that of more bandwith intensive iPhone customers, who faced fair usage caps from the outset. But, by 2010, the operators had switched to some form of limited tiered price plan in conjunction with most handset models. Vivo and Claro, which cater to higher-end consumers and have a better mix of contract subscribers (c20%), have largely withdrawn unlimited data plans altogether. That said, Claro has been offering an unlimited data plan with the iPhone, provided the customer pays the full price (cUSD500) for the 16GB iPhone4. Meanwhile,
35
abc
value player Oi offers tiered pricing, with up to 10GB per month of usage. TIM, which caters to the large C-class of middle income users (roughly 60% of the population) has taken a different approach to pricing, by charging on the basis of time rather than on data volumes. It has also introduced an unlimited offering for pre-pay customers, but with speeds that are reduced once a daily threshold has been reached. In Mexico, America Movils Telcel unit (the clear market leader) offers a so-called unlimited mobile data plan, but with speeds that are throttled to a maximum of 64kbps for the remainder of the month once a user has reached a 500MB fair use threshold. Telcel can afford to be tougher on pricing than most operators in the Americas given its dominant market share, and the fact that it is the only sizeable operator with a 3G service in the key region of Mexico City.
The company guided for revenue growth of between 1% and 4% (as compared to a consensus of around 1%), as well as for stabilising margins. However, it did not lift free cash flow guidance on the basis that it felt it would be advantageous to invest more in the business. Clearly, additional subsidies may be required as ever more customers choose smartphones; but this type of expenditure is registered above EBITDA. The additional investment mentioned therefore implies higher capex. Management remains adamant that the Vodafone network in Europe does not face capacity constraints; nonetheless it does now seem sufficiently confident in the prospects for data services to implicitly guide to modestly rising levels of capex. Again, a crucial point to make here is that Vodafone has historically out-invested most of its mobile rivals. In other words, a need (or desire) to invest more in capex does not indicate (in Europe at least) that its network is somehow deficient. In fact, Vodafone cut its capex during the credit crunch rather less aggressively than most, and so has been outspending its competitors even recently. Vodafones guidance does not therefore represent catch-up spending. We suspect that if a consistent, relatively heavy investor like Vodafone is contemplating rising capex, some of its rivals may have to up the pace of their spending by rather more. We would also point to the track record of AT&T, which initially guided to a rather modest rise in capex, but in the end increased its expenditure by considerably more. While AT&Ts network issues are somewhat unusual (relating in part to its somewhat tortuous technology upgrade path), we would nonetheless argue that there remains scope for operators to revise upwards their guidance for capex, as the extent of the opportunity in mobile data becomes more apparent.
Capacity crunch
We believe that the transformation in data tariff structure is the most conspicuous smoking gun indicating the veracity of our thesis that mobile data capacity is scarce. The introduction of tiered pricing is also clearly helpful in providing evidence that operators should be able to charge more for the extra capacity that they must supply if they are to meet demand. However, there are also indications from other quarters that mobile data network capacity is showing signs of being scarce and even the admission from certain telecoms companies that mobile capex is likely to be on the rise, albeit in a controlled fashion. Given its pan-regional presence, Vodafone is often regarded as a useful proxy for the European mobile sector as a whole. It was therefore particularly interesting to hear managements implicit guidance at its H1 2010/11 results that capex was likely to rise somewhat.
36
abc
There are also more subtle indicators that mobile data services will be subject to capacity constraints. An interesting and surprising example is in the dtente reached between two long-term adversaries, Google and Verizon. The former is, needless to say, the loudest of activists on the subject of net neutrality; while the latter has lobbied hard against such regulatory intrusions. But, in the aftermath of the Comcast court victory over the FCC on net neutrality, there have been efforts by industry participants to broker a solution that would avoid the FCC taking a more active role in internet regulation. One result of this was Google and Verizon reaching a compromise, in which Verizon agreed that net neutrality was a viable proposition on its fixed-line network (which, in Verizons FiOS regions, is all-fibre, and hence has enormous capacity), but in which Google conceded that net neutrality should not be applied to the wireless industry. Implicit in this agreement is the idea that fixed-line capacity is limitless, but that mobile capacity is intrinsically limited, and thus does require rationing. Further to this, the FCC in December 2010 adopted a modified net neutrality policy along similar lines: applying strict net neutrality principles to fixed-line networks but conceding that due to the inherent scarcity of wireless capacity, that it was only practical to allow operators to perform legitimate traffic management, provided that the reasons for doing so were transparent.
cellular network. Ostensibly, the idea here is to build the O2 brand and win new customers, attracted by the operators innovation and generosity. The WiFi coverage should also be useful in going after the fast-developing mobile advertising market: for example, customers could receive an advert for a shop conveyed by a WiFi hotspot located there. Given WiFis very limited range, the advert would reach only customers who found themselves in close proximity, and could therefore pop in to the premises in question. But we suspect that there is more to the move than the pursuit of potential new revenue streams. O2 UK has enjoyed good success with its (initial) iPhone exclusivity, but this has resulted in some well-publicised strains to its network. Although remedial capex has been invested in particularly congested areas like London (O2s Network Performance Improvement Plan earmarked an additional GBP100m in 2010 primarily focused on adding 3G cell sites in major towns and cities), it would hardly be a surprise given the ongoing explosion in data traffic if the network was still feeling under strain. One way to alleviate this would be to use wireline offload, in which traffic that would otherwise hit the (contended) cellular base station is instead routed via WiFi (or, for that matter, a femtocell) to the fixed-line network (where capacity is far greater). Arguably there is a useful parallel here with Softbank in Japan which, just like O2, has had both iPhone exclusivity and its fair share of network problems. In an effort to alleviate the burden being placed upon its cellular network, Softbank has actually been giving out WiFi units free though note it is also engaging in an ambitious capex upgrade programme at the same time. In our view, Softbanks twin-pronged approach really underlines the severity of the capacity issues facing the industry: management
WiFi provides further evidence

Another interesting recent development has been the announcement by O2 UK that it is to move more aggressively into the WiFi hotspot market. With modest additional investment, the operator is seeking to deploy its hotspots in areas like retail outlets and hotels, but will then offer the resulting connectivity to everyone, not just users of its own
37
abc
guidance for 2010 suggested an increase in capex of over 80%, and yet the company still felt the need to embark on a parallel and very ambitious WiFi hotspot strategy. Naturally, the approach being taken by both O2 UK and Softbank also raises questions about the respective roles of WiFi and cellular, and whether there is a risk that the former cannibalises the latter. There are certainly some vocal bears of the mobile space who would argue that WiFi amounts to a powerful competitive challenge to cellular systems. But, were this really the case, it is somewhat difficult to explain why there should be two cellular networks among those that are leading the WiFi charge. In fact, both operators would seem to feel that there is more than enough capacity demand to justify the deployment of both types of network. We covered this topic in some detail in our thematic report The Cell Side last year (April, 2010). A key conclusion was that WiFi infrastructure is simply not scalable: in other words, it is fundamentally incapable of providing any kind of ubiquitous service. As a result, WiFi will remain a nice-to-have rather than a necessity. Customers valuing ubiquitous service (which has been a common feature of all successful mobile services) will continue to need to take a service from a cellular operator. See illustration at bottom of page. As such, it is the
WiFi: although handover has improved the barrier remains a practical one: patchy coverage
mobile operator that owns the intrinsically scarce element in the value chain, and thus stands the best chance of being able to charge for it. Moreover, if mobile data growth comes anywhere near forecasts suggesting it will double every year for a decade (implying something like a thousandfold increase over the course of a decade), then networks will need to perform all the wireline offload of which they are capable (and more tricks besides) in order to cope. So, far from signalling that mobile is set to be displaced, we see moves into WiFi from the likes of O2 UK and Softbank as providing further corroboration for our view that cellular capacity is intrinsically scarce, implying rising capex but also price-based rationing or, to put it another way, pricing power. Note, though, that O2 UK has not been the only British company to recently push into the WiFi space. BSkyB (hitherto primarily a payTV operator with an extensive presence in the unbundled ADSL market) has announced its purchase of The Cloud, a WiFi hotspot startup. This move implies that BSkyB are concerned about the mobile operators ability to deliver Sky Players streamed video content to its customers. The deal might also indicate that BSkyB is concerned about the type of charging structure that the mobile operators could choose to impose
Cellular: hierarchical cell structures support cost-effective ubiquitous service
No service
WiFi
G PRS/EDGE
WCDMA
HSPA
Source: HSBC
Source: HSBC
38
abc
in order to provide the necessary prioritisation to ensure adequate quality of service for Skys video streams. However, for the reasons touched on above, even an acquisition such as this barely scratches the surface in terms of providing coverage. BSkyBs customers will still need to rely on cellular network transmission if they are to be able to access their video content in the majority of locations although perhaps BSkyBs latest acquisition might plausibly improve its bargaining position. Note that BSkyB has shown an excellent track record in terms of being prepared to invest in order to head off future potential challenges. For instance, well in advance of BT being able to bundle IPTV content with its offerings, BSkyB had entered the broadband market by unbundling ADSL through its Easynet acquisition. Such a move need not indicate long-term commitment to the network market in question after all, with the transition to NGA, BSkyB may well return to wholesaling connectivity from BT, rather than unbundling the service (although it is participating in a small fibre trial).
capital intensity, there remains the nagging suspicion in the market that data services are simply intrinsically unprofitable by comparison with their voice antecedents. Indeed, it is frequently said that data is intrinsically dilutive to returns perhaps one reason why the prospect of higher capex in order to deliver it has cast such an oppressive shadow over the sector. Judged from first principles, though, there is no reason to suppose this must intrinsically be the case. Consider the question through the lens of the gross margin, for instance. Voice calls often terminate off-network, necessitating interconnect payments that remain substantial (despite regulators ongoing efforts). Hence, the gross margin on voice services must reflect this resulting in margins often around the 65% level. By contrast, with respect to data services, interconnect is not a factor hence gross margins are closer to 100%. But perhaps the most compelling argument that mobile data can generate attractive profitability is of the simple empirical variety. What is needed is a mobile player that is devoted exclusively to providing data services. Fortunately, such an operator does exist, and provides an enticing insight into the profitability of data, even at a modest (but not irrelevant) level of market share. The network in question is eMobile in Japan (the mobile subsidiary of eAccess): and what should be truly impressive is that, with market share standing at a modest 2.4% as at end-2010, we nonetheless anticipate a full year EBITDA margin of 27% for the mobile division (for the year to March 2011).
SuperEconomics
The commercial appeal of data services remains subject to widespread scepticism. One part of the anxiety on this topic relates to the capex that is required in support. We have argued that levels of capital intensity are likely to rise, and return to their (pre-credit crunch) longer-term norms (in Europe, around 12-13% of sales seems plausible). Unrestrained, the tremendous growth in demand for data bandwidth would, in our view, far outstrip the ability of operators to keep pace (even allowing for innovations like fixed-line offload and MIMO (multiple-input multiple-output) antennas. However, we think that they will be able to keep the growth to within bounds that they are capable of satisfying through the use of tiered plans. But even leaving to one side the question of
The eMobile business

eMobile is a wireless broadband operation in Japan, a country where wireless competition is intense, while fixed-line broadband services are the fastest and cheapest in the world (according to the OECD). Yet, in an environment that looks
39
abc
exceptionally tough (at least on paper), eMobile has flourished. The company has had a number of factors in its favour; some are market specific, but others are more broadly applicable: Competition specifically for mobile data is actually limited. Direct wireless broadband competitors (Willcom, UQ Communications) are currently ineffective competitors because of financial and operating difficulties, and crucially the established cellular operators (NTT DoCoMo, SoftBank, KDDI) are capacity constrained. New network equipment. eMobile launched services in March 2007, benefiting from a Huawei/Ericsson 3G network of an entirely different order to that deployed by NTT DoCoMo in its ground-breaking launch of non-standard WCDMA back in 2001. Business model designed for wireless broadband. eMobile intended from the outset to focus exclusively on wireless broadband. Strong demand from younger users. Despite Japanese wireline broadband being both cheap and fast, eMobile services have tapped a need among younger Japanese users (many of whom live with their parents due to high
rental/housing costs) for a private/personal broadband service. eMobile launched services in March 2007, using equipment from Huawei and Ericsson. From the outset it has seen solid demand for its wireless services (which were delivered almost entirely via datacards for the first three years after launch). Key milestones have included: eMobile reached 1.6m subscribers (market share of 1.5%) and break-even on an EBITDA basis in the quarter ending June 2009. In the quarter ending December 2009, eMobile achieved break-even on an operating profit basis, and moved to operating free cashflow break-even in the quarter ending June 2010. By end-2010, eMobile had 2.924m wireless data subscribers, equivalent to market share of 2.4%. We forecast its EBITDA margin in the year to March 2011 at 27%. eMobile CEO Eric Gan has controversially observed that the job of eMobile was merely to provide access to be the bit-pipe guys (in an interview in Ericsson magazine, December 2008).
eMobile EBITDA (JPYbn) and EBITDA margin trends
15 10 5 0 -5 -10 -15 EBITDA

Source: Company data, HSBC
30% 25% 20% 15% Mar-08 Jun-08 Sep-08 Dec-08 Mar-09 Jun-09 Sep-09 Dec-09 Mar-10 Jun-10 Sep-10 10% 5% 0% Margin, %
40
abc
Given mobile operators largely ineffectual attempts to prevent the surrender of value in the wireless application layer to companies such as Apple, his focus on building a profitable business based on data transport now seems prescient. eMobiles pricing mechanisms have been both sensible and consistent, in our view: Data pricing has been tiered, with recent declines in ARPU due to the increased takeup of plans with a lower minimum monthly fee (but with more expensive overage charges). Netbook subsidies are repaid over a two-year contract. This is as compared to operators in Taiwan, which have run several heavily subsidised, value-destructive offers on netbook and wireless broadband combinations. eMobile customers are required to pay an additional amount over the monthly fee on a 24-month contract that cumulatively equates to the value of the netbook. Unlike operators seeing substantially higher marketing costs for smartphones (such as in Taiwan and Korea), eMobile has seen average subscriber acquisition costs (SACs) fall due to the sharp declines in the cost of procuring wireless broadband modems. Previous company guidance had been for SACs of JPY30,000, to be split equally between device subsidy, commission and advertising. Recent price declines should see the subsidy portion decline substantially we project SACs falling to JPY24,000 in the year to March 2013.
Obviously, with respect to SACs, eMobiles experience may prove less relevant to more conventional operators that will be subsidising smartphones rather than datacards. Nonetheless, the operations profitability (whether considering its market share or its time since launch) is impressive given mobile datas reputation for dilution.
The power of pricing power

The example of eMobile, then, augurs well for the mobile industrys profitability as data services grow in significance. The outlook should be even better if operators are able to enjoy a degree of pricing power in this area. Arguably, eMobile represents an early instance of this phenomenon as well. It owes part of its success, in terms of its degree of pricing power, to the capacity constraints of its chief rivals for example, DoCoMo (which has been saddled with a prestandardised WCDMA platform that has hampered its upgrade path) and Softbank (the network of which is struggling to cope with the volumes generated as a result of the companys iPhone exclusivity).
41
abc
Whatever its precise source, though, there is little to rival the scale of the impact that the arrival of pricing power can have on an industry. In order to drive home this point, it may be useful to consider a handful of examples. The purpose of this exercise is not to inspect the mechanisms conferring pricing power (which are different in each instance even within the telecoms sector, between fixed-line and mobile) so much as to examine the extent of its impact most particularly in terms of valuation. Perhaps the best known example of the sudden appearance of pricing power originates from the oil sector. Lately, OPECs ability to control the oil price has been somewhat diminished as a result of the development of new reserves in Russia, Alaska and the Gulf of Mexico. But its members still own around 79% of the planets crude oil reserves and account for 44% of current production. It might be supposed that higher oil prices would translate into greater profitability for the western oil companies, and thus also, more generous valuations. However, there are a number of other factors in the mix for instance, the heavy taxation that oil tends to attract. Recently, strong demand from emerging markets like China and political tensions in the Middle East have contributed to a significant appreciation in the
OPEC had a dramatic impact on the oil price in the 1970s
120 100 80 60 40 20 0 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009 Crude O il (USD/bbl) (at nominal prices)
price of both oil itself and the stocks of the companies that extract it. There have been two periods of pronounced scarcity in the oil sector. The first period, 1973 to 1980, was triggered by the actions of OPEC. Firstly, it decided to raise the price of oil by 70%, to USD5.11 per barrel. Secondly, it announced its intention to cut production in 5% instalments over time until OPECs members economic and political objectives were met. Oil demand is relatively inflexible in the short term (ie inelastic), and so demand does not tend to fall dramatically in response to a price increase. Therefore, the price increase had to be still more dramatic if it was to curb demand to a level that the new, lower levels of supply could satisfy. In anticipation of this, the market price for oil in fact quadrupled, moving from USD3 per barrel to USD12 in the space of less than a year. However, the read-across into oil company valuations was not straightforward. Of course, these corporations did not derive as much benefit from the price rise as those countries actually owning the oil reserves. Moreover, the sudden nature of the oil price increase led to a recession, which impacted demand in the medium term. So, for example, the share price of Exxon Mobile
Oil sector valuations move in tandem with the oil price
160 140 120 100 80 60 40 20 0 1995 1997 1999 2 001 2003 2005 2007 2009 2011 Bren t crud e (USD/ bbl) (LHS) FTSE Europe O&G Prod Index (RHS)
Source: Thomson Reuters Datastream
1000 900 800 700 600 500 400 300 200 100
Source: BP
42
abc
Steel prices driven by consolidation and increased demand

1200 1100 1000 900 800 700 600 500 400 300 200 1980 1984 1988 1992 1996 2000 2004 2008
Sharp rise in steel demand powered by emerging markets

1400 1300 1200 1100 1000 900 800 700 600 1980 1984 1988 1992 1996 2000 2004 2008
Steel HRC price global av g (USD/tonne)

Source: Bloomberg
World apparent crude steel consumption (mt)

Source: World Steel
actually decreased from USD2.95 at the start of 1973 to USD2 at the end of 1974. Looking slightly further out, though, Exxon Mobiles shares almost doubled to reach USD5 in 1980 (by which time the oil price had reached a peak of USD40 per barrel). The second period of oil scarcity has been more recent, driven in large part by the fact that demand from emerging markets has increased sharply during a period when production has stagnated. Political tensions and conflict in the Gulf region have further exacerbated the situation. From 2003 to its peak in July 2008, the oil price increased by a factor of 4.7x, from USD25 to USD143 per barrel. But on this occasion, because the
appreciation was determined more by basic economics (ie the demand/supply equation) rather than exogenous political factors, companies in the oil and gas sector have directly benefited. The industrys valuation (as measured by the FTSE Europe Oil and Gas production index) rose by 1.6x over this timescale. So, at least in this latter period, the arrival of pricing power in the oil industry had a significant impact on its valuation. We would also argue that a similar phenomenon has been seen in the steel sector. The steel industry has a relatively finite production capacity (at least over shorter time durations) and, in addition, has been undergoing a
Steel price relative to world Iron & Steel EV/EBITDA
Steel industry valuations vs steel price increases

1200 3000 2500 2000 800 1500 600 1000 400 500 0 2002 2003 2004 2 005 2006 2007 2008 2009 2010 2011
1200 1100
14 12 10 8 6 4 2 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
1000
1000 900 800 700 600 500 400 300 200
200
Steel HRC price global av g (USD/tonne) (LHS)
World Iron & Steel EV/EBITDA
Ste el HRC price global av g (USD/tonne) (LHS) Datastream World Iron & Steel index (RHS)
Source: HSBC, Thomson Reuters Datastream
Source: HSBC, Thomson Reuters Datastream
43
abc
process of consolidation (eg Arcelor/Mittal). Meanwhile, in recent years, demand from emerging markets has exploded. The impact on the steel price of this combination of relatively tight supply vs exponential growth has been very marked. Once world apparent crude steel consumption rose above the 900mt per annum mark in 2002, the industry became capacity constrained, and the price began to shoot upwards. By 2008, apparent crude steel consumption had reached around 1,300mt, an increase of 45% from its 2002 level. Over this period, the steel HRC average global price rose 3.8x, from USD223 per tonne to a peak of USD1,079 per tonne. This translated into a massive appreciation in the value of the industry: the world Iron and Steel index rose about seven-fold, from 316 in Jan 2002 to 2,202 in August 2008.
Cisco has increased its global mobile traffic (Terabytes per month) forecast over the past year
4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 2010 2011 Feb-10
Source: Cisco Visual Networking Index
100% CAGR
2012 Feb-11
2013
2014
Exponential mobile traffic growth is the product of three compounding factors: Rising penetration of smart devices such as smartphones, tablets and connected laptops. These devices feature full web browsing and rich media capabilities that in turn spur massive increases in individual bandwidth consumption: as customers migrate from simple voice/text usage to mobile broadband, use of network resource increases typically twenty-fold or more. See illustration.
Exponential demand, linear supply

Our December 2009 report, The Capacity Crunch, set out our belief that the mobile industry faces a major imbalance in the demand and supply of network capacity. Mobile traffic (demand) has doubled each year for the past couple of years and is set to continue growing at an exponential rate for the foreseeable future, perhaps even for the next ten years. Cisco, for example, forecasts that mobile network traffic will grow at a 100% compound rate to 2014. Note also that the firms current forecast (published in February 2011) increases projected volumes from the year-earlier analysis, as shown in the chart below.
Smart devices consume orders-of-magnitude more bandwidth than basic voice/text mobile phones
= = =
X 24*
X 122*
X 515*
Cisco, HSBC. Note: multiples based on Ciscos VNI Mobile 2011
44
abc
As the utility of connected smart devices increases and as the range of mobile applications grows, more and more activities can be executed over the mobile web; hence there will be a propensity for people to use these devices more frequently and for longer periods over time Consumption will also increase indirectly, as mobile applications become ever more sophisticated and web media ever richer. For example, most web video today is quarter video graphics array (QVGA) format offering approximately half the resolution of a standarddefinition (SD) television picture. As mobile video cameras become more ubiquitous and as the quality of video capture on these devices increases, the bandwidth required to upload and download this content wirelessly rises significantly too. A standard definition video file is roughly twice the size of a QVGA format file, while a high definition (HD) file is roughly twice the size again. Bandwidth consumption could thus quadruple purely due to higher quality video formats without the user watching another minute of content
Admittedly, though, this is not the first time that the mobile industry has faced skyrocketing demand for capacity indeed, it has periodically faced capacity crunches a number of times before. This is intuitive when one considers how the mobile market has grown from a handful of developed-market business users in the late 1980s to five billion accounts globally (including 500m mobile broadband users) in a period of a little over twenty years. It has been possible to add sufficient capacity to support this explosive growth in large measure because of the improved efficiency of mobile network technology. It is this innovation that has massively leveraged the comparatively modest increases in spectrum allocated to mobile services over the period, and thus permitted the mobile market to grow to its current size and scope. Successive generations of infrastructure have enabled operators to squeeze 10-20 times more capacity from a given slice of mobile spectrum than did their predecessors. For example: 2G microcellular architectures and antenna sectorisation enabled much more intensive frequency re-use than was possible in firstgeneration (1G) analog networks. Replacing,
Global mobile data traffic 20102015 (TB per month)
7,000,000 6,000,000 5,000,000 4,000,000 3,000,000 2,000,000 1,000,000 0 2010 Nonsmartphones Home Gateways
Source: Cisco
2011 2012 Smartphones M2M
2013 2014 Laptops and Netbooks Tablets Other portable devices
2015
45
abc
say, a four-mile radius analog macrocell with four one-mile radius microcells increases channel capacity four-fold, while splitting each cell into three 120 sectors (known as tri-sectorisation) increases it a further threefold in this example a total capacity gain of 1,200%, purely as a result of these 2G innovations 3G code division multiplexing (CDMA) enabled universal frequency re-use. 2G systems used frequency division multiplexing (FDMA), whereby adjacent cells are required to use discrete frequencies. This had the implication that operators were only able to use a fraction (at best, one-seventh) of their total spectrum in each 2G cell. By contrast 3G CDMA ensures separation between communications by means of a unique code (pseudorandom noise or PN code). This enables every cell to make use of all the frequencies allocated to a given operator. Increasing utilisation of spectrum from oneseventh to seven-sevenths represents a 600% uplift in capacity. In parallel, improved 3G modulation techniques like QPSK (2 bits per symbol) and 16QAM (4 bits per symbol) enable more data to be inserted into each carrier wave cycle than the GMSK (1 bit per symbol) modulation used by 2G GPRS data services thus doubling or quadrupling data capacity. Bringing all these techniques together, it can be seen that the aggregate capacity gain achieved by moving from 2G to 3G is well over 1,000% on a like-for-like basis Innovations such as these have massively improved mobile network efficiency. In turn, this has delivered huge increases in capacity over the past two decades enough to support the compounding demands of mass market adoption
of mobile services and the migration from basic voice and text offerings to mobile broadband. Habituated to this trend of ever-increasing efficiency, it is not surprising that in the face of the current wall of demand created by mass market adoption of mobile broadband there is a widely-held assumption that the latest generation of network technology, 4G/LTE, will perform the same tricks and thereby save the industry from yet another impending capacity crunch. However, we are firmly of the opinion that things will not be so simple this time.
Hitting the Shannon Limit

The problem is that the efficiency of mobile networks is starting to peak. The practical implication of this is that we believe new 4G/LTE technology is barely any more efficient than todays 3G platforms perhaps as little as 30% more efficient on a strict like-for-like basis. This gain is trivial by comparison to previous generational upgrades that have comfortably yielded 1,000-2,000% like-for-like efficiency benefits. The sharply flattening efficiency curve is due to the technology approaching what is known as the Shannon Limit. This is a law of radio physics that governs the maximum amount of error-free data that can be transmitted over-the-air for a given level of noise (or interference). Noise is an unavoidable feature of the mobile environment, with interference created by buildings and other obstructions, signals from other users, and simply due to attenuation of the signal as radio waves travel through the air. The higher the level of noise, the more the network has to compensate by reverting to less data-intensive but more robust modulation techniques as well as dedicating a greater portion of the channel resources to error correction (and thus leaving less for the end-users data). From a customers perspective, the higher the noise level, the lower the data rate that they
46
abc
will receive. For a typically noisy outdoor urban cell, the Shannon Limit is around 2bits/second/Hz. Standard 3G (WCDMA) technology has an optimal performance of around 1bit/Hz, while HSPA-enabled 3G and 4G LTE can increase this to between 1-2bits/Hz. See the grey shaded area in the accompanying illustration. In lower noise environments (for example, where the user is both close and has line-of-sight to the base station, or if the cell is relatively uncrowded, or if it is indoors) then a speed of greater than 2bits/Hz may be achieved. In this type of scenario, the lowering of the noise allows the user to move further up the Shannon curve (up-and-tothe-right in the accompanying illustration). Some reduction in outdoor cell noise may be achieved in the future through implementation of smart beam-forming antennae that can reduce co-channel interference (residual noise from other users). These smart antennas are able to pinpoint a specific mobile device with a narrow beam; this is in contrast to the vast majority of antennas today that blast signals out to a large portion of the
Shannon Limit sets a ceiling on mobile spectral efficiency
sector in an attempt to find the specific user. However, note that smart antennas will only improve one aspect of cell noise (co-channel interference), and hence significant improvements in bits/second/Hz as a result of implementing beam-forming antennas seem unlikely.
The building blocks of efficiency

The spectral efficiency of a radio technology is largely the product of three different areas:
Multiplexing whereby radio spectrum is divided up in order to support multiple simultaneous connections Modulation the process of inserting information into a radio carrier wave Coding or more properly, forward error correction coding (FEC). This is the process of transmitting redundant information in the knowledge that some original data will be lost or corrupted along the way
Comparing 3G and 4G across these three fundamental building blocks of spectral
6
Shannon limit
5
Achievable rate (bps/Hz)
Shannon limit with 3dB offset OFDMA CDMA (HSPA)
4 3 2 1 0 -15 -10 -5 0
Required SNR (dB)
Lower interference environments e.g. isolated hotspots, femtocells
Inaccessible Region
10
15
Typical loaded mobile network (outdoors)
20
Source: Alcatel Lucent, HSBC
47
abc
efficiency, the first observation to highlight is simply the degree of similarity seen between the two platforms: Modulation options for each are identical (standard 16QAM with the option to use more data-intensive 64QAM if conditions permit). Indeed, underlining the relevance of the Shannon Limit is the fact that 64QAM (6 bits per symbol) will likely only function in around 5-10% of the cell area, where radio conditions are at their most benign Similarly, forward error-correction techniques for 3G and 4G are near-identical, each based on turbo coding (albeit with 4G/LTE featuring a minor tail biting improvement) The fundamental similarities between 3G and 4G in terms of modulation and coding mean that by a simple process of elimination any improvement in spectral efficiency that exists between the two generations must be principally derived from differences in multiplexing. 3G WCDMA employs code division multiplexing (systemised as CDMA) while 4G LTE uses orthogonal frequency division multiplexing
(systemised as OFDMA). Both techniques achieve universal frequency re-use, meaning that all frequencies may be utilised in all cells. The only major difference between the two is that while CDMA is a single-carrier transmission technology, LTE uses multi-carrier transmission on the downlink (though note that LTE also uses single-carrier on the uplink). By splitting the signal into multiple sub-carriers, 4G/LTE is able to assign groups of sub-carriers to users based on their bandwidth needs: if the user requires a higher data rate, then a larger group of sub-carriers is assigned; if the user requires a lower data rate, then fewer sub-carriers are allocated. This more granular approach of 4G/LTE versus 3G enables the better matching of mobile network resources to demand with the flexibility of the multi-carrier structure able to fill bandwidth requests more precisely. However, this advantage is generally only available with certain traffic patterns when the network is relatively unloaded. As more users fill the cell, though, radio resource allocations will, ceteris paribus, be governed by a policy that supports users more equitably even if
3G HSPA (WCDMA) assigns resources in relatively large blocks (different colours signify different users)
4G LTE (OFDMA) is more granular enabling better matching of radio resources to demand (different colours signify different users)
T5
T4
Time
Codes
T3
T2
T1
Time
T1 T2 T3 T4 T5
Sub-carriers
1589
= 23 sub-carriers
Source: HSBC
Source: HSBC
48
abc
MIMO theory undermined in practice

This is not to suggest that all innovation will grind to a halt and capacity/efficiency gains with it. However, in our view, it does seem likely that future gains will tend to be rather more marginal than those achieved in the past. Among the most promising new technological developments is the use of MIMO (multiple input multiple output) antenna arrays, which are supported by both 3G/HSPA+ and 4G/LTE. Most antennas today are single input single output (SISO), and incorporate just a single antenna on the device that communicates with a corresponding antenna at the base station. By contrast, MIMO systems use multiple sets of antennas on both the device and at the base station. There are two (mutually exclusive) benefits to this: By transmitting duplicate information, MIMO can improve the reliability of connections in a process known as transmit diversity. This is typically of greatest benefit to users at the edge of the cell, as it is they who experience the highest levels of interference Alternatively, the MIMO antennas can transmit different information from one another, with specialised algorithms at the receiver able to constructively combine signals from the different paths taken by these signals into one, integrated information stream. This creates an additional, spatial multiplexing efficiency gain
Transmitter
Source: HSBC
In theory, moving from todays single input single output (SISO) antennas to 2 x 2 or 3 x 3 MIMO arrays should double or treble the efficiency (and thus the data capacity) of the system. However, achieving these gains in the real world is more difficult, if for no other reason than the complexity of differentiating between the separate transmission streams. MIMOs spatial multiplexing techniques rely on differences in the timing or incidence (angle of arrival) of the received signals. When antennas are spaced too closely together, there is insufficient difference between the signals for the system to work a problem known as spatial correlation. In order to mitigate spatial correlation issues, MIMO antennas need to be spaced multiple wavelengths apart on the device. This creates major issues for smaller devices such as smartphones, which may not have the dimensions to accommodate anything more than the most basic 2 x 2 arrays, where the relative gains versus SISO are smaller.
Receiver
this means that some customers experience download speeds that fall short of the peak capabilities of their device (or that are inadequate to the application or service that they are trying to access). Hence, as cells become more crowded, the greater flexibility of OFDMA is negated, diminishing its relative efficiency gain versus 3G.
MIMO can improve link reliability or increase spectral efficiency via a spatial multiplexing gain
49
abc
WiFi is friend, not foe to cellular

We see WiFi as a useful complement to cellular infrastructure, as a means of offloading traffic from increasingly overburdened cellular networks. We do not believe that WiFi can supplant cellular services, due to a number of technical factors that compromise its appeal as a commercial first-line public access service. These issues include: Range limitations: WiFis lack of scalability will always imply patchy coverage Limited capacity: spectrum assigned to WiFi at suitable frequencies is scarce and unlicensed, so it is contended by a wide range of other devices with equal priority. Moreover, note that WiFi is no more spectrally efficient than leading cellular technologies Limited functionality: WiFi compares poorly to cellular services not only in terms of mobility but also in terms of service quality and security. And because it cannot provide ubiquity, it likewise lacks the secondary added value of capabilities like location awareness, which is an important component in many mobile data applications Around 30% of traffic (both voice and data) typically originates in the home, while a further 20% is typically generated in offices and other places sufficiently close to fixed-line points of presence to use WiFi instead of cellular. Given the prospect of spiralling mobile network traffic, any
WiFi and femtocell percentage offload of smartphone and tablet traffic 2010 2015
China U.S. Brazil France Japan Russia Germany U.K.

Source: Cisco VNI Feb 2011
20% 21% 23% 31% 32% 35% 38% 40%
23% 30% 22% 38% 28% 42% 37% 42%
role that WiFi can play in offloading a portion of this growing burden will surely be welcome. Ciscos VNI index forecasts that (depending on the country in question) roughly 20-40% of traffic from smartphones and tablets will be offloaded to WiFi or femtocells (the formers cellular equivalent technology) see table above. However, we believe that it is unlikely that WiFi can fulfil a more ambitious role, for a series of reasons: Firstly, there is actually precious little suitable WiFi spectrum available. Most WiFi systems exploit the 2.4GHz Industrial Scientific Medical (ISM) band, which is comprised of 100MHz of unlicensed public spectrum. WiFi itself works in broad 22MHz channels; with the inclusion of guard bands, this implies that only three nonoverlapping channels can be accommodated within the ISM spectrum. The practical ramification of this is that no more than three different service providers can efficiently operate WiFi hotspots in the same approximate location.
Only three broad 22MHz WiFi channels can be established in the all-important 2.4GHz band
1 2 3 4 5 6 7 8 9 10 11 12 13 14 Channel 2.412 2.417 2.422 2.427 2.432 2. 437 2.442 2.447 2.452 2. 457 2.462 2.467 2.472 2.484 Center Freque ncy (GHz)
22 MHz
Source: Wikimedia Commons
50
abc
Moreover, it is important to stress that the ISM band is unlicensed. This does confer the benefit that it is free for operators to use (in contrast to dedicated cellular spectrum which is typically auctioned at very high prices). The drawback, though, is that public spectrum is also shared with a host of other devices and services that rely on short-range radio connections. These include Bluetooth-equipped electronics, microwave ovens, remote controlled toys, baby monitors and wireless CCTV links (to name but a few). No one user or device has precedence above any other. So, while operators can opportunistically utilise WiFi to offload cellular traffic, they cannot rely on it. WiFis short-range is a further major barrier to its assuming a more ambitious role. The widelydeployed 802.11g standard is capable of covering a maximum range of around 30m, while the newer 802.11n standard merely doubles this, to around 70m. The ability of network technology to accommodate growth (whether in terms of expansion of coverage, user growth or demand for higher bandwidths) is known as scalability; and it is precisely scalability that WiFi lacks, in our view. A few hundred square feet can be covered with a relatively inexpensive WiFi access point. A few hundred square miles, however, requires a literally geometric increase in investment. For example, it would take 522 x 70m-radius WiFi hotspots to cover the same geographic area as a one-mile radius cellular microcell (see illustration opposite). Even at a modest cost of USD500 per access point, this would total USD261,000 for access electronics alone, compared with perhaps USD15,000 for a 3G cellular base station. As far as the technology itself is concerned, WiFi is no more (nor less) spectrally efficient than cellular technologies like 3G (WCDMA/HSPA) or 4G (LTE) or even WiMAX (IEEE 802.16). All
WiFis inherent range limitations means poor scalability 1 microcell= 522 WiFi hotspots (802.11n)
Source: HSBC
of these draw from broadly the same pool of constituent techniques: spread spectrum (code division multiplexing) or orthogonal frequency division multiplexing, similar modulation options (QPSK, 16QAM and 64QAM, the latter only in benign radio conditions) and coding (low density parity check, LDPC, or turbo codes). With WiFi no more (nor less) efficient than 3G, it therefore largely comes down to a question of how much additional radio resource does WiFi bring to the table? The answer: a further 60MHz globally, shared by all WiFi operators and users (and a host of other devices too) which is hardly in and of itself a game-changer. 60MHz would represent a fraction of the total spectrum that is otherwise allocated to cellular services typically several hundred MHz in total in most developed markets, for example. So while cellular spectrum globally supports 5bn accounts, including 500m mobile broadband users (a figure expected to double to 1bn by the end of 2011, according to Ericsson forecasts), WiFi spectrum currently supports perhaps only a couple of hundred million users at most.
51
abc
So when considering the question of WiFi as a potential substitute for cellular, it is useful to think from first principles. Does WiFi bring significantly more spectrum to the table? No only around a further 60MHz (net of overhead). Does WiFi squeeze any more out of this spectrum than cellular technologies? No WiFi uses broadly the same techniques as cellular; it is generally no more (nor less) spectrally efficient. So is it likely that WiFi could handle the wholesale migration of the worlds cellular users (even just cellular broadband users) onto a thin sliver of unlicensed public spectrum? Clearly not, in our view.
basics approach to adding more mobile capacity. This will involve offloading traffic to WiFi and femtocells wherever possible, but it will also mean increasing substantially the density of cellular networks, through techniques like cell splitting. This latter term refers to the shrinkage of cell sizes, in order to reduce demand on each base stations radio resources. While it is a relatively expensive, time-consuming and complex process, it does have the advantage of being a tried-andtested method of adding substantial capacity. We believe that WiFi can play an important supporting role in offloading traffic from overburdened cell towers Cisco for example forecasts that between 20-40% of total smartphone and tablet traffic can be diverted to fixed line via WiFi and femtocells. However, we are highly sceptical that it can play any more than a supporting role to cellular, due to its manifest constraints (its use of public spectrum that is itself in limited supply; the fact that, with only three channels available, co-location becomes a major challenge; the shared nature of its radio frequencies; and its intrinsically short range). It is therefore our view that, facing the threat of a capacity crunch and without the prospect of new technologies to save the day, operators will need to deploy more cellular infrastructure. This will necessitate an increase in capex, and hence our conviction Overweight call on mobile equipment market leader, Ericsson. However, in our opinion, investment in cellular infrastructure is about considerably more than merely avoiding an imminent network collapse. We believe that capex is likely to emerge as a powerful source of competitive advantage in a way that has not, frankly, been the case in the 2G voice-centric world. The reason for this lies in the difference between circuit-switched voice and adaptive IP data services.
Importance of capex re-emerges

So, to re-cap: mobile traffic is set to continue doubling each year for the next several years, driven by rapid smart device adoption, growing usage patterns and indirect increases in consumption as applications become more sophisticated. At the same time, however, the industrys ability to supply significant amounts of additional capacity is starting to be constrained by the hard laws of radio physics (specifically the Shannon Limit). This is what underlies the fact that 4G technology is barely any more efficient (perhaps only by 30% or so) than todays 3G networks, on a strict like-for-like basis. Some small efficiency improvements may be possible going forward, thanks to innovations like MIMO and smart, beam-forming antennas. However, we expect these gains to be relatively modest and particularly so when set against historic gains from previous generational upgrades (ie 2G succeeding 1G, and 3G succeeding 2G each of which transitions delivered orders-of-magnitude improvements in capacity). With the idea of any quick fix from innovative technology breakthroughs likely off the table, we believe the industry will have to take a back-to-
52
abc
2G voice is a circuit-switched service. Circuit switching creates a dedicated channel which runs at a constant bit rate (9.6-14.4kbps depending on the voice codec), which in turn means that as long as the user is within range of the base station they will receive a broadly uniform quality of service. (Poor call quality is typically the result of being out of range, while call dropping is typically due to either moving out of range, or the cell being over-crowded). The key point to note here, though, is that voice users close to the base station experience broadly the same level of service as those who are much farther away (but still within range). But, while this is the case with 2G voice, it is emphatically not true of a 3G or 4G data-centric world. 3G and 4G are adaptive systems. Adaptive networks change their method of operation based on prevailing radio conditions. In this sense, 3G and 4G base stations might be thought of as tool boxes, offering an array of tools ranging from sophisticated precision engineering devices at one end to basic hammers and chisels at the other. The choice of tool depends on the conduciveness of the radio environment. When radio conditions are benign, with low levels of noise (such as when a user is close to the base station and has line-ofsight), the network can utilise the most sophisticated techniques to cram more data into the radio waves (and simultaneously allocate less of the radio resource to error-correction). But if the user is far from the base station and is subject to high levels of interference (for example, from intervening buildings and other users in the cell), then the base station must revert to the most basic (but robust) of tools. These are less adept at cramming 1s and 0s into the radio wave, and must also incur a greater overhead in terms of error connection.
So while distance matters little in providing a voice service, it matters greatly in the world of mobile data. The closer a user is to a base station, the faster the data speeds that they will receive; and vice versa. This fact should benefit the larger networks. Because they will have more base stations than their rivals, their users will on average be closer to a base station, and being such should enjoy superior bandwidth. Note that two operators might have the same capex/sales ratio, and even the same ratio of customers to base stations but if one operator is twice the size of the other, with twice the number of base stations, then despite the parity in capital intensity, the fact that the average distance of its customers to a base station is shorter implies it should be able to provide a materially superior data service. Of course, a final point to make is that, thanks to their scale, the larger operators should in any case be able to purchase infrastructure at a per unit discount to their smaller rivals which should compound the scale-friendly nature of the effect described above.
53
abc
Operator IPTV has mixed track record; often a response to cable OTT eliminates barriers to entry, ushering in a new wave of rivals Set-top boxes fast becoming a Trojan horse for OTT services
Introduction
In one sense the most popular service running over fixed-line connectivity remains voice. However, thanks to years of aggressive price competition (often stimulated by regulatory intervention) and the arrival of VoIP, voice revenues represent an ever-diminishing portion of the total. Operators everywhere have taken strenuous efforts to rebalance into contracted revenues, in particular line rental and broadband. Given this context, what are the fixed-line applications of the future? One especially stands out: IPTV the provision of a payTV service over a broadband line. However, this is also an area that arouses profound suspicions. Investors are (quite rightly, in our opinion) acutely nervous about the prospect of any telecoms operator moving out of its natural comfort zone and into the unknown territory that is the media sector. Can telecoms operators make any kind of success of this field? The answer, in our opinion, simply depends on what they are trying to achieve. We would argue that most operators would be best served seeing IPTV in modest terms: as a way of improving customer loyalty, as well as a means of stimulating the market for underlying connectivity by demonstrating the power of video delivered over a superfast broadband infrastructure. In developed markets, it seems unlikely that telecoms operators
will have the right competencies to match existing media players or the technology/internet companies that are also eyeing this territory (although arguably there is still everything to play for in some of the emerging markets). The problem for the telecoms operators is that they are not merely going into battle against powerful existing media brands. Indeed, both traditional telecoms and media companies face a potential barrage of new entrants in the form of so-called over-the-top (OTT) competitors piping payTV over broadband connections. Previously, entry into the payTV market has been difficult, due to the very limited number of conduits available to take payTV content into the home (eg digital satellite or digital terrestrial broadcast platforms). But the arrival of superfast broadband infrastructure means that suddenly any entrant has access to a conduit able to convey programming to viewers. As a result, a host of new rivals will appear very likely led by some of the biggest and most innovative companies in the world: the likes of Apple, Google and Amazon.
Telecoms IPTV: a brave effort

Previously, if a customer wanted to receive a video service, they generally had a choice between a cable provider and/or a set of satellite/terrestrial broadcasters. As we argued in our DisContent thematic report (September,
54
abc
2009), the defining feature of this system is that the number of conduits into the home (ie broadcast channels, whether over satellite, cable, or terrestrial multiplex) is very limited. This has understandably made those conduits (such as digital terrestrial broadcast spectrum) very expensive. An entrant with video content wanting to break into the market and distribute its programming therefore faces a real challenge in this environment. But, with the arrival of broadband (and particularly superfast broadband), suddenly any player with content has access to a distribution mechanism. Broadband is, in essence, a conduit available to anyone with content and, with the removal of such a massive barrier to entry, it is hardly surprising that there should be such a large number of players looking to take advantage of the new opportunities on offer. One of the challenges for the telecoms operators is that many of the potential entrants are high-profile and popular brands from the worlds of the internet, technology and retail, as well as well-established media entities. However, the first entrants into the payTV space using broadband telecoms pipes have typically been the incumbent themselves. Most have now launched a payTV service to augment their more conventional voice and broadband service
proposition. Incumbent video offerings typically make use of IPTV an acronym denoting any television service that is provided over a broadband connection using internet protocols. (It generally also indicates that the service is a managed one, rather than a best-effort video stream such as that provided by YouTube).
45m IPTV users globally

In terms of sheer customer numbers, IPTV might not rival the penetration of mobile, but it has nonetheless grown to a material size over a comparatively short period of time. We estimate that there were more than 45m IPTV users in the world in January 2011, a 40% year-on-year increase. IPTV subscriptions represented 8.2% of total broadband lines in Q3 2010, as compared to 6.2% in Q3 2009 (as shown in the graph below). If we exclude cable and mobile broadband accounts, we estimate that IPTV penetration of broadband lines would be closer to 12%. Measured by absolute IPTV customer numbers, Europe is the most significant region, accounting for 46% of global IPTV connections in Q3 2010, followed by Asia (with 35%) and the Americas (with 18%). The Middle East and Africa contributed less than 1% of the total. France is the world leader, with nearly 10m IPTV connections, followed by China (9m including HK) and the US (7m).
Global broadband users (million) and IPTV penetration as a percentage of broadband lines
600 500 400 300 200 100 0 Q2 08 Q3 08 Q1 07 Q2 07 Q3 07 Q4 07 Q1 08 Q4 08 Q1 09 Q2 09 Q3 09 Q4 09 Q1 10 Q2 10 Q3 10 4.0% 2.0% 0.0% 10.0% 8.0% 6.0%
Top 10 IPTV countries, Q3 2010 (IPTV connections, millions)

10 8 6 4 2 0 Fra Chi USA SKor Jap Ger Bel Spa Ita
Global BB users
IPTV penetration
Source: Point topic, HSBC
Source: Point Topic
55
abc
One clear driver of IPTV is simply the deployment of incumbent NGA platforms. As operators roll out fibre-based infrastructure capable of supporting superfast broadband speeds, it becomes much more straightforward to accommodate the bandwidth demands of HDTV streams. Note that Akamai suggest a good quality HDTV picture requires at least 5Mbps; the company estimates that, currently, only around 22% of global internet connections achieve this type of throughput. This would seem to indicate that there is plenty of potential future upside for IPTV services, as the reach of NGA services extends further. But however appealing the prospect of revenues from an entirely new market area, we suspect that the most compelling case for operators to provide IPTV services is defensive. Cable operators are a particular threat to the incumbents, and make extensive use of bundled propositions, combining all the various services a customer might take via a physical communications connection. Furthermore, the cable companies have become a particular threat now that relatively inexpensive DOCSIS3.0 upgrades (which take very little time to deploy) enable them to deliver fibre-type broadband speeds. Where cable operators are pitching a triple-play including payTV and (superfast) broadband, incumbent operators have little choice but to follow. It is therefore perhaps not surprising that there does seem to be a clear statistical relationship between the degree of IPTV penetration in a given country and the magnitude of cable broadband market share there. The accompanying graph plots IPTV subscribers as a percentage of incumbent broadband lines against cable market share of broadband for European operators. The positive relationship indicates that the stronger the cable operator in a market, the harder the incumbent has pushed IPTV.
Incumbent IPTV service penetration of incumbent broadband lines (x-axis) vs cable market share of broadband (y-axis)
50% 40% 30%

SWI BT KPN BEL PT
20% 10% 0% 0%
TI
TEF DT FT
20%
40%
60%
80%
100%
We would therefore identify four different groups of countries: Firstly, markets like Belgium, Switzerland, Portugal and the US, where the push into IPTV has likely been driven by the need to compete effectively against upgraded cable networks Secondly, markets where IPTV has not been much of a focus because the local cable competition is lacklustre, such as Italy and Spain Thirdly, a number of (sizeable) anomalies: countries where telecoms operators have been aggressive with IPTV, despite the fact that cable is weak examples being France and Hong Kong Fourthly, countries where cable competition might well have spurred on IPTV services, had factors like regulation permitted it
Spurred by cable: Bel, Swi, Ger, UK, US, Japan
In a series of markets it seems likely that the presence of powerful cable competitors has been a substantial factor in spurring the incumbents into action with IPTV service offerings. Belgium provides a good example of this, with the incumbent Belgacom doubtless keenly aware of
56
abc
the threat potentially posed by cable rival Telenet (now part of Liberty Global). Belgacom was among the earliest to launch IPTV services in Europe, and now has around 20% market share of the national TV market as at the end of Q3 2010 with the rest belonging to cable operators (predominantly Telenet). Satellite and DTTV are very small in Belgium (respectively 2% and 4% of the market). IPTV penetration reached 60% of Belgacoms broadband connections by Q3 2010. Indeed, IPTV has become a relatively material item for the incumbent, and now provides 16% of its consumer revenues (growing at 43% year-on-year in Q3 2010). The companys triple-play package, which includes fixed-line voice and broadband (12Mbps with a 50GB cap) as well as more than 70 TV channels, starts at EUR52 per month. Telenets equivalent triple-play packages start from EUR45 per month (for 15Mbps broadband with a 15GB cap and 70 digital TV channels). Swisscom is another European incumbent that faces competition from one of Liberty Globals cable operations. The Swiss incumbent started providing IPTV services in 2007 and had secured around 10% of the TV market by the end of Q3 2010 by which point, 23% of its broadband lines were taking an IPTV service. Swisscoms triple play package starts at CHF99 per month and includes line rental, free voice calls to the Swisscom fixed-line and mobile networks,
10Mbps broadband and a TV service with over 160 channels. The incumbent is competing against cable operator Cablecoms triple-play package (CHF75 per month for fixed-line calls, 20Mbps broadband and more than 120 TV channels). Turning to Portugal, incumbent Portugal Telecom (PT) has made a good measure of progress with its IPTV offering. The company has secured control of 28% of the payTV market in a period of only three years (although payTV in total only represents about 47.5% of total TV households). Nearly 80% of PTs broadband lines take an IPTV service. The companys triple-play offering includes 70 channels as well as video-on-demand, 12Mbps broadband and unlimited voice, all from EUR42 per month. Zon, the larger of the cable operators, has a 34% market share in broadband and 60% in TV. Important content such as football is available to both PT and ZON on the same basis. The two major operators presently offering an IPTV service in Germany are Deutsche Telekom (DT) and Hansenet (which is part of Telefonica), under the Alice brand. Meanwhile, Vodafone is presently re-launching its IPTV offering. DT is the largest IPTV provider, though, with 1.4m subscribers (a circa 3.5% share of the TV market), which represents only 11.8% of the companys broadband lines. DT offers IPTV packages both on a stand-alone basis and bundled together with broadband. The price of its stand-alone packages
European incumbents: IPTV performance and market environment IPTV subs % total BB subscribers TV channels Exclusive content IPTV competitors Cable competitors
FT DT TEF TI BT KPN BEL SWI PT TLSN TDC

3.2m 1.4m 0.8m 0.4m 545K 1.2m 920k 358k 0.8m 450K 128k
35.8% 11.8% 13.6% 5.4% 9.9% 46.5% 59.2% 23.1% 79.8% 39.9% 9.9%
130+ Football rights + cinema Iliad, SFR 120+ Football rights Hansenet 30+ None Orange, Jazztel 100+ None Fastweb, Vodafone 70+ None Talk Talk, BSkyB 50+ None Tele2 70+ Football rights+exclusive content none 160+ none 70+ None Sonaecom, Vodafone 40+ none Telenor 38+ none Dansk Bredbnd
Numericable KDG, UnityMedia, KBW Ono No cable in Italy Virgin Media Ziggo, UPC Telenet, Woo, Numericable Cablecom Zon ComHem Stofa
57
abc
range from EUR27.95 to EUR39.95 per month over a 16Mbps DSL connection. The bundled packages prices range from EUR44.95 to EUR49.95 per month over a 16Mbps DSL connection. The TV packages provide 120 channels, although DT also offers 10,000 movies and soaps on an on-demand basis. DTs offering needs to be compelling, because it faces stiff competition from the cable segment, which has been in the process of a major network upgrade to enable it to provide superfast broadband services. Moving across the Atlantic, IPTV has been a major focus for the US RBOCs, Verizon and AT&T. Both have invested heavily in upgrading their access infrastructure to high capacity fibre systems since 2004. Although not strictly an IPTV service (as it is actually broadcast over fibre on dedicated wavelengths), Verizons FiOS TV now has 3.5m customers, equivalent to 28% penetration in the areas where it is available for sale. AT&T uses more economical IPTV switched video over FTTN/VDSL, and now has 3m Uverse video customers. The RBOCs motivation for shifting into the payTV world was in large measure provided by the increasing threat posed by the cable operators with their triple-play bundles. And while Verizon and AT&T have made some solid progress, their IPTV subscriber bases remain small by comparison to the 100m payTV households in the US the bulk of which continue to be serviced by cable and satellite providers (see accompanying chart). In Japan, NTT has developed its IPTV platform both to maintain competitiveness with cable operators as they upgrade broadband capabilities, and to provide an additional 'hook' for households to upgrade from DSL. This service is now demonstrating good momentum, with 1.8m households were subscribing to NTT's optical broadcast service at end 2010, equivalent to a
penetration rate of 12.6% of its fibre accounts. This total includes both the FLETs TV service, where NTT rebroadcasts SKY Perfect content, and the Hikari TV IPTV service, which is operated by NTT Communications. FLETs TV prices begin low (at JPY700, cUSD9 per month), and so have succeeded in attracting some of those cable customers looking to save on their costs, as well as winning viewers new to payTV services. After a period during which NTTs IPTV service lacked appealing content relative to its cable peers, NTT has now been able to redress the balance. In addition to rebroadcasting terrestrial TV, NTT offers more than 70 IPTV channels, including around 40 in HD. Half of current users take the unlimited service, which allows access to about 13,000 video-on-demand programmes; recently released movies are available for JPY820 (cUSD10) each. By contrast, KDDI has focused on broadband and VoIP rather than IPTV. As at the end of September 2010, only circa 10% of its fibre customers took its IPTV service, doubtless reflecting the fact that it has a weaker channel and content line-up relative to that provided by NTT.
IPTV not a focus: Italy, Spain
Cables presence, though, is patchy even in developed markets. Several European countries are notable for having either an, at best, limited
US payTV market share end 2010
Telco 7%
Satellite 33% Cable 60%
Source: Companies, HSBC estimates
58
abc
cable segment, or none at all with Spain a case of the former and Italy of the latter. Telecom Italia (TI) and Fastweb are the two main IPTV providers in Italy, although this service is still minor as compared to satellite representing only 20% of the payTV market and 5% of the total TV market. Telecom Italias IPTV customers are equivalent to only 5% of its broadband lines. TIs chief rival, Fastweb, offers a basic triple-play service with more than 30 TV channels, 10Mbps broadband and fixed-line voice all starting from EUR27.45 per month. In addition, Fastweb and Sky Italia (the largest digital satellite player) have entered into a commercial agreement whereby customers can choose services from either within a joint plan, and then receive a unified bill with dedicated call centre support. This could provide TI with a serious challenge, as it combines strong content from Sky with the high-speed broadband capability of Fastweb. Meanwhile, in Spain, IPTV is now being offered not only by the incumbent Telefonica, but also by the likes of Orange and Jazztel. However, in total, IPTV represents around 20% of the payTV market (with payTV in turn accounting for around 25% of total TV households). Telefonica is the largest player, with a market share in IPTV of more than 90%. But this still only represents 14% of the companys broadband connections in Spain. Telefonicas IPTV package consists of 30 channels and access to a video-on-demand catalogue, and is available for a monthly price starting at EUR10.
IPTV not response to cable: France, Hong Kong
By Q3 2010, France had close to 10m IPTV subscribers, out of 26m homes and 21m broadband lines. The French market has been offering IPTV since 2003, when Iliad launched its triple-play offers over ADSL. Subsequently, not only France Telecom (FT) but also other alternative operators have launched similar bundles: hundreds of channels, video-on-demand, payTV and also catch-up TV. The success of these services has been facilitated by the fact that FTs local loops are shorter in length than in most countries (and also employ thicker wires), which means that the service can work well with even basic ADSL technology. Another factor, though, was the fact that the local loop was made available on an unbundled basis both relatively swiftly and at attractive tariffs (at least so far as FTs competitors are concerned). So France is rather atypical in the sense that IPTVs success was not triggered by a perceived need on the part of the incumbent to address troublesome cable competition (indeed, cables market share of broadband in France is only 6%). Instead, it was Iliads early and innovative move into IPTV that forced competitors to replicate its offer. The above said, it is worth pointing out that most households in France (unlike those in Belgium or Switzerland) still rely on digital terrestrial TV (DTTV) for the bulk of their TV consumption. We estimate that analogue TV and DTTV still provide the main mode of access to television for about 45% of French households. This compares to about 24% for satellite and 22% for IPTV. But IPTV is nevertheless a very popular complement to DTTV services in France. However, while IPTV services might have achieved a good measure of uptake, this is not to suggest that they are necessarily lucrative especially so far as the incumbent is concerned. FT has bought exclusive content, whereas its
In contrast, certain other markets have embraced IPTV services even without the presence of a convincing cable threat. France arguably provides the best example of this, given the relatively limited nature of cable deployment in this market.
59
abc
competitors have preferred to focus on reselling standard channels and video-on-demand catalogues. In particular, FT spent EUR625m to purchase key football rights for the period 201012, and also acquired movie content for the Orange cinema series of channels (at a cost of EUR249m). Oranges payTV offering attracted 1.7m clients as at the end of Q3 2010, a figure equivalent to 53% of its standard broadband IPTV users, but seems unlikely to have delivered attractive returns. As a result, FT has announced that it will not be bidding for the next round of football rights, as well as that it now intends to merge its Orange channels with those of Canal+. Turning to Hong Kong, incumbent PCCW does face a cable competitor, but its enthusiasm for IPTV services has had a long track record, dating all the way back to the turn of the last decade. PCCW now runs the markets largest payTV service based on an IPTV platform. The offering is bundled with the companys market leading broadband service. Indeed, PCCW is the only quad-play (fixed-line voice, broadband and IPTV together with mobile) operator in Hong Kong (with market shares of 57%, 55%, 46% and 14%, respectively). As noted above, PCCW does face cable competition, but it has generally been ineffectual although the company was also doubtless encouraged to push IPTV thanks to the supportive topology of its network (short loop lengths serving dense urban neighbourhoods).
IPTV but for regulation: Brazil, Mexico
gain superiority. Infrastructure permitting, then, IPTV offered by the telcos would seem to make most sense in some of the emerging markets. But, as it happens, the necessary regulation has not always been forthcoming. For example, the development of IPTV in Brazil has hitherto been stymied by the countrys media laws, which prohibit the offering of IPTV over terrestrial telecoms networks. This has resulted in the Brazilian payTV market up to this point comprising a virtual duopoly between a dominant cable operator, NET Servios, and a dominant satellite player, Sky Brasil. However, there are signs that this may be about to change. The telecoms and media regulator Anatel has been keen to foster greater competition in payTV, and so has supported proposed changes to the media law (PLC116) currently awaiting review by the Brazilian senate that would allow telecoms operators to offer IPTV. Although precise timing remains unclear, we believe the necessary changes to legislation will take place in H1 2011. In fact, Anatel has already removed language prohibiting IPTV over telecoms networks in its licence renewals for 2011. From a practical point of view, though, further growth in IPTV will still depend on the roll out of superfast fibre-based broadband platforms. The Brazilian government is keen to improve broadband availability, although the focus here is more on bridging the so-called digital divide for basic broadband rather than on the higher-speed technologies (FTTP or FTTN/VDSL) that are needed for good quality IPTV. The success of alternative operators like GVT might also stimulate further activity. GVT (part of Vivendi) has been offering superfast broadband (ADSL2+, VDSL and FTTP) services in lucrative areas of its home region II; additionally, it now has plans to expand into Rio de Janeiro (region I)
There are good grounds for doubting developed world telecoms operators expertise with media content, and hence for scrutinising their IPTV plans with due caution. However, there is arguably a very genuine opportunity in the emerging markets. Here, operators are often able to more fully leverage their scale and network advantages. Meanwhile, local content may make it more difficult for the US technology giants to
60
abc
as well as Sao Paulo. This should help spur the telecoms operators into fibre investments of their own that are suitable for IPTV services. The IPTV market in Mexico has a similar story to that in Brazil, with current legislation prohibiting the dominant telecoms player (Telmex) from entering the market. What is different, however, is that unlike Brazil the prospect of imminent changes to the legislation seems less likely. Telmex had struck a deal with the government in which it agreed to allow cable operators to interconnect with its fixed-line infrastructure, thereby enabling these companies to enter the voice telephony market. In exchange, Telmex was to be permitted to enter the pay-TV market. However, although cable companies are now permitted to offer telephony, Telmex has yet to receive permission to offer TV (of any kind IPTV, satellite or otherwise). The disagreement between the government and Telmex on this issue has rumbled on for several years, and shows little sign of resolution (although we would add that visibility on this issue is entirely absent and hence there is also the potential for a positive surprise).
increasingly insist that the connectivity be offered to all-comers on equal terms, it is no longer very apparent that expertise in NGA platforms (such as possessed by the incumbents) is much of a relevant consideration. The ability to bundle IPTV with other associated products (particularly broadband itself) obviously still holds appeal, but again regulation now often permits the incumbents competitors to readily replicate this proposition. Over-the-top (OTT) providers are so-called because their offering runs on top of the connectivity provided by a telecoms operators infrastructure. It is very plausible that certain OTT players will not only give the incumbents own IPTV services some stiff competition, but could also challenge existing media brands that are today well-established in the payTV market. Most high-profile among the OTT new entrants are the likes of Google, Apple and Amazon. Indeed, if the challenge confronting telecoms operators looks tough, spare a thought for the payTV stalwarts. For telecoms operators, what is at stake are potentially attractive additional revenues, but likely at low margins (as operators would always have to pay out much of the retail revenue captured in fees to the content owner). For payTV players, what is at stake is their existing livelihood. Admittedly, due to factors such as the complexities of rights arrangements and the power of their existing brands in the entertainment space, it seems likely that todays payTV providers will remain a powerful force in the OTT world, and their sites a popular destination for those seeking video content. But there remains the simple problem that, with the barriers to entry now much reduced, there are likely many companies besides the existing payTV players that will suppose this market holds promise.
IPTV vs OTT
As outlined above, the response to operators IPTV services has been mixed and in only very few countries could the service be described as an unqualified success. However, the incumbents are not the only potential providers of IPTV offerings. Indeed, one of the startling features of superfast broadband in particular is that it opens up a conduit to the home capable of carrying HDTV streams that is potentially available to anyone with content to distribute. Because of their knowledge of the underlying connectivity, it is only natural that the incumbents should want to exploit this opportunity. But they will not be alone. And given that regulators
61
abc
However, OTT services have a very long way to go before they can hope to challenge the viewing figures of mainstream broadcasters. That said, though, those wishing to downplay OTTs significance should beware the deceptive s-curve displayed by so many new technologies: penetration rates should not be extrapolated in a linear fashion from the early adoption phase, as there is usually a point at which uptake rapidly accelerates. In summary, then, the number of market participants is likely to swell significantly, and among the new arrivals will probably be some of the largest, most powerful companies on earth. It is hard to see how this could be an incremental positive for todays payTV companies, even if they are able to retain a significant position in the new OTT order. In order to give an impression of just how many players are involved in the contest for a share of payTV revenues, the section below surveys the way in which this struggle has already manifested itself in the consumer electronics space.
business of video gaming. But both of these devices are also capable of streaming video content from OTT suppliers. Many DTTV and satellite boxes, while primarily intended to enable their owners to watch broadcast content, also contain an Ethernet interface enabling them to handle certain video streams. In many cases, these boxes are produced to work in conjunction with a specific payTV companys broadcast output, but there are also generic standards emerging that work on an open basis, and will thus facilitate access to OTT content from a wide variety of sources. Dedicated streaming devices have also begun to make their appearance, an example being the Roku box in the US, a USD99 device that enables its owners to view content streamed from the likes of Netflix and Amazon. Another example would be the AppleTV settop box, which has recently been re-vamped and re-launched. The GoogleTV platform is also making its entrance, and will be available in set-top boxes initially manufactured by Logitech, in addition to some TVs. Televisions themselves are also getting in on the act. The latest premium models are often internet enabled, meaning that they can directly access streamed video content from OTT sites such as Googles YouTube. Similar functionality is also appearing in supporting devices, such as Blue-ray players. Just to further complicate matters, many participants have their fingers in more than one pie. For example, Sony has its own suite of products such as internet TVs and the PS3, but is also collaborating with the GoogleTV project. Additionally, there are a number of conscious efforts on the part of various participants to establish some common standards. For instance,
The fight for your living room

In living rooms right across the world, a fierce contest is now underway indeed, dear reader, it is more than likely taking place in your very own home. The weapons of choice are the otherwise anonymous plastic boxes that typically house consumer electronics equipment. Sitting beneath the worlds television sets are an ever expanding array of devices, ostensibly devoted to enabling their owners to variously access content provided via conventional means, such as DTTV, satellite, DVD and so on. But almost all of these units today also include what could be regarded as a Trojan horse element, intended for use with OTT video services. Consoles such as the Microsoft Xbox 360 or Sony PS3 might superficially appear to be dedicated to the (increasingly important)
62
abc
in Europe, the Hybrid broadcast broadband TV (HbbTV) consortium has achieved a good measure of traction, with a double digit number of German broadcasters now utilising the format to provide services like catch-up TV. Sony, mentioned just a moment ago, is again a participant. (Note Project Canvas, now renamed YouView, is approximately a UK equivalent of HbbTV, and two may converge in future). There are two asides to make at this stage. The first is that, amid all the intense rivalry on display here, there is clearly the risk that an undue degree of fragmentation takes place, potentially both confusing the consumer and introducing excessive costs for those that own content rights but must port them between numerous different standards and electronic programme guides (EPGs). By way of example, note that platforms often use different DRM (digital rights management) technology: that appropriate for streaming to a games console is not necessarily compatible with that working on an internet TV even if both units share the same manufacturer. The second is that, despite the panoply of acronyms, there remains one practical issue within the home that still lacks a good uniform solution: most houses are simply not wired so as to enable the television set in the living room to take a service from a data feed. A variety of technologies are in contention to address this need, although none would currently seem to provide all the answers. There are new wireless technologies like wireless HD (WiHD) and wireless home digital interface (WHDI) but of course these are not incorporated into the majority of pre-existing consumer equipment, and WiHD is restricted to line-of-sight implementations. WiFi remains the most widely deployed domestic wireless data technology, but does not generally have the bandwidth to support IPTV streaming,
particularly if in HD. Another alternative is powerline technology, which enables data to be streamed over the one form of existing wiring seen in almost all homes, ie that providing electricity. Earlier implementations lacked the bandwidth to cope with HDTV, but there are now faster upgrades available.
The name of the game

Why all this potential duplication of effort? The players in this field would naturally each like to control the customers gateway into the world of streamed video services, hence the desire to see their software interface alongside (or inside!) consumers televisions. A great deal of jockeying for position, as well as hedging of risks, is therefore to be expected. The problem for telecoms operators in this context is the sheer number of players (often with powerful brands) that are eying this space. And while the telecoms network may be an absolutely vital component in the overall service (after all, OTT content can only reach the consumer over telecoms pipes), the problem is that this element is somewhat invisible (indeed, this is the usual dilemma for any service company). In contrast to telecoms players, electronics and technology companies will not have to deploy substantial capital to reach the home; but if it is their EPG that the customer consults to select content, then however unjustly the chances are that it is with them that the consumer will primarily associate the service. Some customers may come to OTT via an orderly decision process, having selected their optimum supplier. However, many others (particularly those less interested in technology) will likely stray into this territory, rather than making a definite plan from the outset. There is a good chance that whichever company is first able to entice them into experimenting with a service may secure them as a customer for its platform.
63
abc
Several telecoms operators have therefore made efforts to get their own set-top boxes into the home, to ensure that their streamed video services are competing for the eyeballs of any customers curious to see what sort of video streaming service their living room equipment can provide. Plainly, if the telecoms operator can exert some form of control over the electronic programme guide in particular, it would be in the powerful position of being able to steer customers towards one video service over another. Even was its own IPTV service to wither, a telecoms operator able to act as a gatekeeper in this fashion would doubtless find it very lucrative. But the problem here is simply getting heard amidst all the activity from so many players in this space. Orange/FT, for example, has made a clear effort to get its own-branded equipment into customers homes, via its Livebox. However, the results have been mixed. In an admirably forward looking move, Orange gave access to the boxs application programming interfaces (APIs), with a view to encouraging the emergence of innovative applications (rather like Apple has successfully achieved with its iPhone). However, this has failed to stimulate the appearance of any kind of developer ecosystem, although this is not to detract from the fact that the devices presence in many living rooms will have made Oranges services an easy upgrade path for its customers. Telecoms operators interested in a set-top box presence are not confined to those with a fixedline focus. Vodafone has recently announced boxes for the German and Spanish markets that are capable of receiving broadcast TV (from a variety of conventional platforms) alongside OTT video via broadband. A particular concern to all market participants (whatever their origins) will be AppleTV. It should almost go without saying that Apple has a formidable reputation in terms of designing and
manufacturing highly desirable consumer electronics. Rivals will be concerned that the AppleTV device itself will lure in customers, particularly those that are already enthusiastic devotees of the brand. Previous iterations of the AppleTV product were arguably underpowered (the device would not support 1080p HDTV) and overpriced (both in terms of the unit itself and the HD movies offered over it). Many competitors may have been relieved that Steve Jobss second iteration of the AppleTV product was not more radical. The first version sold in (by Apples standards) disappointing quantities, not even shifting 1m units. Its replacement offers a greater range of VoD and is also cheaper, yet it hardly looks like a game changer. For this paradigm shift, we would argue that it is necessary to look elsewhere specifically to the iPad. In future, these devices may provide the most immediate viewing platform, with consumers opting to send a streamed signal to a more conventional TV only when they know that several companions are interested in seeing the same programme. Furthermore, the iPad and its tablet equivalents look like the best interface by which to navigate through the likely bewildering array of content available from OTT sources. In future, the purchase of a programme via an OTT video-ondemand service might involve a transaction via an iPad, rather like the purchase today of an app. So, even on the assumption that customers continue viewing their programming on television screens (rather than on the tablets themselves), Apple through its iPad has control of the platform that could effectively become the equivalent of the remote control in countless households. Hence Jobs may have concluded that with or without AppleTV his company already owns the most obvious gateway into the OTT world for many future viewers.
64
abc
Where Apple goes, Google is usually not too far behind, and OTT video services are no exception. The GoogleTV offering seeks to meld together internet and TV content, capitalising on some of the companys traditional strengths in search. Following the conventional web model, the platform is fully open. As usual, Google will be looking to monetise the opportunity via advertising (clearly it has its eyes on the enormous television advertising market). Interestingly, though, Google has chosen to include some relatively traditional partners: for instance, Sony and DirecTV (indeed, working alongside a conventional US satellite broadcaster will undoubtedly raise some eyebrows). But Googles most powerful weapons remain its brand and core search competence (giving users an easy way to find programming amid the otherwise confusing landscape of OTT). But, lest it be supposed that the greatest threats to media and telecoms companies seeking to monetise the OTT opportunity originate entirely out of technology companies based on the west coast of the US, the intentions of various supermarkets also merit mention. In the US, Walmart has already entered the OTT field, while in the UK, Tesco also has its eyes on this area. Indeed, Tesco is actually looking to sell its own internet connected set-top boxes, which will have video-on-demand functionality. Any ambitions held by Tesco in the OTT field are highly relevant to the UK market because of the enormous power and clout of its brand and retail presence.
Indeed, many telecoms operators do seem to be retreating from the idea that they can muscle in on the traditional broadcast space. For instance, PCCW in Hong Kong decided that it was simply too expensive to renew its English Premier League rights exclusivity. Meanwhile, the intervention of the regulator has persuaded FT that it should re-evaluate how to go forward with its Orange Cinema and Sport channels (following complaints from Vivendi concerning their exclusivity). It made sense to acquire exclusive rights when it was possible to use these as a mechanism to attract customers to a given distribution platform. But with regulators increasingly intolerant of such exclusive arrangements (and not just with respect to the telecoms operators), this is no longer such an appealing strategic option. The content rights business model is, instead, moving in the more open direction implied by OTT. But at least the regulators are looking at exclusivity arrangements established by all parties (and not just at those put in place by telecoms players). This should limit the ability of dominant payTV players to draw viewers onto their proprietary distribution platforms (eg DTTV or satellite) via exclusive content. In turn, this should leave telecoms networks (including cable) as the most natural and flexible content delivery solution a fact that operators should benefit from through their connectivity revenues. As an example of this process, note that Ofcom has referred BSkyB to the UK Competition Commission on the grounds that it might be necessary to reduce Skys ability to act on incentives to exploit market power, by requiring it to provide wholesale access to linear and S[ubscription] VoD premium movie content on regulated terms. What makes this so important is that the availability of a wide range of video content via OTT platforms is very likely to prove one of the
What operators can do

What should operators do to confront the challenges that beset their IPTV businesses? Most lack a brand that would naturally be associated with entertainment; nor do they own exclusive content, or enjoy the benefits of having (say) a powerful internet brand.
65
abc
most powerful drivers of uptake for telecoms NGA fibre services. Regulators will be aware of this, and given that the deployment of such platforms is in many places regarded as a priority for national competitiveness, may view sympathetically the need for there to be attractive video content available over NGA connections, in order to improve the economics of the roll out. Alas, the matter of winning payTV customers involves more than just assembling the right programming. For example, BT had long been looking to obtain access to some of BSkyBs sports content in order to boost the appeal of its Vision service. Having at long last (thanks to regulatory intervention) secured this programming, it undertook a substantial advertising campaign, highlighting the cheap availability via its offering of Sky Sports 1 at just GBP6.99 per month. Admittedly, the new content did enable BT to improve its Vision net adds for the quarter (to 24k from 14k in the previous three months), but this is hardly an impressive acceleration, especially given the marketing undertaken. Perhaps, then, the issue is really one of brand: operators are simply not seen as credible suppliers of video entertainment. Another alternative for the telecoms companies would be to extend their remit into elements of the applications layer other than IPTV that are perceived as being more functional in nature (and thus closer to the operators core competency of providing connectivity). DT is among those that have developed services of this type. For instance, customers are offered a network-based address book that can be used via PCs as well as mobile phones, on-line storage accessible by a variety of means, including through TVs equipped with DTs set-top box, control of IPTV programme recording via PCs and mobile handsets, and so on. But, while a
useful feature set, it will be apparent that many (if not quite all) of these capabilities are available from the big internet and technology brands, and customers may well gravitate towards using these providers rather than their telecoms supplier. Another alternative would be to exploit one unique selling point that telecoms operators do possess in the form of their extensive call centre resources. Operators are often heard complaining that they are generally on the receiving end of the telephone call when something goes wrong with a consumers set up, even if the problem is more likely to be with a supplier of the electronics or software. Since the technology companies responsible for the latter components of the service are usually much harder to reach, the customers first port of call is to ring their telecoms supplier. While it might be frustrating for operators to have to take a call about a problem that is not necessarily in any way their responsibility, it is perhaps worth considering whether if this is going to happen in any case it is worth making a virtue of the situation? Operators support capabilities are an advantage that could be leveraged with the suppliers of consumer electronics to persuade them to enter into partnership, and (at the very least) incorporate pathways guiding users to the telecoms companies OTT services.
66
abc
Mobile applications
Operators may regret failure to push RCS as a defensive measure Connectivity revenues at risk in markets lacking app ecosystem Soft SIM threat overstated, as not in vendors own best interests
Introduction
The mobile applications layer is now firmly in the possession of companies from the technology industry. Indeed, even the term application has effectively become synonymous with Apple and its app store. Even established device vendors are struggling, and Apples most powerful rival in this space looks set to be Google, with its Android platform. The investment markets have not unreasonably long since written off the operators chances in this arena, and hence the rise of the likes of Apple and Google in this field has not come at the expense of telecoms valuations. But there is a risk that the operators find themselves not only cut out of the upside from applications, but also dis-intermediated from services that have traditionally been their preserve. For instance, applications may become the gateway to communications services and make no acknowledgement of the underlying connectivity upon which they depend. In this context it is disappointing not to see the operators showing more commitment to platforms like the Rich Communications Suite (RCS) initiative, which might at least have helped them to contain certain threats. That said, though, the operators do continue to need the applications ecosystem to drive innovation and thus the take-up of data services
that will exploit the connectivity they provide. In a country such as China, where there is a risk that the ecosystem could become stunted (because the like of Apple and Google are unlikely to have any hands-on control), this might in turn actually hamper the take off of the entire mobile data market segment. The operators, then, do need the applications ecosystem; but the reverse is also true: the technology players need the operators, which foot most of the bill for the mobile device upgrade cycle. Consequently, we think it unlikely that vendors such as Apple will either launch as mobile virtual network operators (MVNOs) or introduce soft SIM functionality (which makes phones re-configurable to other networks) as such moves would risk jeopardising their relationship with the very companies that provide the bulk of their revenues.
RCS: Relegated to Connectivity Service?

We have long warned of the potential danger to the operators of leaving the technology sector in control of the smartphone software and applications layer (for instance, see Avoid Android, 9 November 2007). However, it is now abundantly clear that the operators have almost entirely ceded this space and, as a result, it will
67
abc
be technology companies like Apple and Google that harvest the potentially rich pickings here. The financial markets have essentially accepted this state of affairs as a fait accompli, and there has been little noticeable effect on operators' valuations as a result chiefly because investors had long-since given up on their ability to monetise the more sophisticated elements of the data opportunity (anything much, that is, beyond connectivity). However, it is worth asking whether the markets currently fully appreciate the extent of the territory sacrificed to the technology sector. The point here is that, once the applications layer is outside of the operators control, everything other than connectivity potentially becomes a service that is delivered OTT, rather than by the network operator. This applies not only to new services like IPTV, but also to the most traditional of telecoms fare, such as voice telephony. Not surprisingly, the operators are making some (late) attempts to at least regain a measure of influence given that to retake control looks wellnigh impossible at this stage. One such initiative
Number of applications available by platform
is the Wholesale Applications Community, an effort by a broad swathe of operators to provide a common platform for developers (over both different handsets and networks), potentially streamlining development processes and multiplying the addressable market. However, such initiatives are almost certainly a case of too little, too late, in our view particularly given the long-standing hostility felt between the developer community and the telecoms industry due to the latters historic inflexibility as regards the design and revenue-sharing models for mobile applications. Another telecoms initiative is the GSM Associations (GSMA) Rich Communications Suite (RCS), an effort to produce a set of standards enabling mobile networks to provide the type of functionality otherwise associated with software-led platforms. For example, RCS has a presence capability, able to inform users whether their contacts are available for communication, in the same type of way that instant messaging systems work today. This is plainly a very useful capability if users are to be persuaded to continue to rely on switched mobile
350,000 300,000 300,000 250,000 200,000 150,000 150,000 100,000 50,000 0 Apple Android end-2009
Source: Distimo, HSBC estimates
600% 500% 400% 300% 200% 25,000 6,000 4,500 100% 5,000 0% Nokia Ov i end-2010 BlackBerry Grow th y -o-y Window s 7
130,000
20,000
18,000
68
abc
calls for their voice communications needs, rather than electing instead to use OTT VoIP applications. However, RCS has attracted limited support from the industry, with most operators non-committal. There have been initial trials, for instance in the South Korean market, but a typical complaint relates to the fact that it has proven difficult to monetise the services in question. The situation bears comparison with the lack of enthusiasm shown for the IP multimedia subsystem (IMS) or the GSMAs OneAPI initiative: two other platforms that have inspired little enthusiasm, likely as a result of the opaque nature of the perceived business case. Naturally, this state of affairs has a knock-on effect with the handset vendors, creating something of a chicken and egg problem (ie without RCS-enabled handsets, why push the services, but without the services, why produce RCS-enabled handsets?). Alas, a clear revenue upside opportunity may prove to be a luxury: the battle is arguably rather to retain ownership of the customer in relation to their communications needs in particular, given the way in which social networking sites like Facebook could become their users embarkation point for all their communications needs. For the telecoms operator to be dis-intermediated in this way might not actually cost much revenue in the short-term (as the customers communications still rely on its connectivity), but very likely entails longer-term risks.
them for innovative new services. Indeed, it is difficult to see how mobile operators can keep up, given that telecoms is an industry composed of large (and therefore generally slower moving) organisations where standardisation is regarded as fundamental. Hence their very nature means that they lack the flexibility of technology companies to invest in new services without any clear picture of how the revenue model will develop.
APIs offer web developers access to key operator network functions to add greater value to applications
Source: HSBC
The above said, operators do retain certain advantages. They are likely more trusted than internet names in terms of security, have extensive customer support facilities (an area where many technology players have almost no capability whatever), and should if the necessary resources are committed to the relevant development work be able to leverage their control of the underlying connectivity (for instance, by providing QoS-related functionality). Note that much of the effort with RCS is to provide developers with APIs (essentially software interfaces) enabling developers to access and leverage the capabilities of the mobile network, for instance to initiate a call or a text message. Arguably, there are now just too many APIs competing for the attention of developers and (as mentioned above) this will likely leave developers graduating by default towards those
The API wars

Facebook already provides a series of APIs for third-party applications to hook into, and most developers (given their IT background) are more at home co-operating with another technology player rather than a telecoms operator. Moreover, the greater the popularity of social networking sites (and their associated applications), the more likely it becomes that mobile users will look to
69
abc
provided by fellow technology companies, rather than to those provided by the telecoms industry. But even if developers can be attracted to exploit this platform, given the operators are no longer in control of the applications layer, there is no guarantee that the application will in any way signal to the user that they are utilising the telecoms networks connectivity in just the same way that Skype draws no attention to the fact that it relies on telecoms bandwidth. So, from a branding perspective, even were RCS to be readily adopted, the operator may still lose out as a result of its having ceded control of the applications layer. It is worth pointing out at this stage, though, that we do not regard VoIP as a practical technology for mass market voice communications over mobile at present. As detailed in greater depth in our thematic reports Frequonomics and SuperFrequonomics (March and September 2010 respectively), 3G platforms are intrinsically unsuited to the quality of service that is demanded by VoIP. However, VoIP should work better with 4G/LTE, where it is the only means of carrying voice (given the new technologys lack of circuit switched capability). It will, though, in our view, take a number of years for 4G/LTE platforms to mature, and for handsets using this technology to become mass market. Moreover, most operators have every incentive to retain lucrative voice traffic on their circuit-switched, metered 3G (and 2G) infrastructure, which will also tend to limit the pace of transition towards VoIP. Hence, over the next few years, the threat looks limited from either VoIP or the fact that the operators may no longer exercise control over how users initiate their voice calls. But, looking out rather longer term (which, admittedly, could mean 5 years plus), losing control of this important interface could spell real problems with regard to branding. After all, even if the reality of
the situation is that operators are best advised to stick to what they know that is, essentially, providing the proverbial dumb pipe it would perhaps be best if they could retain greater engagement with the customer, so as to limit churn and avoid being perceived merely as a commodity supplier (albeit one owning and operating an intrinsically scarce resource).
Flourishing ecosystems best

Had the operators displayed real aptitude and enthusiasm for the applications layer, then the capture of this segment by players from the technology sector might have been perceived as more of a distinct negative. However, the reality of the situation has been that, although some extravagant claims were made for the scope of the operators opportunity during the internet bubble, most investors have long-since written this off. So, while we might decry the mobile players failure in the application space, in a scenario where there is no associated value being ascribed to them, arguably the most important consideration is that someone pursues the opportunity whatever their identity. In other words, if there are no good reasons to suppose the telecoms companies will get applications right, then it is better that the challenge be taken up by those fit for the task (ie the likes of Apple and Google) rather than the whole field be neglected. Provided the applications layer is properly served, customers are likely to want to take advantage of the exciting services on offer, and therefore to take out data plans and thus the operators benefit via the resulting connectivity revenues. If, on the other hand, the applications layer remains rudimentary, the incentive for customers to adopt data connectivity services will simply not be present. To summarise: if operators have not the aptitude to take the whole pie (ie applications + connectivity), better that a rival provide the applications so that the operator can take the
70
abc
connectivity half of the pie, rather than the absence of applications mean there is no associated demand for connectivity (ie no pie at all!). The global nature of brands like Apple and Google might lead one initially to suppose that their involvement in the applications layer would ensure that this field could develop rapidly in every important market. However, there may well be at least one substantial exception to this: China. It seems most unlikely that Apple or Google will be left in charge of running the applications layer in China, a responsibility that is more likely to be given to the telecoms operators. However as elsewhere on the planet there is little reason to suppose that these companies have the right skill set to make a success of this enterprise. This might not dim the appeal of iconic devices such as the iPhone, which possess clear status appeal, but sales of high-status devices need not translate through into data connectivity revenues. The risk is, therefore, that without the proven track record of innovation of the global technology leaders, data connectivity could be a damp squib in China.
branding and reputation to attract customers. Operators would know from experience the power of these brands, and the stronger operators would also be aware that their weaker rivals might find it difficult to resist such an opportunity. Although the technology player in question would own the retail customer, they would still be paying the mobile network operator a wholesale fee. This could be perceived by smaller network operators as a short cut to gain market share; and the knowledge that the smaller players would find such a proposition compelling might force the hand of a larger operator into itself accepting the proposal (with the justification that, if it were to refuse, one of its rivals would get the business instead). But, in our view, there are good reasons to suppose that the technology players will not in fact take this route. The first of these revolves around returns. We certainly believe that telecoms in general (including mobile) generates an attractive return profile, but we would not claim the ROIs are as high as those seen from successful technology companies. So the first question is, why would the technology players want to dilute their returns by becoming, in effect, mobile operators? This is arguably particularly the case given that, as a MVNO, they would be perceived as responsible for something that would not actually be under their direct control: the quality of the mobile network that they were reselling. Mobile coverage will always be patchy, given issues of interference, in-building penetration difficulties, and so on. At present, it is the telecoms operators' brands that take the hit for the inevitable problems of mobile reception. It seems unlikely, in our view, that the technology players would want to be on the receiving end of the blame for this type of issue, especially when they do not directly control the infrastructure in question.
Softly softly on Soft SIMs

Given the tremendous success that technology players like Apple have had in terms of building their presence in the mobile applications layer, it is hardly surprising that concerns have been raised as to the possibility of their attempting to extend this power base further. For instance, consider the scenario where they decided to push into the connectivity space itself the mobile operators' heartland or, alternatively, where they chose to disrupt this space by introducing innovations such as 'soft SIMs' (of which more below). How might technology players like Apple (or Google) go about making a move into the connectivity layer? One route would be for them to launch as MVNOs, leveraging their powerful
71
abc
There is also the problem that any move into the MVNO space by the technology players would effectively involve competing against their own customers. It might wrongly be supposed that the customers of a company like Apple are the people who emerge from shops having purchased an iPhone. But this would be a simplification, because most of those 'buying' an iPhone pay nothing like its full list price, instead taking a handset subsidised by a mobile operator. So, if the identity of the 'customer' is defined as the party paying the largest element of the list price, then this is usually likely to be a mobile network. Clearly, if (for example) Apple was to decide to launch as a MVNO, it would thus become a direct competitor at a retail level to all the mobile networks in the country in question. This would doubtless leave the operators much less keen to subsidise Apple's devices. Thus Apple might lose out to its handset rivals, in particular to those producing equipment powered by Google's Android platform. Apple might (not unreasonably) argue that its devices have such a hold on the popular imagination, that operators would simply have to continue subsidising its devices; but, nonetheless, they would surely become more wary of doing so, as well as keener to find alternatives. It seems unlikely that this risk is worth running simply in order to gain the relatively thin margins of a wholesale operator, particularly given the associated risks around brand/network quality outlined above.
mobile customer deciding to churn between two networks would have to remove the SIM card of their previous operator, and replace it with one provided by their new operator. Not only is this process fiddly, but there is also the strong possibility that the original operator will have 'locked' the mobile phone such that it will only work with one of that operator's SIM cards. Mobile networks do this in order to ensure that, once they have subsidised a handset's retail price, the customer does not simply churn to a competitor (after all, they need to earn a return on the handset subsidy). Note that handsets can be unlocked by those with the requisite know-how, and while this is a service that is widely available, it is also somewhat disreputable. Soft SIMs make the process of switching mobile operator much simpler by enabling the user via on-screen software to re-configure the device to work on different mobile operators specific frequency bands as well to be recognized by their AAA (authentication, authorization and accounting) servers. For instance, there need be no fiddling around removing casings and manipulating diminutive slivers of plastic. But, from a business model perspective, there is a clearly a major potential headache here. Just as with regard to the risk vis-a-vis technology players becoming MVNOs, the problem is that to introduce soft SIMs would be to work very clearly against the operators' interests. And, in a business model where the operator (via the subsidy) is, in effect, the real customer (since it is responsible for paying the bulk of the handset list price), this is plainly highly risky. Operators subsidise handsets in the hope that they will attract customers to their network. If customers are able to churn as soon as they have their handset (courtesy of soft SIM functionality), then there is little incentive for the operator to subsidise the device at all. This would leave the customer having
Soft SIMs counter-productive

What, then, of the risk of technology players introducing so-called soft SIMs? (News reports in November 2010, for example, suggested that Apple was planning to integrate a soft SIM into the iPhone, before the firm beat a hasty retreat following the vocal concerns of several mobile operators). At this stage, the phrase soft SIM may require some explanation. Traditionally, a
72
abc
to pay up front the full list price for the handset. In such circumstances, there is little doubt but that customers would upgrade their handsets much less frequently, and thus that the handset vendor would take a considerable revenue hit. It would therefore be surprising, in our view, if the technology companies were prepared to risk the arrangement in which they currently find themselves. It is perhaps worth reviewing again in outline the present business model. A handset vendor like Apple develops expensive, premium devices with a very substantial list price that, were it to be reflected in the retail price, would surely deter many would-be users. But the retail price is then substantially reduced through the subsidy of the operator, enabling the likes of Apple to sell far more devices than they would otherwise do and yet Apple still receives the full list price. This is clearly an advantageous state of affairs for the vendor: Apple in effect gets to sell far more of its devices than it would if its retail price was its list price. Why jeopardise such a beneficial business model?
One set of circumstances in which this might be justified would be in the scenario where the operators looked intent on moving into Apple's territory. Given the operators' poor showing in terms of developing the applications layer, and the tremendously successful entry into this space of the Apple App Store in particular, the technology players could be forgiven for thinking that this now represents their own home territory. As such, belated efforts by the operators to get involved, such as the Wholesale Application Community (WAC), would likely appear threatening (notwithstanding the low chances of success that we would ascribe to this particular initiative). So, one possible explanation for the sudden interest in soft SIMs is that Apple was well aware of the counterproductive nature of this functionality, but wanted to send a warning shot across the operators' bows in view of developments such as WAC. In summary, therefore, we are sceptical about the concept of technology companies either becoming MVNOs or introducing soft SIMs. Neither innovation would seem to make sense from a business model perspective, although both might make for useful threats in order to persuade the operators that the mobile applications layer is no longer a suitable goal for them.
73
abc
Winners and Losers

The scarcity is in the connectivity, so network investment vital Fixed-line winners: cable plus telecoms operators investing in fibre Mobile winners: scale players to enjoy quality of service advantage
Introduction
The emergence of pricing power in both fixed-line and mobile telecoms should be a powerful positive right across the industry, particularly after having been absent for so long. We would stress, however, that because pricing power is being driven by the connectivity layer, the benefits are by no means evenly distributed. In terms of the fixed-line subsector, we believe that the bestplaced operators are those who are investing in fibre-based next generation access (NGA) infrastructure, which by dint of the high capital intensity of such projects is ensuring scarcity in the market for super-fast broadband provision. Incumbents furthest ahead deploying NGA will be the first to benefit; among the Europeans this includes smaller incumbents like Swisscom and KPN of the Netherlands, which are generally further ahead than their peers in larger markets. In the CEEMEA region, Israeli incumbent Bezeq looks best-placed, while in the US, Verizon, which has deployed FTTP to nearly 16m homes, stands to benefit most, we believe. However, even ahead of the best-placed telecoms operators, we think that the greatest benefits will actually accrue to the cable companies (Germanys KDG, multinational Liberty Global, Belgiums Telenet, Virgin Media in the UK and ZON in Portugal, as well as recently re-floated Danish incumbent TDC which also has cable assets). This is because digital cable
platforms can take advantage of a rapid and relatively low cost upgrade to NGA via DOCSIS3.0 modem technology. In the mobile subsector, we favour larger operators with the necessary scale to out-invest their smaller rivals. Here again the application of capital should be a positive differentiator, enabling operators with deep resources to deploy a more densely-built mobile network. This should mean shorter average distances between their customers and their base stations, which in turn should result in a faster and more robust data connection clearly a significant source of competitive advantage. As a consequence, we prefer the mobile exposure of scale players like Vodafone, Telenor, NTT DoCoMo and KT. We believe that the tremendous growth in mobile data volumes will mean that operators will need (and want) to invest more in capex. The mobile equipment vendors ought to benefit from this, with our favourite being market-leader Ericsson, which has the highest exposure to the mobile sector. In the emerging markets, the dynamics are somewhat different. There should still be the opportunity to monetise scarcity in terms of connectivity, and hence we would suggest stocks like TIM Part in Brazil. However, additionally, there may still be some scope for operators to
74
abc
capture a portion of the applications layer: we would advocate STC and MTN on this basis. We see the losers on the themes discussed in this report as being those companies that have been slow to invest in their core connectivity capabilities, like Telecom Italia and TPSA in Poland.
themes considered in this and previous documents, set against a basket of valuation measures. We highlight stocks that screen well on the topics discussed in the current report, but which also do well on our broader analysis of thematic issues. In Europe, this applies to Vodafone (obviously a key beneficiary of this particular reports central theme too) as well as to KPN and Telefonica.
Developed world: winners and losers on all themes

The specific theme examined in this report the emergence of scarcity and its ramifications in terms of bestowing a measure of pricing power upon the operators is certainly topical, and merits special focus, in our view. However, this theme is only one of many that are relevant to our overall evaluation of the companies in the telecoms sector. The accompanying chart therefore summarises our assessment of developed telecoms stocks on all the various
CEEMEA: winners and losers on this theme

We believe that there is an opportunity for telecoms operators in the connectivity layer in both developed and emerging markets. However, emerging markets are somewhat distinct from their developed counterparts, in that there may still also be some scope for operators to monetise the applications layer as well. In view of the
Stocks assessed in terms of full range of strategic themes (x-axis) vs valuation on a basket of measures (y-axis)
5.00
4.50
Key: O/W = N/W = U/W = (size = mkt cap)
Tnet Elisa
4.00
MOBI TLSN BT
3.50 Valuation relative
Swiss Vod
Tef
Belga TA PT Tele2
KPN
FWB
3.00
CT
2.50
VZ DT SNX Tnor OTE
AT&T
2.00
KDDI TI FT NTT
1.50
Bouy
Thematic relative
1.00 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00
Source: HSBC
75
abc
different scope of their opportunities, below we set out winners and losers in the CEEMEA region as an example of emerging market operators prospects.
In the CEEMEA region, there has been a relatively slow transition to next generation access (NGA) superfast broadband at the fixed-line operators. Ultimately, this shift could potentially move the competitive landscape away from unbundling and towards wholesale in the most mature telecom markets in CEEMEA, like in CE3. The slow speed of deployment is mostly due to the desire on the part of the telecoms incumbents to negotiate first with the regulator some guarantee for a minimum return or a more accommodating regulatory environment. Typical examples of this approach have been seen at TPSA in Poland and Bezeq in Israel. In 2009, TPSA entered into an agreement with the regulator that requires the company to provide 1.2m high-speed broadband lines. This will require an investment of cPLN3bn over three years, but has helped to ease regulatory pressure somewhat. Due to the delays in NGA roll-outs, we have not yet seen signs of a migration from unbundling to wholesale. And note that in some countries, like Russia and in Africa, ULL has not even been implemented. However, increased investment in NGA has precipitated a degree of consolidation between small cable operators in CE3, as well as triggering the acquisition of smaller cable TV operators by CEE incumbents. Recent acquisitions of regional cable operators by Magyar Telekom in Hungary and by MTS in Russia are examples. In most CEEMEA telecoms markets, regulation is not particularly stringent as regards net neutrality. In fact, this issue has only really become a topic of discussion for countries like CE3 that are
converging towards European telecoms regulation. The absence of net neutrality regulation is clearly a positive for emerging market telecom operators, since it will enable them to manage their resource (and available capacity) more efficiently, and even to charge a premium for prioritisation. Another recent positive regulatory trend is for the media and cable companies tends to be increasingly compared with the telecoms operators, and included within the remit of new regulation. Hence, the regulatory advantage previously enjoyed by cable operators is gradually eroding. A similar conclusion could be drawn vis--vis taxation: for example, the new communications tax in Hungary applied to both telecoms and cable operators. The monetisation of fixed-line connectivity varies greatly between countries in CEEMEA. Key factors to consider are affordability, levels of income disparity within the population and also the level of availability of substitutes. For instance, Polish subscribers spend a relatively modest amount on broadband compared with customers in Czech or Saudi. In 2010, TPSA reduced fixed-line broadband pricing, through which it intends to reengage with the core market. However, broadband competition is fierce in Poland, mainly thanks to the cable operators, which are upgrading their networks to DOCSIS3.0. Overall, we therefore believe that TPSA should be seen as vulnerable to the cable threat. The Saudi environment is much more supportive when it comes to the monetisation of broadband connectivity. There are cable or altnets competitors offering fixed-line broadband; mobile broadband (via dongles) are the only alternative. Currently VoIP is banned in Saudi Arabia, with the consequence that broadband is not a direct substitute for voice services. There is also likely to be a healthy demand for broadband in the long
76
abc
term, given Saudis attractive demographics, favourable regulatory environment, lack of entertainment options and the absence of alternative technologies. In Turkey, the main inhibitor of fixed-line broadband growth in Turkey is PC penetration. Note that there is only a small difference between PC penetration and broadband penetration (ie households with a PC but no broadband). The attempt to increase prices for unlimited packages in 2011 to adjust for sharply rising inflation and to impose fair usage quotas could negatively impact subscriber and ARPL growth in the near term. Furthermore, the nominal price increase does not represent pricing power in real terms, given the c7% inflation in Turkey that is expected by HSBC economists in 2011. However, positives for Turk Telekom include the absence of regulation promoting net neutrality and the fact that local loop unbundling remains unattractive for altnets.
able to learn from developed-world operators missteps. For instance, it is interesting that comparatively few have launched all-you-can-eat data tariffs, with most instead (very sensibly) opting for plans with usage caps. Another mistake (in our view) committed by the developed-world players was to cede the phenomenally important applications layer to the technology sector. In the CEEMEA region, MTS and some of the African operators like MTN look well positioned. All have introduced capped data tariffs, face relatively little competition from fixed-line broadband and have upside via content and/or VAS. MTN is well placed in terms of pricing power on mobile data. All the mobile operators in Africa currently offer tiered data plans there is no all you can eat data plan currently. Data usage in South Africa has increased 49% since December 2009, with MTNs data revenues growing 42% year-on-year in H1 2010. Even more impressive is the 58% year-on-year growth seen in non-SMS data revenues, which accounted for 11% of MTNs South African service revenue. Among its larger operations, Nigeria currently has the lowest proportion of revenue (4.2%) coming from data services, the key reason for this being simply a lack of required bandwidth. However, with the landing of various submarine cables in Africa, we expect the capacity bottleneck issues to be removed soon. This should, in our view, allow MTN to capture the vast growth potential for data usage in Nigeria. However, there are risks relating to the entry of a credible new entrant (Bharti). In Russia, growing demand for data is providing mobile connectivity with momentum. 3G USB modem numbers have now overtaken fixed-line broadband subscribers. In Q3 2010, VAS (SMS included) amounted to 20.5% of revenue in Russia (a 2.1 ppt year-on-year increase), boosted by the expansion of the 3G network and the fast
Mobile connectivity
As discussed in our SuperFrequonomics report (September 2010), data is not a story confined to the developed world indeed, in many ways, it is actually operators in emerging markets that are better placed on this theme. Emerging markets are, naturally enough, further behind in data adoption, but as a result should have more of the growth ahead of them. Affordability issues will do much to dictate the speed at which data services penetrate the customer base, but the pace at which smartphones and PCs are falling in price is remarkable: it should not be long before USD100 smartphones make their appearance. The provision of mobile broadband services in many emerging markets is also inherently less competitive, by simple virtue of the fact that fixed-line platforms (the obvious alternative supplier of broadband) are often limited in their geographic reach. And there is arguably some prospect that emerging market players may be
77
abc
growth seen in data traffic. We believe that pricing power is unlikely to be undermined significantly in the medium term, as the competitive environment has remained quite rational in Russia. However, there have been also cases of private reselling of connectivity and the illegal sharing of capacity. While the impact of this could potentially be significant, it should be mitigated as long as data tariffs remain tiered. We would also expect the potential issues with illegal sharing of capacity to gradually fade, as the demand for higher speeds takes hold.
absence of net neutrality regulation should provide significant pricing power and advantage to the incumbents. In Saudi, STC formally launched IPTV services in August 2010 under the InVision brand and is bundling its services through triple-play offerings of voice, broadband and IPTV. This should now leave them well-placed to monetise the fixedline applications layer.
Mobile applications
By contrast with their developed peers, emerging market operators should enjoy several advantages when it comes to leveraging the application layer. For example, they will benefit from a degree of familiarity with their local culture and language that global technology brands will find it difficult to replicate. They should also have less cause for concern over one application in particular that will likely cause headaches for their developed-world peers: VoIP. This service is blocked in many emerging markets and, not infrequently, with the governments approval. Hence, we believe that Africa and MENA should display good growth in the medium term. We expect low-cost smartphones based on the Android platform to drive internet usage in Africa. However, language will present a barrier to would-be developed world competitors. So, for instance, we believe players like MTN should be able to avoid the fate of being dis-intermediated from the applications layer. Some successful non-voice mobile applications are already evident in the emerging markets. For example, mobile payment could become a significant revenue generator (and now provides around 14% of revenues for SafariCom). The fact that most countries in Africa have a large unbanked population creates an appealing opportunity in these countries especially. MTNs mobile payment service, Mobile Money, is showing a good take-up. For instance, MTN Uganda already has almost 16% of its subscriber
There has been limited success in the monetisation of fixed-line data applications by operators in emerging markets. IPTV is one of the most obvious potential services, but is chiefly considered as a bolt-on tool for retaining subscribers and competing at par against cable operators. For example, in Poland, where cable penetration is high, users are reluctant to pay for IPTV unless real premium content is provided something that the incumbent is now making progress in. TPSA signed a 10-year content agreement with TVN group to cross-sell each others services, with a special focus on multiplay bundles. Telecoms operators have had a degree of success with their portals. TPSAs Wirtualna Polska is one of the most frequently visited websites in Poland, but monetisation remains elusive. The hope is that TPSA will be able to venture into adjacent sectors like payment and e-commerce. Meanwhile, competition in terms of content is already pronounced UPC has been providing cable services in CE3 since the late 1990s. OTT is not yet a threat, but inevitably is a looming risk for the future. As for the other CEEMEA countries, especially Russia and MENA, factors like the complexities of rights arrangements, power of existing brands and
78
abc
base using the service. In total, MTN has c2.2m mobile payment subscribers across its various operations. We reiterate our view that, over the medium term, mobile payments will contribute approximately 6% of revenues at a margin of around 20% for the African mobile operators. In Saudi Arabia, services based on news, entertainment and religious content offer good growth prospects. The increase in network coverage and online Arabic content should also drive strong mobile broadband uptake over comin years, in our view. Data volumes conveyed over Mobilys network have increased by 70% yearon-year, reaching 85 terabytes per day. Mobily remains the market leader in the mobile broadband segment, with 60% market share, and has been able to command a pricing premium, reflected in its high data ARPU of cUSD55. Pricing has remained stable in Saudi Arabia for broadband services, unlike other more competitive markets, such as Egypt. Turning to CE3, we expect low-cost smartphones based on the Android platform to drive mobile data usage. However, language will likely continue to present a modest barrier to would-be developed world competitors eyeing these markets. Hence operators like TPSA may not yet be completely dis-intermediated from the mobile applications layer.
stem from a slight increase in Brazil mobile ARPU after adjusting for Q4 2010 equipment sales. We summarise the main changes in the table below.
America Movil: Change in estimates (USDm) 2011e 2012e 2013e
Revenue Previous Difference % EBITDA Previous Difference % Net profit Previous Difference %
Note: All figures on an adjusted basis. Source: HSBC estimates
52,893 52,617 0.5% 21,844 21,755 0.4% 9,171 9,124 0.5%
54,769 54,346 0.8% 22,948 22,805 0.6% 9,793 9,710 0.9%
56,224 55,698 0.9% 23,928 23,744 0.8% 10,415 10,301 1.1%
We value America Movil using a DCF approach employing a WACC of 8.2% with a risk-free rate of 3.5%, equity risk premium of 5.0%, country risk premium of 1.0%, beta of 1.0, and debt-tototal capital ratio of 30%. Our AMX.N ADR target price is USD64. Under our research model, for stocks without a volatility indicator, the Neutral band is 5ppts above and below our hurdle rate for Mexican stocks of 9.5%, or 4.5-14.5% around the current share price. Our target price of USD64 per ADR implies a potential return of 12.7% which is within the Neutral band; thus, we maintain our Neutral rating. Key upside risks to our Neutral ratings, in our view, include synergy delivery above our expectations and greater success of converged fixed-mobile service offerings than we anticipate. Key downside risks, we believe, include merger synergies falling below those of the market, and deteriorating trends at Telmex, which represents around one-fifth of group sales.
Ratings, target price & risks

Below we summarise our target prices, valuation methodologies and highlight the main risks for each of the incumbents analysed in this report. Note that our ratings on these operators reflect our work on numerous themes, and not only the issues on which the present report focuses on.
America Movil: Maintain Neutral rating and USD64 target price

We have tweaked our estimates following the Q4 2010 results conference call. Most of the changes
AT&T: Maintain Neutral rating and USD30 target price

We reduce our forecasts for AT&T following the loss of iPhone exclusivity and its fourth quarter
79
abc
results. We summarise the main changes in the table below.

AT&T: Change in estimates (USDm) 2011e 2012e 2013e
Belgacom: Maintain Neutral rating and EUR30 target price

We have a Neutral rating on Belgacom with a DCF-based target price of EUR30. In our DCFbased valuation, we use a risk-free rate of 3.5% (from 4.0% earlier), a market risk premium of 5.0% (from 4.5% earlier) and a beta of 1.1 to derive our target price. Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below our hurdle rate for Europe-ex-UK stocks of 8.5%, or 3.5-13.5% above the current share price. Our 12-month target price implies a potential return of 11%, which is within the Neutral band and hence we reiterate our Neutral rating. The key upside risk, in our view, is rapid macroeconomic recovery leading to higher-thanexpected revenue and EBITDA growth; the key downside risk is a price war in mobile triggered by the introduction of handset subsidies, leading to a higher-than-expected deterioration in margins.
124,847 127,041 -1.7% 43,676 45,164 -3.3% 13,983 14,843 -5.8%
126,302 129,076 -2.1% 44,817 46,151 -2.9% 14,987 15,812 -5.2%
128,943 132,243 -2.5% 46,297 47,643 -2.8% 15,654 16,575 -5.6%
We value AT&T using a DCF approach employing a WACC of 6.3% (from 6.7% previously) with a risk-free rate of 3.5% (4% previously), equity risk premium of 3.5%, beta of 1.1, and debt-to-total capital ratio of 30%. The impact of the decrease in our forecasts on our DCF-based valuation is broadly offset by a reduction in the WACC, and hence we maintain our target price for AT&T at USD30. Under our research model, for stocks without a volatility indicator, the Neutral band is 5ppts above and below our hurdle rate for US stocks of 7.0% or 2.0-12.0% around the current share price. Our USD30 target price implies a potential return of 5.4%, which is within the Neutral band; thus, we reiterate our Neutral rating on AT&T shares. Key downside risks, in our view, include continued weakness in wireline revenues, and AT&T taking longer than anticipated or being unable to expand its wireless service margins from current levels in absence of iPhone exclusivity. Key upside risks, we believe, include continued high share of net additions despite Verizons iPhone launch and wireless ARPU growth.
BT: Maintain Neutral rating, raise target price to 200p (from 153p)
We lift our target price to 200p (from 153p), driven partly by the changes in our estimates post Q3 results and partly by the reduction in our contingent liability assumption for pension deficit an intrinsically volatile number to GBP3.0bn (from GBP5.7bn earlier) to reflect the changes in the pension indexation to CPI from RPI. The accompanying table summarises the changes to our forecasts. Our WACC assumption of 8.0% is unchanged (cost of equity of 8.7%, cost of debt of 8.4% and a debt-to-capital ratio of 25%).
80
abc
BT: Change in estimates GBPm FY 2011e FY 2012e FY 2013e
5.5% (based on long-term bond yield) and equity risk premium of 5.5% and a beta of 0.8. Downside risks, in our view, include negative regulation that will support alternative operators but not provide similar relief to Bezeq; launch of a new MNO with regulatory protection (ie sub pricing for national roaming); IEC moving to install fibre network in selected regions within Israel, offering third parties an alternative to Bezeqs transmission services.
Revenue Old New % change EBITDA reported Old New % change EPS reported (p) Old New % change
Source: HSBC estimates
20,284 20,173 -0.5% 5,433 5,620 3.4% 14.7 18.2 23.9%
20,090 20,067 -0.1% 5,645 5,813 3.0% 17.9 20.6 15.2%
20,112 20,083 -0.1% 5,791 5,974 3.2% 20.2 23.1 14.2%
Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below our hurdle rate for UK stocks of 7.5%, or 2.5-12.5% around the current share price. Our target price of 200p implies a potential return of 7.9%, which is within the Neutral band; therefore, we reiterate our Neutral rating on the stock. A primary risk both upside and downside to our Neutral stance relates to the triennial pension fund review. BT has agreed to payments of GBP525m into the pension fund for the next three years and committed to further payments increasing at 3% CAGR for the next 17 years. However, the Pensions Regulator has expressed its disquiet at this approach, and is presumably concerned that this time frame is simply overlong. Additional risks to the upside include the potential for BT to beat expectations on its cost savings. Additional risks to the downside include any fresh problems emerging within Global Services as a result of the UK governments austerity measures.
DT: Maintain Underweight rating and EUR9.5 target price

DT: Change in estimates EURm Dec-10 Dec-11 Dec-12
New Revenue HSBC EBITDA company EBITDA HSBC EPS HSBC (EUR) Old Revenue HSBC EBITDA company EBITDA HSBC EPS HSBC (EUR) Change Revenue HSBC EBITDA company EBITDA HSBC EPS HSBC
62,317 19,555 18,550 0.69 62,677 19,892 18,887 0.73 -0.6% -1.7% -1.8% -5.5%
60,787 19,059 18,054 0.68 60,747 19,429 18,424 0.73 0.1% -1.9% -2.0% -7.0%
60,630 19,098 18,093 0.75 60,241 19,388 18,383 0.79 0.6% -1.5% -1.6% -5.8%
Bezeq: Maintain Overweight rating and target price of ILS11.6

We maintain our Overweight rating on Bezeq with a target price of ILS 11.6 per share. Our DCF valuation and target price are based on a cost of equity of 11%, incorporating a risk-free rate of
We use a DCF methodology to arrive at our oneyear forward target price of EUR9.5. We have assumed a risk-free rate of 3.5%, an equity-risk premium of 5.0%, a beta of 1.1 and terminal growth rate of -1%). We have adjusted our numbers to reflect the Q4 currency movements and rolled forward the valuation to year-end 2011e, leading overall to an unchanged target price of EUR9.5. Our target price implies a potential return of -4.4%, which falls below the HSBC Neutral band of 3.5% to 13.5% for European non-volatile stocks. Thus, we reiterate our Underweight rating on the stock.
81
abc
Key upside risks include a quick and sustainable turnaround of T-Mobile US, a strong USD and macro recovery in Eastern Europe.
eAccess: Maintain Overweight (V) rating, cut target price to JPY79,000 (from JPY88,000)
We continue to apply a multiple of 9x to March 2012e EPS to derive our target price for eAccess, based on average trading history. Applied to our new March 2012 EPS estimate of JPY8,732, this yields a new target price of JPY79,000, down 10% from JPY88,000 previously. Under our research model, the Neutral band for volatile Japanese stocks is defined as the hurdle rate of 7% for Japan, plus or minus 10ppt, which translates into -3% to 17% relative to the current share price. Our new target price of JPY79,000 for eAccess shares implies a potential return of 54.9% (including prospective dividend yield of 1.2%). We therefore maintain our Overweight (V) rating on eAccess stock. The principal downside risks to our rating are as follows: Lower profitability than expected in the wireless broadband business, driven by an increase in SACs (we believe the companys current SAC guidance of JPY30,000 is conservative) Greater-than-expected churn in the ADSL business, possibly caused by increased discounting by fibre broadband suppliers Increased competition in the wireless broadband business from UQ Communications, in particular causing erosion of margins from cheaper pricing
Forecast changes
This is a result of eMobile handsets and the network upgrade to 42Mbps being delayed we expect strong sales in the final quarter, and in 2011 generally. Our estimates vs consensus and our previous forecasts are outlined in the table below.
eAccess: Change to estimates JPYbn Mar-11 Mar-12 Mar-13
Consensus Sales 189.8 EBITDA 57.4 EBIT 24.8 Net Income 9.5 EPS 2,949 HSBC new estimates Sales 183.5 EBITDA 57.4 EBIT 24.1 Net Income 8.5 EPS 2,856 HSBC old estimates Sales 186.1 EBITDA 58.5 EBIT 26.9 Net Income 11.1 EPS 3,756 HSBC vs Consensus Sales -3.3% EBITDA -0.1% EBIT -2.7% Net Income -10.6% HSBC vs previous estimates Sales -1.4% EBITDA -1.9% EBIT -10.4% Net Income -23.9% EPS -23.9%
228.4 76.8 36.7 20.7 5,880 219.0 83.2 45.4 30.2 8,732 228.6 83.7 45.7 33.8 9,757 -4.1% 8.3% 23.7% 46.0% -4.2% -0.6% -0.6% -10.5% -10.5%
252.0 88.4 46.8 27.4 8,227 253.5 91.2 55.9 41.0 11,846 252.0 99.7 62.4 51.0 14,743 0.6% 3.2% 19.4% 49.6% 0.6% -8.5% -10.4% -19.7% -19.7%
Ericsson: Maintain Overweight rating and SEK95 target price

We value Ericsson using a DCF approach. We employ a WACC of 9.1% based on a risk-free rate of 3.5%, equity risk premium of 5.5%, beta of 1.25 and debt to capital ratio of 20%. Our target price is SEK95, implying a 15.7% potential return based on the SEK82.1 closing price of 11 February; we have an Overweight rating on the stock. Ericssons current PE is 12.5x 2011 (HSBCe), which compares with a historical range (2005-10) of 8.5-17.5x; our target price of SEK95 implies a PE of 14.5x.
We reduce our FY March 2011 estimates substantially, reflecting lower profitability in the quarters ending September and December 2010.
82
abc
Downside risks to our Overweight rating in the shorter term include continued softness in network orders as these tend to lag the economic cycle. Price pressure from Chinese rivals Huawei and ZTE and a trend toward network sharing among mobile operators may both adversely affect Ericssons revenues.
Key downside risks in our view include: Higher competition in the French market, although Iliads entry into the mobile market has been impacting the share price already, in our view A significant M&A transaction would probably weigh on FT shares as investors would fear dividend cuts and execution risks, although this is not our view and the company has not stated that it has any such plans Execution risks in the UK during the integration process
France Telecom: Maintain Overweight rating and EUR21 target price

We have cut 2010 net profit forecasts by 16% mainly due to the reclassification to net income of the Orange JV in the UK. Other changes in our estimates are immaterial.
France Telecom: Change in estimates EURm 2010e 2011e 2012e
KPN: Maintain Overweight rating and target price of EUR15

After the Q4 2010 results, we have fine-tuned our forecasts, cutting revenue forecasts by 2% to 3% and EBITDA forecasts by 1% to 2% in the medium term (see table below). We have used a DCF valuation to derive our 12month forward target price of EUR15, which implies a potential return of c27%. We have assumed a risk-free rate of 3.5%, a market risk premium of 5.0%, and asset beta of 1.1. Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate for Europe ex-UK stocks of 8.5% (3.5-13.5%). We reiterate our Overweight rating.
KPN: Change in estimates EURm Revenues Old New % change EBITDA Old New % change
Revenue New Old difference % EBITDA New Old difference % Net profit New Old difference %
45341 44839 1.1% 15608 15524 0.5% 4604 5478 -16.0%
45439 44920 1.2% 15504 15682 -1.1% 4725 4696 0.6%
45165 44585 1.3% 15460 15392 0.4% 4842 4693 3.2%
We value France Telecom using a DCF-based sumof-the-parts (SOTP) methodology to arrive at our target price of EUR21. For our SOTP DCF, we assume a discount rate of 8.5% for France, the UK, Spain and the Enterprise business, 11% for Poland and 9.2% for the Rest of the World division. Under our research model, for French stocks without a volatility indicator, the Neutral band is five percentage points above and below our hurdle rate of 8.5%, or 3.5-13.5% above the current share price. Our France Telecom target price of EUR21 implies a potential return of 29.3%, which is above the Neutral band; therefore, we reiterate our Overweight rating on the stock.
2011e
2012e
2013e
13,633 13,423 -2% 5,460 5,472 0%
13,938 13,579 -3% 5,561 5,519 -1%
14,254 13,782 -3% 5,669 5,557 -2%
83
abc
Key risks to our Overweight rating include: E-Plus delivering lower EBITDA margin than we expect or losing market share to competition More aggressive price-based competition by Dutch cable that could hit KPNs broadband recovery
Saudi Arabia, equity beta of 1.0) and cost of debt of 3.3% pre tax, using a debt-to-asset ratio of 30% We have revised our earnings estimates upwards by 11.9% in 2011 and 11.6% in 2012 to incorporate strong pricing power for mobile operators in Saudi Arabia in the lucrative broadband business. Our base case estimates suggest data as a percentage of total revenues would go up from 18% in 2010 to around 29% by 2012, driven by strong subscriber growth along with pricing power for telecom companies in Saudi Arabia. Our capex estimates also go up to incorporate increased network expenses on account of high data usage. Based on our revised WACC assumptions and changes in estimates, our target price rises to SAR71 from SAR69. Under our research model, for Saudi Arabia stocks without a volatility indicator, the Neutral band is 5ppt above/below the hurdle rate of 9.5%. Our new target price implies a potential return of 29.7% from the current price. As this is above the Neutral band for non-volatile Saudi stocks of 4.5-14.5% under the HSBC research model, we maintain our Overweight rating on the stock. The key downside risks to our rating include: (1) an aggressive price war with Zain KSA that erodes the profitability of mobile operators in Saudi Arabia; (2) STCs aggressive build-up of fixed network negatively affecting mobile broadband growth in the Kingdom; and (3) a deterioration in the broader Saudi market with an adverse impact on the stock price, as Mobily is a growth story.
KT Corp: Maintain Overweight rating and KRW60,000 target price

We maintain our Overweight rating on KT Corp with a PE-based one-year forward target price of KRW60,000. We continue to apply a multiple of 8x forward earnings, based on average trading history. Under our research model, for Korean stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate of 10% (ie 5-15%). Applying a multiple of 8x to our 2011 EPS estimate of KRW7,441 yields an unchanged target price of KRW60,000. This implies a 52% potential return including the 6.3% dividend yield in 2011e. We therefore maintain our Overweight rating on KT stock. The key downside risks include: Action by the KCC that mandates the provision of free voice minutes in smartphone tariffs More competition than anticipated in wireline from LG U+ and SK Broadband and in the wireless business from SKT
Mobily (Etihad Etisalat): Maintain Overweight rating, raise target price to SAR71 from SAR69
We continue to use a three-stage DCF model to value Mobily. We have adjusted our WACC to 10.1% compared with 10% earlier. Our revised WACC is a function of cost of equity of 13% (risk-free rate of 3.5%, market risk premium of 9.5%, including inflation differential of 3.5% in
84
abc
Mobily: Change to estimates SARm Old estimates Revenues EBITDA EBITDA margin Capex Net Profit New estimates Revenues EBITDA EBITDA margin Capex Net Profit Change Revenues EBITDA EBITDA Margin (bp) Capex Net Profit
2011e
2012e
2013e
17,021 6,595 38.7% -3,278 4,262 17,930 7,148 39.9% -3,454 4,768 5.3% 8.4% 112 5.4% 11.9%
17,896 6,851 38.3% -3,221 4,402 18,645 7,420 39.8% -3,580 4,912 4.2% 8.3% 151 11.1% 11.6%
18,377 7,071 38.5% -3,014 4,544 19,116 7,599 39.7% -3,319 4,986 4.0% 7.5% 127 10.1% 9.7%
The entry of new operators in many markets could potentially threaten its ARPU and margins in these markets, which can significantly affect its valuation. Fluctuations in ZAR and local currencies against USD and relative movements against each other could have a significant impact on the valuation.
MTS: Maintain Overweight rating and target price of USD29

Our DCF valuation and target price are based on an unchanged cost of equity of 11.5%. We use the one-year average US 10-year treasury bond yield (currently 3.5%) as the risk-free rate (we previously used 6.0%). We arrive at a market risk premium of 8% (was 5.5%), calculated using the relative USD volatility of the local equity market to the US equity market (time-weighted average over the past 10 years). We use a beta of 1.0 (unchanged). Currently we assume no goodwill write-off and no synergies from the Comstar merger in our estimates. Under our research model, for stocks without a volatility indicator, the Neutral band is 5% above and below the hurdle rate for Russian stocks of 11.5%. This translates into a Neutral band of 6.516.5% around the current share price. Our target price implies a potential return of c47%, which is above the Neutral band; hence, we reiterate our Overweight rating. Downside risks, in our view, are execution risks with potential difficulties in the integration of Comstar with MTS and an unsuccessful bid for a 3G licence in Ukraine. RUB fluctuation versus the USD continues to have an impact on the ADR price and is, in our view, a significant risk. A collapse in crude oil prices would have a negative impact on the Russian currency and unemployment level as well as private consumption demand.
MTN: Maintain Overweight rating and target price of ZAR148

We value MTN on a SOTP basis using countryspecific cost of capitals ranging from 14.2% to 20.2%. Our SOTP valuation and target price are based on a cost of equity of 14.7%, incorporating a risk-free rate of 9.2%, equity risk premium of 5.5% and a beta of 1.0. Under our research model, for South African stocks without a volatility indicator, the Neutral band is 5ppt above/below the hurdle rate of 10%. This translates into a Neutral band of 5-15% around the current share price. Our target price implies a potential return of c18%, which is above the Neutral band; hence, we reiterate our Overweight rating. M&A activity is a significant risk to MTNs valuation, in our opinion. Other key potential downside risks include political and economic instability in MTNs areas of operation, particularly Iran, Syria and Sudan, and operational and regulatory risks across its operations. A recession in global commodity prices could weaken demand from the economies where it operates, most of which are commodity-driven.
85
abc
NTT DoCoMo: Maintain our Overweight rating, raise target price to JPY172,000 (from JPY165,000)
Our primary valuation methodology for NTT DoCoMo is PE-based, as we believe this better reflects the way investors view the stock relative to domestic peers and sectors. We continue to apply a multiple of 13x to our March 2012 EPS estimate of JPY13,197. This yields a new target price of JPY172,000, up 4.2% from JPY165,000 previously. Under our research model, the Neutral band for non-volatile stocks is 5ppt above and below the hurdle rate of 7% for Japan. This translates into 2-12% around the current share price. Given the potential return (including prospective dividend yield) of 17% implied by our JPY172,000 target price, we maintain our Overweight rating. The primary downside risks to our rating are: Greater than expected tariff competition in mobile data: We expect DoCoMo to match SoftBanks smartphone plan at cJPY4,000. Prices lower than this would be a negative for the sector. As we explain in this report, we base our revenue forecasts on this assumption, but recognise that announcement of a price cut could be negative for sentiment. Faster than expected voice revenue declines: We assume that DoCoMos voice ARPU decline bottoms out in 2-3 years as a result of migration to the Value Plan concluding. Higher than expected investment in India: We expect further investment in the Tata-DoCoMo joint venture in India, to fund 3G network deployment, but believe this should be both relatively small and expected by investors. We increase our revenue and profit estimates for NTT DoCoMo after faster than expected smartphone growth in the quarter ending December 2010. Changes to our estimates, and our forecasts vs consensus, are outlined in the table below.
NTT DoCoMo: Change to estimates JPYbn Mar-11 Mar-12 Mar-13
Consensus Sales 4,253 EBITDA 1,559 OP 844 Net Income 499 EPS 11,940 HSBC Sales 4,251 EBITDA 1,616 OP 885 Net Income 518 EPS 12,460 HSBC new estimates Sales 4,197 EBITDA 1,561 OP 842 Net Income 493 EPS 11,849 HSBC vs Consensus Sales 0.0% EBITDA 3.7% OP 4.8% Net Income 3.9% HSBC vs previous estimates Sales 1.3% EBITDA 3.5% OP 5.1% Net Income 5.2% EPS 5.2%
4,329 1,569 864 514 12,348 4,356 1,653 937 548 13,208 4,296 1,593 899 526 12,683 0.6% 5.4% 8.5% 6.7% 1.4% 3.8% 4.3% 4.3% 4.1%
4,395 1,598 902 538 12,990 4,597 1,731 1,024 598 14,391 4,451 1,602 917 535 12,892 4.6% 8.3% 13.5% 11.1% 3.3% 8.0% 11.7% 11.8% 11.6%
Portugal Telecom: Maintain Neutral rating and EUR9 target price

We value PTs domestic business on a DCF basis (risk-free rate 3.5%, market risk premium 5.0% and asset beta of 1.1). We value PTs stake in the Oi Group/Telemar by applying a 25% discount (to take into account the complex governance) to the fair market value of Telemar. Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate for Europe ex-UK stocks of 8.5% (ie 3.5-13.5%). Our 12-month target price of EUR9 implies a potential return of 5%, which is within the Neutral band; hence, we maintain our Neutral rating. Key risks to our Neutral rating include. The key upside risk, in our view, is if PT were to acquire a stake of more than 22.38% in the Oi Group for the same consideration of BRL8.4bn .This could occur if the existing
86
abc
shareholders of TNL and TMAR do not fully subscribe to the rights issues The key downside risk, in our view, is if the macroeconomic condition were to become tougher, affecting the domestic business more than expected
Saudi Telecom Company (STC): Maintain Overweight rating, cut target price to SAR48 (from SAR49)
We continue to use a three-stage DCF model to value STC. We now use a WACC of 10.7% compared with 10.5% earlier. We use the longterm EBITDA contribution of 60% from domestic operations to calculate the weighted average cost of equity for STC. International COE is taken to be 15%. For domestic operations, we use a risk free rate of 3.5%, market risk premium of 9.5% including inflation differential of 3.5% in Saudi Arabia and equity beta of 1.0. Our weighted cost of equity for STC is 13.8%. Using a long-term debt-to-equity ratio of 30%:70%, we obtain a WACC of 10.7%. Cost of debt is unchanged at 3.8% for the group.
We have revised our earnings estimates upwards by 5% in 2011 and 8.2% in 2012 to incorporate strong pricing power for mobile operators in Saudi Arabia in the lucrative broadband business. Our capex estimates also go up by 13.9% in 2011 and 16.7% in 2012 to incorporate increased network expenses on account of high data usage as the migration to NGN continues for STC. Based on our revised WACC assumptions along with changes in estimates, our target price falls to SAR48 from SAR49. Under our research model, for Saudi Arabia stocks without a volatility indicator, the Neutral band is 5ppt above/below the hurdle rate of 9.5%. Our new target price implies a potential return of 19% from the current price. As this is above the Neutral band for nonvolatile Saudi stocks of 4.5-14.5% under the HSBC research model, we maintain our Overweight rating on the stock. The key downside risks to our rating include: 1) increased FX losses from international operations; 2) Zain KSA getting into a price war in Saudi, leading to erosion of ARPUs for all mobile operators; 3) overpayment for international acquisitions; 4) delay in NGN implementation, and 5) higher-than-expected capex in international operations such as those in India and Indonesia.
STC: Change in estimates SARm Old estimates Total Revenue EBITDA EBITDA margin Capex Net Profit New Estimates Total Revenue EBITDA EBITDA margin Capex Net Profit Variation Total Revenue EBITDA EBITDA margin (bp) Capex Net Profit
2011e
2012e
2013e
52,026 19,872 38.2% -9,478 8,723 54,477 20,631 37.9% -10,791 9,157 4.7% 3.8% -33 13.9% 5.0%
52,582 20,099 38.2% -8,607 8,767 54,975 21,104 38.4% -10,043 9,484 4.6% 5.0% 16 16.7% 8.2%
53,700 20,530 38.2% -8,124 9,154 55,796 21,479 38.5% -9,178 9,626 3.9% 4.6% 26 13.0% 5.1%
Sprint Nextel: Maintain Neutral (V) rating and USD5 target price
We reduce our profit estimates for Sprint Nextel following its fourth quarter results. We summarise the main changes in the table below. We value Sprint Nextel using a DCF approach employing a WACC of 8.1% (down from 8.4% previously) with a risk-free rate of 3.5% (4% previously), equity risk premium of 3.5%, beta of 1.5, and debt-to-total capital ratio of 30%.
87
abc
Sprint Nextel: Change in estimates USDm 2011e 2012e 2013e
Swisscom: Change to estimates CHFm Revenue New Old change EBITDA New Old change Net profit New Old change
2010e
2011e
2012e
32,348 32,250 0.3% 5,917 6,140 -3.6% -1,069 -335 NM
32,822 32,363 1.4% 6,188 6,327 -2.2% -14 572 NM
33,553 32,725 2.5% 6,367 6,482 -1.8% 492 1,180 -58.3%
12,008 11,994 0% 4,696 4,692 0% 1,898 1,897 0%
11,951 11,987 0% 4,741 4,771 -1% 2,018 2,044 -1%
12,243 12,254 0% 4,840 4,858 0% 2,093 2,116 -1%
The impact of a decrease in our profit forecasts on our DCF valuation of Sprint Nextel is broadly offset by a reduction in the WACC. Accordingly, we maintain our target price for Sprint Nextel at USD5. Under our research model, for stocks with a volatility indicator, the Neutral band is 10ppts above and below our hurdle rate for US stocks of7.0% or -3% to 17.0% around the current share price. Our target price of USD5 indicates a potential return of 8.7%, which is within the Neutral band; thus, we reiterate our Neutral (V) rating on Sprint Nextel shares. Downside risks, we believe, include operational performance below consensus expectations. Upside risks that we see include potential consolidation the US wireless sub-sector.
The key downside risks to our Overweight rating on Swisscom are: a worsening in competition conditions in Switzerland for both fixed line and mobile owing to lower tariffs in either retail or wholesale (ULL pricing); Swisscom continuing to be conservative with its dividend payout in FY 2010; and a worse-than-expected tax investigation outcome for Fastweb.
TDC: Maintain Overweight rating and DKK58 target price

TDC: Change to estimates FY 2011e New Revenue reported EBITDA reported Operating profit reported Old Revenue reported EBITDA reported Operating profit reported Change Revenue reported EBITDA reported Operating profit reported
FY2012e DKKm 26,200 9,873 4,796
FY2013e DKKm 26,307 9,945 5,000
DKKm 26,143 9,794 4,572
Swisscom: Maintain Overweight rating and CHF470 target price

We have marginally revised our forecasts, although this has no material impact on valuation. We have an Overweight rating on Swisscom with a DCF-based target price of CHF470. In our DCF, we assume a risk-free rate of 3.5% (from 4.0% earlier), a market risk premium of 4.0% (from 4.5% earlier) and beta of 1.1.
26,220 9,925 4,624 -0.3% -1.3% -1.1%
26,393 10,082 4,923 -0.7% -2.1% -2.6%
26,594 10,184 5,152 -1.1% -2.3% -2.9%
We have slightly revised down our estimates to reflect Q4 2010 reported numbers. We use two separate methods to calculate an equity valuation range for TDC: first, a DCF approach; and second, a comparison to peer group EV/EBITDA multiples and dividend yields. Our target is the mid point of both approaches. We believe longterm a pure DCF-based approach is legitimate. However, as TDC just recently returned to the
88
abc
market, it will take time for investors to reacquaint themselves with the stock. During that process, a balanced approach of DCF and peer multiple valuation seems appropriate to us. We calculate a target price of DKK58 as the mid point of our DCF and peer multiple valuations. Our DCF assumes a WACC of 7.2% (RFR 3.5%, MRP 4.5%, beta 1.1) and a (cautious) 0% terminal growth rate. Under our research model, for Danish stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate of 8.0% (3.0-13.0%). Our target price implies a potential return of 25%; we maintain our Overweight rating. As for all incumbents, regulation is a significant source of downside risk for TDC, and there is specific concern that the introduction of a wholesale bitstream broadband product over TDCs cable network could undermine the incumbents access network advantage. The utility companies fibre ambitions also represent a continued threat. Price erosion in traditional products could more than offset growth from newer areas such as quad-play.
Tele2: Change to estimates SEKm New Revenue reported EBITDA reported Operating profit reported Old Revenue reported EBITDA reported Operating profit reported Change Revenue reported EBITDA reported Operating profit reported
FY 2011e
FY2012e
FY2013e
40,769 10,672 7,054 39,774 11,046 7,662 3% -3% -8%
42,043 11,653 7,799 40,246 11,725 7,906 4% -1% -1%
42,700 12,321 8,263 40,687 12,401 8,358 5% -1% -1%
In our current valuation methodology we assume a debt-to-capital ratio of about 25%. Please note our target price of SEK160 includes the extraordinary dividend of SEK27 proposed by Tele2 for FY 2010e; post adjusting for this, our implied target price would be SEK137. The potential return from the current price of SEK148.7 implied by our SEK160 target price is 8%, which is within the 4% to 14% Neutral range for non-volatile Swedish (ex-UK) stocks under HSBCs research model; thus we retain our Neutral rating on Tele2. Better than expected margin performance from the Swedish mobile division, where Tele2 has been focusing on gaining higher share of post paid subscribers could be a potential positive driver for the stock. Key downside risks include a failure to maintain the Russian growth momentum given the lack of 3G spectrum and a collapse of cash generation of its arbitrage businesses, in our view.
Tele2: Maintain Neutral rating, raise target price to SEK160 (from SEK155)
We value Tele2 based on HSBCs DCF methodology (assumptions: risk-free rate 3.5%, market premium 5.5%, debt-to-capital ratio 25% and a beta of 1). The upward revision of our target price to SEK160 from SEK155 reflects a slight increase to our revenue estimates (primarily due to higher growth expectations for Swedish mobile and a small increase to subscriber numbers for Russian mobile) and a reduction to our WACC assumption to 8.3% from 8.6%. On a DCF basis we arrive at one-year forward target price of SEK160.
Telecom Italia: Maintain Underweight rating and EUR1.05 target price

We are consolidating Telecom Argentina from Q4 2010 following the approvals by the Argentine regulators that have resulted in some significant changes in our forecasts. Considering that the competition has worsened in the domestic fixed and domestic mobile continues to underperform the market, we are cutting our domestic forecasts
89
abc
downwards and this is also reflected in the cutting of our EPS forecasts. We have used a DCF valuation to derive our oneyear forward target price. We assume a risk-free rate of 3.5%, a market risk premium of 5.0% and asset beta of 1.1 to arrive at an unchanged EUR1.05 per share. Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate for Europe ex-UK stocks of 8.5% (3.5-13.5%). The potential return implied in our valuation of 0% is below the Neutral band; we therefore reiterate our Underweight rating .
Telecom Italia: Change to estimates EURm Revenues Old New % change EBITDA Old New % change EBIT Old New % change EPS Old New % change
Telefonica: Maintain Overweight rating and target price of EUR22

We maintain our Overweight rating on TEF and our DCF-based one-year forward target price of EUR22. We have assumed a risk-free rate of 3.5%, an equity risk premium of 5.0% and asset beta of 1.1. Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate for Europe ex-UK stocks of 8.5% (ie 3.513.5%). Our 12-month target price of EUR22 implies a potential return of 20%, which is above the Neutral band; hence we maintain our Overweight rating. Key risks to our Overweight rating include: A delayed recovery of the Spanish economy (HSBC economist Madhur Jha forecasts GDP growth of -0.4% in 2010 and +0.6% in 2011) A currency crisis in LatAm that would hurt the growth momentum Telefonica or competitors introducing irrational mobile data pricing in Spain, UK or Germany, limiting the potential for restoring revenue growth in Telefonicas developed markets
2011e
2012e
2013e
26,803 29,695 11% 11,335 12,221 8% 5,934 6,371 7% 0.12 0.11 -4%
26,732 29,927 12% 11,233 12,159 8% 6,132 6,661 9% 0.13 0.13 -1%
26,690 30,067 13% 11,180 12,142 9% 6,367 6,921 9% 0.14 0.14 0%
The key upside risk to our rating is a rebound in the economy, which could lead to lower revenue pressure in Italy. We estimate a 1-3% domestic revenue decline in 2011 and 2012 after a 7% revenue decline in 2010; an improvement in the top line would also be an upside risk to our rating. We are factoring in aggressive MTR cuts beyond 2011 to reach EUR4c in 2012e and EUR2c in 2013e. More benign MTR cuts would represent upside to our rating.
90
abc
Telekom Austria: Maintain Neutral rating, raise target price to EUR11 (from EUR10)
Telekom Austria: Change to estimates EURm New Revenue reported EBITDA Reported Operating profit Old Revenue reported EBITDA Reported Operating profit Change Revenue reported EBITDA Reported Operating profit
Telenor: Maintain Overweight rating, cut target price to NOK111 (from NOK115)
Telenor: Change to estimates NOKm New Revenue reported EBITDA reported Operating profit reported Old Revenue reported EBITDA reported Operating profit reported Change Revenue reported EBITDA reported Operating profit reported
FY 2010e
FY 2011e
FY 2012e
FY 2011e
FY 2012e
FY 2013e
4,618 1,641 573 4,588 1,614 575 1% 2% 0%
4,570 1,586 569 4,540 1,591 563 1% 0% 1%
4,600 1,598 649 4,466 1,543 580 3% 4% 12%
100,544 30,510 14,493 100,957 31,481 16,151 0% -3% -10%
105,789 33,580 17,811 105,905 34,055 19,354 0% -1% -8%
109,174 35,526 20,368 110,151 36,127 21,990 -1% -2% -7%
We value Telekom Austria based on a simple DCF. Our valuation assumes a risk-free rate of 3.5%, market risk premium 5.0% and beta of 1.1. Our target price upgrade to EUR11 from EUR10 reflects rolling forward our valuation to FY 2011e and a slight increase in our estimates. Our EUR11 target price implies a potential return of 4.6%, which is above the 3.5% to 13.5% Neutral range for non-volatile Austrian stocks under HSBCs research model; thus, we maintain our Neutral rating on the stock. Key upside risks include a quicker macro recovery, consolidation of the Austrian mobile market and a share-buyback announcement as originally envisaged in January 2009 (though we do not foresee any announcements in the immediate future). Key downside risks include prolonged revenue decline across the CEE and TKA failing to stabilise the fixed-line business.
We value Telenor based on a sum-of-the parts valuation, in which we ascribe 4.5x EV/EBITDA for Nordic business (Fixed+ Mobile), 5x EV/EBITDA for CEE operations and 6x EV/EBITDA for emerging Asian market subsidiaries (except DiGi, which we now value based on HSBCs target price of MYR29/share; DSOM.KL, MYR25.9, OW). We value Vimpelcom based on our target price for Vimpelcom of USD18 (revised down from USD22, in our last report dated 18 January; VIP.N, USD14.3, N(V)). Our NOK111 target price implies a potential return of 21%, which is above the 5.5% to 15.5% Neutral range for non-volatile Norwegian stocks under HSBCs research model; thus, we maintain our Overweight rating on the stock. Our target price revision to NOK111 (from NOK115) primarily reflects the downward revision to our target price on Vimpelcom, slightly offset by a higher valuation for DiGi, which we now value based on HSBCs target price of MYR29, from 6.0x EV/EBITDA previously, and a slightly increased valuation for the Nordic business driven by margin improvements.
91
abc
Telenor: Sum-of-the-parts valuation (NOKm) Business Stake 2011e Sales 2011e EBITDA Multiple (x) Valuation method and ratio Valuation Val / share % of EV OLD Val/share
Telenor Norway (fixed+mobile) Telenor Denmark (fixed + mobile) Telenor Sweden (fixed + mobile) Nordic Operations Pannon GSM - Hungary ProMonte - Montenegro Telenor Mobile-Serbia Eastern European Operations Telenor Pakistan DTAC - Thailand GrameenPhone DiGi.Com - Malaysia Asian Operations Broadcast Value of core businesses EDB-Ergo Vimpelcom Others Value of associates Indian venture market value Value of minorities Appraised EV Net debt Pension deficit ( 2010a) Contingent liabilities (anticipated licences in Bangladesh and Thailand) Appraised equity value Number of shares Appraised share price
100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 65.5% 55.8% 49.0% 100.0% 27.2% 39.6%
25,816 7,268 9,978 4,565 631 2,678 4,946 14,852 7,027 10,808 9,083 70,003
10,311 1,840 2,534 1,650 275 1,068 1,519 5,173 3,432 4,744 2,322 33,396
4.5 4.5 4.5 5.0 5.0 5.0 6.0 6.0 6.0 9.1 7.2 4.2
EV/EBITDA EV/EBITDA EV/EBITDA EV/EBITDA EV/EBITDA EV/EBITDA EV/EBITDA EV/EBITDA EV/EBITDA HSBC TP of MYR29 EV/EBITDA Market Cap HSBC TP of USD18 Book value Cash burn until breakeven in 2014
46,399 8,279 11,404 66,082 8,249 1,373 5,341 14,963 9,117 31,037 20,590 42,981 103,724 16,720 201,489 765 54,127 844 55,735 -10,279 -42,656 204,289 -19,276 -1,381 -3,000
180,632 1,631 111.0
28 5 7 41 5 1 3 9 6 19 13 26 64 10 124 0 33 1 34 -6 -26 125 -12 -1 -2

111
23% 4% 6% 32% 4% 1% 3% 7% 4% 15% 10% 21% 51% 8% 99% 0% 26% 0% 27% -5% -21% 100%
28 5 7 39 7 1 4 12 6 19 13 17 50 10 117 0 40 1 41 -6 -21 131 -13 -1 -2

115
The key downside risks to our valuation and estimates, in our view, are: 1) higher than expected cash outflow in the Indian operations following the 2G licence debate; 2) failure to maintain competitive shareholder remuneration; and 3) loss of influence in Vimpelcom.
TeliaSonera: Maintain Neutral rating and SEK58 target price

TeliaSonera: Change to estimates SEKm New Revenue reported EBITDA reported Operating profit reported Old Revenue reported EBITDA reported Operating profit reported Change Revenue reported EBITDA reported Operating profit reported
We use HSBCs standard three-stage DCF to value the core business and a peer multiples approach to value the associates and minorities not under our coverage. We assume a risk-free rate of 3.5%, market risk premium of 5.5%, debtto-capital ratio of 25% and beta of 1.1. Our valuation methodology yields an unchanged target price of SEK58. We have slightly revised our estimates to reflect the currency movements. The potential return from the current price of SEK54.5 implied by our SEK58 target price is 6.4%, which is within the 4% to 14% Neutral range for non-volatile Swedish stocks under HSBCs research model; thus, we retain our Neutral rating on TeliaSonera. The key downside risks would be a significant fall in domestic business margins or any obstacles to an agreement with Alfa group to combine their
FY 2011e
FY 2012e
FY 2013e
106,041 37,242 24,644 107,473 37,625 25,443 -1% -1% -3%
109,481 38,409 25,766 108,740 38,604 26,111 1% -1% -1%
110,850 38,941 25,873 109,814 39,120 26,396 1% 0% -2%
92
abc
holdings in Turkcell and MegaFon, and value destructive acquisitions. The main upside risks would be a decision that could lead to higher shareholder returns, stabilisation of the Swedish fixed business and dividends from Megafon (which we believe has limited visibility in the short term).
deterioration in the Australian competitive environment may put this recovery at risk.
TIM Participaes: Maintain Overweight rating and BRL7.50 target price

We value TIM Participaes using a DCF approach, employing a WACC of 10.1% with a risk free rate of 3.5%, equity risk premium of 5.0%, country risk premium of 3.0%, beta of 1.0, and debt-to-total capital ratio of 30%. Our DCF-based target price for TIM common shares (TCSL3.SA) is BRL10.00. To value the preference shares (TCSL4.SA), we apply a 25% discount to our core valuation of TIM common shares based on the historical trading relationship. This results in a target price of BRL7.50 for the preference shares. We attain our TSU.N ADR target price of USD45 by converting our TCSL4.SA target price to USD using HSBCs 2011 average BRL/USD forecast of 1.67 at the time we last updated our target price on 16 November 2010. Under our research model, for stocks without a volatility indicator, the Neutral band is 5ppts above and below our hurdle rate for Brazilian stocks of 11.5%, or 6.5-16.5% around the share price. Since our target prices imply potential returns above the Neutral band (TCSL4.SA 24.6%, TCSL3.SA 37.0%, TSU.N 22.0%), we rate all TIM share classes Overweight. Downside risks, we believe, include execution risks surrounding the companys turnaround plan. Competition in the Brazilian mobile market remains high, and although we believe that it has moderated somewhat, an uptick remains a risk to TIM.
Telstra: Maintain Overweight rating and AUD3.40 target price

Given depressed earnings in FY 2011 and a lack of medium-term visibility, we continue to use a dividend discount methodology (DDM) to value Telstra stock. We apply a cost of equity of 8.5%, a beta of 1.1, and a terminal growth rate of 0%. This yields our unchanged target price of AUD3.40. For stocks without a volatility indicator, the Neutral band is five percentage points above and below the hurdle rate for Australian stocks of 8.5%, or 3.5-13.5% around the current share price. Our target price of AUD3.40 implies a potential return of 16.8%. Therefore we maintain our Overweight rating for Telstra stock. The key downside risks to our rating are NBN risk. With the passage of the June 2010 Heads of Agreement deal not finalised and subject to change, regulatory risk remains. A more hostile attitude to Telstra (including more threats of break-up) would result in a likely rejection of a deal by Telstra shareholders, and further regulatory uncertainty. Operational risk. Our forecasts are in line with consensus for FY June 2011, but substantially ahead of consensus for FY June 2012 based on our view of Telstras sustainable competitive advantage in both its wireline and wireless networks. A further
93
abc
TPSA: Maintain Underweight rating and PLN15.3 target price

TIM Participaes: Ratings and target prices (BRL except TSU ADR) Ticker New rating Previous rating Target Current price price Upside
Turk Telekom: Maintain Underweight rating and TRY6.8 target price

We value Turk Telekom using the average of our SOTP and DCF valuations. Our DCF valuation gives a fair value of TRY6.7, while our SOTP fair value is now TRY6.9. The DCF valuation is based on a cost of equity of 13.5%. We use the one-year average US 10-year treasury bond yield (currently 3.5%) as the riskfree rate. We arrive at a market risk premium of 10%, calculated using the relative USD volatility of the local equity market to the US equity market (time-weighted average over the past 10 years). We use a beta of 1.0. The SOTP uses WACCs of 13.5% for the fixed-line business and 13.2% for the mobile business. Under our research model, for stocks without a volatility indicator, the Neutral band is 5ppt above and below the hurdle rate of 13.5% for Turkey. This translates into a Neutral band of 8.5-18.5% around the current share price. Our 12-month target price implies a negative potential return of 4.5%, including a dividend yield of 9%. As this is below the Neutral band, we maintain our Underweight rating. Main upside risks, in our view, include lowerthan-expected inflation, higher-than-expected dividends, high growth in Turk Telekoms subscriber acquisitions, higher-than-expected mobile ARPU and broadband ARPL, brisk IPTV off-take, and a delay in the entry of competition in the fixed-line mass market.
TSU (USD) TCSL4 TCSL3
OW OW OW
OW OW OW
45.00 7.50 10.00
34.38 5.76 7.05
30.9% 30.2% 41.8%
Source: HSBC estimates. Prices at close 10 January 2011
Our DCF valuation and target price are based on a cost of equity of 11.8%. We use the one-year average US 10-year treasury bond yield (currently 3.5%) as the risk-free rate. We arrive at the market risk premium of 7.5%, calculated using the relative USD volatility of the local equity market to the US equity market (time-weighted average over the past 10 years). We use an unchanged beta of 1.1 to reflect the earnings and price uncertainty arising out of the DPTG dispute. Under our research model, for stocks without a volatility indicator, the Neutral band is 5ppt above and below the hurdle rate of 11% for Poland. This translates into a Neutral band of 6-16% around the current share price. Our 12-month target price implies a potential return of only 1.4%, despite including the dividend yield of c9%. That is below the Neutral band, so we maintain our Underweight rating. A quick resolution of the DPTG claim in favour of TPSA would be a significant upside risk. Other upside risks include rational competition in the mobile segment, less competition than expected from cable operators, and a better-than-expected macro and regulatory environment.
94
abc
Verizon: Maintain Overweight rating, raise target price to USD41 (from USD37)
Vodafone: Change in estimates GBPm 2011e 2012e 2013e
Our new USD41 target price implies a potential return of 12.7%, which is above the Neutral band; thus, we maintain our Overweight rating on Verizon shares. Key downside risks, in our view, include weaker than anticipated sales response to the launch of the Verizon iPhone, longer than anticipated weakness in the wireline business, and moves by Verizon and Vodafone Group to restructure their ownership of their Verizon Wireless partnership.
Sales New Old %change EBITDA New Old %change EPS* (p) New Old %change
45,555 44,790 1.7% 14,659 14,611 0.3% 17.1 16.2 5.6%
46,476 45,232 2.8% 14,825 14,833 -0.1% 17.5 16.4 6.3%
47,648 45,840 3.9% 15,168 15,081 0.6% 18.6 16.7 11.4%
Note: *Company definition of adjusted earnings. Source: HSBC estimates
Vodafone: Maintain Overweight rating, raise target price to 230p (from 190p)
We value Vodafone via a DCF methodology, and have arrived at a new target price of 230p, up from 190p previously benefiting partly from change in our forecasts post the companys Q3 results and partly from the GBP2.8bn share buyback programme (of which Vodafone has already completed GBP1.1bn to 31 December 2010). Our WACC assumption of 8.0% (cost of equity of 9.1%, cost of debt of 6.3% and debt-to-capital ratio of 25%) is unchanged. The accompanying table summarises the changes to our forecasts. Under our research model, for stocks without a volatility indicator, the Neutral band is five percentage points above and below our hurdle rate for UK stocks of 7.5%, or 2.5-12.5% around the current share price. Our Vodafone target price of 230p implies a potential total return of 27.8%, which is above the Neutral band; therefore, we maintain our Overweight rating on the stock. Key downside risks, in our view, include: an intensification of competition and adverse regulatory impact in the already challenging Indian market; the improvements seen in the European operations faltering; or the upcoming spectrum auctions proving more expensive than anticipated.
We increase our forecasts following Verizons fourth quarter results and the firms launch of the iPhone. We summarise the main changes in the table below.
Verizon: Change in estimates (USDm) 2011e 2012e 2013e
110,599 107,446 2.9% 36,697 36,533 0.4% 6,539 6,529 0.2%
118,385 110,593 7.0% 39,640 38,031 4.2% 7,791 7,468 4.3%
123,006 114,394 7.5% 41,235 39,651 4.0% 8,662 8,392 3.2%
We value Verizon using a DCF approach employing a WACC of 6.3% (adjusted down from 6.7% previously) with a risk-free rate of 3.5% (4% previously), equity risk premium of 3.5%, beta of 1.1, and debt-to-total capital ratio of 30%. Reflecting the increase in our forecasts and reduction in the WACC, our target price rises to USD41 from USD37 previously. Under our research model, for stocks without a volatility indicator, the Neutral band is 5ppts above and below our hurdle rate for US stocks of 7.0% or 2.0-12.0% around the current share price.
95
abc
This page has been left intentionally blank
96
abc
Company profiles
97
abc
America Movil (Neutral, TP USD64)

Fixed line connectivity
America Movil (AMX) is now a diversified player having merged its mobile assets with the Mexican incumbent Telmex as well as Telmex Internacionals roster of South American cable and altnet properties. To the benefit of fixed incumbents (although less so altnets), unbundling of the local loop either does not exist or is not actively supported by the regions regulators. Generally speaking, regulators in Latin America often lack the necessary experience, resources or political clout to pursue radical agendas like unbundling or net neutrality. The only exception so far in is Chile where a law upholding net neutrality principles was passed in July 2010.
careful to differentiate between pure smartphone use and tethering (using a smartphone as a laptop modem) which attracts a 20% premium in markets like Mexico for example. In Mexico, America Movils c73% market share and 3G leadership afford it a certain degree of pricing power in data. In Brazil, although voice price competition remains fierce (America Movil has noted the price of a voice minute fell by 20% in 2010), there is greater price discipline in data between the data market leaders, Claro and Telefonicas Vivo.
Fixed line applications

In Mexico, regulatory challenges continue to prevent Telmex from offering pay TV. Meanwhile, in Brazil, America Movils has cable (NET Servios, Brazils leading cable player) as well as rapidly growing satellite operations (Embratels satellite TV customer base quintupled in 2010). It is widely expected that Brazils media law which currently prohibits telcos from offering IPTV will be changed this year; the regulator Anatel has already started removing this restriction from telco licence renewals as it seeks to foster more payTV competition.
Mobile connectivity
Smartphone penetration is low in Latin America (excluding QWERTY devices like BlackBerrys, we estimate penetration of true smartphones across the region is in the low single digits). America Movil however has a strong focus on data which is paying dividends with Q4 2010 group mobile data revenues growing c37% year-on-year in constant currency terms. Mobile data now represents typically c2025% of service revenues in most countries. We expect continued steep price erosion of Androidbased smartphones, primarily from Asian manufacturers to continue expanding the addressable market for mobile data in Latin America. Towards the end of Q4 2009, America Movils Brazilian unit, Claro, experienced some isolated pressure from data users on its 3G network before it was comprehensively built out. Having watched the difficulties created by unlimited mobile data plans in developed markets, operators in the region have adopted sensible tariff structures: tiered pricing, data caps with overage or speed throttling are normal practises in the region. Additionally, America Movil, and most of its competitors are
Mobile applications
Although operators in emerging markets with local knowledge and powerful domestic content allies (America Movil joint-owns Net Servios with powerful Brazilian media group Globo), we still believe it will be difficult for mobile operators to capture a significant portion of the applications value chain (ie above any revenue share under the Android Market model).
Investment thesis
We rate America Movil Neutral. Although wellpositioned to benefit from mobile capacity scarcity, we expect consolidation of Mexican fixed-line (Telmex) to drag significantly on group growth. In fixed line, the scarcity opportunity is not being fully exploited with little NGA investment and little or no unbundling to flush out anyway.
98
abc
Financials & valuation: America Movil

Financial statements Year to 12/2010a 12/2011e 12/2012e 12/2013e Key forecast drivers Year to 12/2010a 12/2011e
Neutral
12/2012e 12/2013e
Profit & loss summary (USDm)
Revenue EBITDA Depreciation & amortisation Operating profit/EBIT Net interest PBT HSBC PBT Taxation Net profit HSBC net profit
Cash flow summary (USDm)
48,130 19,553 -7,216 12,337 -989 10,982 12,200 -3,179 7,195 7,853
52,893 21,844 -7,286 14,558 -1,077 13,823 13,823 -4,147 9,171 9,171
54,769 22,948 -7,438 15,510 -1,026 14,740 14,740 -4,422 9,793 9,793
56,224 23,928 -7,622 16,306 -973 15,619 15,619 -4,686 10,415 10,415
Mexico Revenues Brazil Revenues Group Revenues Group EBITDA Group mobile subscribers (000)
19,865 11,939 48,130 19,553 225,025
21,489 13,172 52,893 21,844 239,801
21,760 13,461 54,769 22,948 251,741
22,114 13,499 56,224 23,928 261,340
Valuation data Year to 12/2010a 12/2011e 12/2012e 12/2013e
Cash flow from operations Capex Cash flow from investment Dividends Change in net debt FCF equity
14,832 -6,508 -11,757 -1,034 100 8,747
17,216 -8,573 -8,573 -1,275 -5,092 8,003
17,633 -8,793 -8,793 -1,431 -4,925 8,565
18,413 -8,683 -8,683 -1,609 -5,645 9,467
EV/sales EV/EBITDA EV/IC PE* P/Book value FCF yield (%) Dividend yield (%)
2.6 6.5 3.3 14.7 4.3 8.0 0.9
2.3 5.6 3.2 12.4 3.5 7.3 1.1
2.1 5.1 2.9 11.3 2.9 7.8 1.3
2.0 4.6 2.7 10.4 2.4 8.6 1.5
Note: * = Based on HSBC EPS (fully diluted)
Balance sheet summary (USDm)
Intangible fixed assets Tangible fixed assets Current assets Cash & others Total assets Operating liabilities Gross debt Net debt Shareholders funds Invested capital
9,150 33,319 20,595 9,229 72,188 15,976 25,981 16,752 26,787 37,860
9,150 34,606 16,379 4,863 69,602 16,721 16,522 11,659 32,599 38,552
9,150 35,961 20,484 8,687 75,317 17,135 15,421 6,734 38,477 39,774
9,150 37,022 23,356 11,341 79,537 17,497 12,430 1,089 44,807 40,691
Issuer information
Share price (USD) Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
56.78 Target price (USD) AMX.N 114,501 59 Mexico Richard Dineen
64.00 Potent'l return (%) 12.7
Bloomberg (Equity) AMX US Market cap (USDm) 114,501 Enterprise value (USDm) 121,468 Sector Wireless Telecoms Contact 1 212 525 6707
Price relative Ratio, growth and per share analysis Year to Y-o-y % change 12/2010a 12/2011e 12/2012e 12/2013e
64 59 54 49 44 39 34 29 24 19 2009
America Movil
Revenue EBITDA Operating profit PBT HSBC EPS

Ratios (%)
14.3 13.5 9.9 1.5 -3.9
9.9 11.7 18.0 25.9 18.7
3.5 5.1 6.5 6.6 9.2
2.7 4.3 5.1 6.0 8.8
64 59 54 49 44 39 34 29 24 19 2010
Rel to MEXICAN I.P.C. IDX
2011
2012
Revenue/IC (x) ROIC ROE ROA EBITDA margin Operating profit margin EBITDA/net interest (x) Net debt/equity Net debt/EBITDA (x) CF from operations/net debt
Per share data (USD)
1.3 23.5 31.1 12.9 40.6 25.6 19.8 62.5 0.9 88.5
1.4 26.7 30.9 14.9 41.3 27.5 20.3 35.2 0.5 147.7
1.4 27.7 27.6 15.5 41.9 28.3 22.4 17.0 0.3 261.8
1.4 28.4 25.0 15.2 42.6 29.0 24.6 2.3 0.0 1690.5
Source: HSBC
Note: price at close of 11 Feb 2011
EPS Rep (fully diluted) HSBC EPS (fully diluted) DPS Book value
3.34 3.87 0.51 13.34
4.59 4.59 0.64 16.33
5.01 5.01 0.73 19.70
5.46 5.46 0.84 23.47
99
abc
AT&T (Neutral, TP USD30)

The FCCs 2003 Triennial Review effectively withdrew support for unbundling (known in the US as UNE-P, Unbundled Network Elements Platform) of incumbent local access fibre. The shift in regulation encouraged incumbents like AT&T and Verizon to then embark on extensive fiber investments. Unbundled access providers in fibred areas of the US have since all but disappeared. While unbundling regulation has favored incumbents, the FCC has taken a hard line on net neutrality. This is understandable given the strategic importance of US Internet companies (whose products and services may be exported, unlike those of domestic-focused telcos). In a December 2010 decision, the FCC passed a new policy that maintains strict net neutrality rules for fixed-line networks while acknowledging the unique scarcity of wireless capacity by permitting some shaping of mobile traffic by operators based on reasonable network management criteria.
AT&T was among the first operators globally to shift from flat-rate unlimited plans to tiered data bundles either USD15 for 200MB or USD25 for 2GB (overage charges apply above these limits). Although other US operators are for the moment sticking with unlimited plans, we believe there is longer-term consensus that tiered plans represent the future. AT&T and Verizon are the pricesetters in the US market, having several advantages that their rivals lack (iPhone carriage and broader and/or higher quality network coverage, larger marketing budgets and so forth). Given the pressure on its network it is unsurprising that AT&T has been among the first to exploit public WiFi to offload traffic where possible. AT&T has 23,000 hotspots in the US, including expanded hotzones in places like New York Citys Times Square. The firm is also offering femtocell products under the Microcell brand which retail for cUSD150.

AT&Ts U-verse IPTV service had nearly 3m customers by end-2010, penetration of around 14% in areas available for sale. As with Verizon, AT&T competes directly against cable and selectively against satellite (AT&T also re-sells DirecTV satellite payTV outside U-verse markets).
Mobile connectivity
The US mobile data market is highly developed. Over 60% of AT&Ts postpaid base has an integrated device (including QWERTY devices such as BlackBerrys); we estimate 35-40% of its base has a true smartphone. Following the rapid adoption of smartphones due mainly to the firms exclusive carriage of the Apple iPhone June 2007February 2011, AT&Ts mobile network came under intense pressure with widely publicized network problems in San Francisco and New York from mid-2009. This prompted a substantial increase in AT&Ts wireless capex with 2010 spend increasing over 50% versus the prior year. A major focus was the addition of a further c2,000 3G base stations during 2010 as well as boosting backhaul capacity.
Mobile applications
In our view AT&T, like Verizon, has been decisively outflanked by Apple and Google in the mobile applications space. We see little prospect that any US carrier can successfully re-take this ground from arguably the two most powerful technology companies in the world.
Investment thesis
We are Neutral on AT&T. Although it should benefit from growing mobile scarcity, trends in wireless may weaken, we believe, following loss of iPhone exclusivity which ended February 2011.
100
abc
Financials & valuation: AT&T

Neutral
12/2012e 12/2013e
124,399 40,080 -19,462 20,618 -2,996 18,384 20,905 1,123 19,507 13,379
124,847 43,676 -19,454 24,223 -3,175 21,848 21,848 -7,865 13,983 13,983
126,302 44,817 -19,153 25,664 -3,087 23,417 23,417 -8,430 14,987 14,987
128,943 46,297 -19,505 26,792 -3,214 24,460 24,460 -8,806 15,654 15,654
AT&T revenues AT&T EBITDA adjusted AT&T EBIT adjusted AT&T PBT adjusted AT&T EPS adjusted
124,399 41,659 22,959 19,963 2.25
124,847 43,676 25,023 21,848 2.34
126,302 44,817 26,504 23,417 2.51
128,943 46,297 27,674 24,460 2.62
32,010 -20,302 -21,449 -9,916 -3,549 17,779
32,653 -19,257 -21,607 -10,120 -2,050 13,396
32,936 -19,204 -19,204 -10,580 -3,152 13,732
33,617 -19,101 -19,101 -10,897 -3,618 14,515
1.8 5.5 1.0 12.6 1.5 11.3 5.9
1.8 5.0 1.0 12.2 1.5 8.5 6.1
1.7 4.8 0.9 11.4 1.4 8.7 6.3
1.7 4.6 0.9 10.9 1.4 9.2 6.4
134,121 103,963 19,951 1,437 275,958 26,755 66,167 64,730 112,122 229,843
132,231 105,656 22,507 4,000 279,118 25,965 66,680 62,680 115,559 230,429
131,001 106,937 22,653 4,000 280,154 25,824 63,528 59,528 119,888 230,767
130,331 107,204 22,917 4,000 280,897 25,508 59,910 55,910 124,565 230,944
Issuer information
28.47 Target price (USD) T.N 168,286 100 United States Richard Dineen
30.00 Potent'l return (%)
5.4
Bloomberg (Equity) T US Market cap (USDm) 168,286 Enterprise value (USDm) 219787 Sector Diversified Telecoms Contact 1 212 525 6707
32 30 28 26 24 32 30 28 26 24 22 20 18 2010
Rel to S&P 500

Ratios (%)
1.1 -2.7 -4.1 -1.6 10.7
0.4 9.0 17.5 18.8 3.8
1.2 2.6 5.9 7.2 7.2
2.1 3.3 4.4 4.5 4.5
22 20 18 2009
AT&T
2011
2012
0.5 6.6 12.5 7.8 32.2 16.6 13.4 57.6 1.6 49.5
0.5 7.3 12.3 5.8 35.0 19.4 13.8 54.1 1.4 52.1
0.5 7.5 12.7 6.1 35.5 20.3 14.5 49.5 1.3 55.3
0.6 7.6 12.8 6.4 35.9 20.8 14.4 44.8 1.2 60.1
Source: HSBC
3.28 2.25 1.69 18.87
2.34 2.34 1.73 19.32
2.51 2.51 1.78 20.04
2.62 2.62 1.84 20.83
101
abc
Belgacom (Neutral, TP EUR30)

Belgacom has one of the most advanced NGA roll-outs in Europe, with 76% VDSL coverage at end Q3 2010. As a result, we expect Belgacom to close 10% to 15% of its local exchanges by 2013, switching off around 40% of the unbundled lines. This will force alternative operators such as Mobistar to replace cheap ADSL (EUR7.5 per month) with expensive VDSL bistream wholesale offers (EUR13.9 per month). In determining the VDSL wholesale price, the regulator has applied a 15% mark-up to the cost incurred by Belgacom. Hence, though regulated, we believe the move of competitors to wholesale VDSL from unbundling will be beneficial for Belgacom. There is no active discussion on net neutrality in Belgium. We believe the Belgian regulation on net neutrality is likely to be in line with EU recommendations, which currently appear pragmatic (authorising prioritisation in the event of congestion).
EUR0.03 per MB. We see pricing power as Belgacoms main competitor, Mobistar, is also experiencing strong demand for data (+117% volume and 63% revenue growth in 2010) and also has (low) caps of 50MB and 100MB.

Belgacom was among the earliest to launch IPTV services in Europe, and now has around 20% market share of the national TV market as at the end of Q3 2010 with the rest belonging to cable operators (predominantly Telenet). IPTV penetration had reached 60% of Belgacoms broadband connections by Q3 2010. Indeed, IPTV has become a relatively material item for the incumbent, and now provides 16% of its consumer revenues (growing at 43% y-o-y yearto-date Q3 2010).
Mobile applications
Belgacom has joined the Rich communication suite initiative to help develop more data services which we see as a positive. But, although we see the soft SIM threat as overstated in most European countries, the situation may be different in Belgium. There are no handset subsidies (apart from Telenets MVNO) and a vendor such as Apple would not take a big risk by taking the soft SIM route. But Belgium is more an exception in Europe and it may not be worth vendors adopting a very different strategy in a small market.
Mobile connectivity
Mobile data demand is picking up in Belgium, with Belgacoms advanced data services revenues growing by 11% y-o-y in 9M 2010. Given there are no handset subsidies in Belgium, smartphone penetration here is behind most of Europe, but, with smartphone prices falling, we expect penetration to increase rapidly. The mobile networks in Belgium do not seem to be under any strain as yet. Belgacom commented in its Q3 2010 conference call that it still had a huge amount of capacity to fill on its 3G networks. Belgacom has adopted tiered data plans. It has four mobile data plans catering to various categories of users, starting with a package of 50MB for EUR5 and with a high-end package of 1GB for EUR25. Additional usage is charged at
Investment thesis
We believe the company has the right network assets (fixed and mobile) and strategy (quad-play offers) to monetise fixed connectivity, but it still faces stiff competition from cables Telenet and VOO. On the mobile data front, Mobistar, which has exclusivity for iPhone, has been ahead of Belgacom in monetising the opportunity so far, but we see a good potential here for Belgacom too. We maintain our Neutral rating and EUR30 target price on Belgacom.
102
abc
Financials & valuation: Belgacom

Neutral
12/2011e 12/2012e
Profit & loss summary (EURm)
Cash flow summary (EURm)
5,990 1,967 -706 1,261 -117 1,144 1,134 -241 904 749
6,589 2,398 -802 1,596 -106 1,489 1,052 -256 1,218 679
6,622 1,933 -776 1,157 -102 1,055 1,055 -264 768 673
6,693 1,951 -728 1,223 -98 1,125 1,125 -337 762 717
Belgacom Revenues Belgacom EBITDA adjusted Belgacom EBIT adjusted Belgacom Net income adjusted Belgacom FCF definition
5,990 1,957 1,261 904 798
6,589 1,961 1,596 1,218 900
6,622 1,933 1,157 768 893
6,693 1,951 1,223 762 889
1,406 -597 -631 -684 -119 1,037
1,578 -730 -683 -702 -180 1,413
1,546 -671 -683 -701 -162 896
1,549 -678 -690 -714 -145 827
1.9 5.8 2.4 11.6 3.4 10.7 7.7
1.7 4.7 2.2 12.8 3.2 14.6 8.1
1.7 5.7 2.2 12.9 3.2 9.3 8.2
1.6 5.6 2.3 12.1 3.1 8.5 8.2
Balance sheet summary (EURm)
2,711 2,420 1,945 471 7,450 1,771 2,187 1,716 2,521 4,834
3,548 2,351 2,135 500 8,334 2,481 2,036 1,536 2,679 5,053
3,560 2,246 2,277 637 8,382 2,507 2,011 1,374 2,733 4,938
3,571 2,196 2,432 782 8,500 2,579 2,011 1,229 2,784 4,839
Issuer information
Share price (EUR)
27.02 Target price (EUR)
Reuters (Equity) BCOM.BR Market cap (USDm) 12,380 Free float (%) 42 Country Belgium Analyst Nicolas Cote-Colisson
Bloomberg (Equity) BELG BB Market cap (EURm) 9,135 Enterprise value (EURm) 11211 Sector Diversified Telecoms Contact 44 20 7991 6826
33 31 29 27 25 23 21 19 17 2009
Belgacom

Ratios (%)
0.1 3.3 8.5 8.5 0.3
10.0 21.9 26.5 30.2 -9.6
0.5 -19.4 -27.5 -29.2 -0.9
1.1 0.9 5.7 6.6 6.6
33 31 29 27 25 23 21 19 17 2010
Rel to BEL-20
2011
2012
Per share data (EUR)
1.2 20.2 31.3 13.0 32.8 21.1 16.8 67.9 0.9 81.9
1.3 21.3 26.1 16.6 36.4 24.2 22.6 52.7 0.6 102.7
1.3 15.3 24.9 10.3 29.2 17.5 19.0 46.4 0.7 112.5
1.4 16.5 26.0 10.2 29.1 18.3 19.9 40.9 0.6 126.1
Source: HSBC
2.82 2.34 2.08 7.87
3.79 2.12 2.18 8.35
2.39 2.10 2.23 8.51
2.37 2.23 2.22 8.67
103
abc
BT Group (Neutral, TP 200p)

BT is now well underway with its NGA upgrade deployment, and ultimately targets passing twothirds of the countrys premises with either FTTN/VDSL or FTTP. In our view, Ofcoms regulation is not as sympathetic as is warranted by a project of this scale, but BT has nonetheless been given a good degree of flexibility on setting its pricing. And although the regulators armoury will include techniques like unbundling of the sub-loop, we believe this is uneconomic in the majority of the country (although it will doubtless be in the interests of BTs retail competitors to express continuing interest in this approach). There is already good evidence that the prospect of fibre-based, superfast broadband is causing some BT competitors to rethink their strategies. For instance, Orange (thought of primarily as a mobile operator, but also a provider of ADSL services) has chosen not to invest in upgrading its platform, and is shifting instead to taking a wholesale product from BT. Presumably it reasoned there was little point in lavishing expenditure on ADSL ULL infrastructure, just as fibre is becoming widely available. Since BTs Openreach division must make available its fibre product to all-comers on equal terms, there is every possibility that BTs retail market share will come under ongoing pressure: some of its competitors may prove more adept at selling the connectivity than BT itself. But the key point is that, whereas the company is paid only GBP7.4 per month for an unbundled copper line, it could receive double this for wholesale fibre. Moreover, what constitute additional revenues for BT are extra costs for those rivals that rely on the incumbents access infrastructure. Their need to be able to digest these cost hikes should result in a more price-disciplined market in general.
Ofcoms views on net neutrality look sensible and pragmatic. While insisting on due transparency, the regulator argues that business models incorporating charges for prioritisation could in fact bring certain benefits. Meanwhile, BT is moving to address the opportunities available in the OTT video market by launching its own CDN specifically targeting those with video content. BT Vision, the incumbents IPTV service, will doubtless be an early customer, but the CDNs capabilities will be sold to third parties as well.

To date, BT Vision has struggled. This is perhaps not surprising given the intense competition from BSkyB in particular. Ofcom has recently intervened to give BT and others access to BSkyBs sports content, and this has now been incorporated into the Vision line-up. However, despite heavy advertising, customer net adds have responded only slowly. We think the YouView OTT platform (previously Project Canvas) should provide a boost to the whole concept of IPTV, but its set-top boxes are now delayed until early next year. Even then, we are more confident of BTs ability to monetise this type of service through selling the underlying connectivity rather than by acting as a payTV retailer.
Investment thesis
BT has made progress in tackling its various operational challenges (for instance, in its Global Services division), and management has now been able to move on to address m strategic issues in particular with regard to the upgrade to a fibre access platform. However, that said, the pension overhang remains an issue and the key factor that underpins our Neutral stance. While the financial markets have improved since the time of the triennial review, we would caution that the Pensions Regulator will want reassurance that the schemes obligations can be met if economic conditions deteriorate once more.
104
abc
Financials & valuation: British Telecom

Neutral
03/2012e 03/2013e
Profit & loss summary (GBPm)
Cash flow summary (GBPm)
20,859 5,162 -3,039 2,123 -1,158 1,007 1,445 22 1,028 1,046
20,173 5,620 -3,027 2,593 -942 1,705 1,868 -294 1,410 1,349
20,067 5,813 -2,963 2,850 -811 2,054 2,122 -455 1,598 1,531
20,083 5,974 -2,938 3,036 -732 2,321 2,382 -530 1,789 1,718
BT Adjusted revenues BT Adjusted EBITDA BT Adjusted PBT BT Adjusted EPS (GBp) BT FCF
20,911 5,639 1,735 18.0 2,106
20,173 5,847 2,016 20.4 2,061
20,067 5,913 2,222 22.3 2,183
20,083 5,974 2,382 23.7 2,143
3,888 -2,509 -2,794 -265 -1,977 1,347
4,235 -2,737 -3,235 -543 -1,355 1,643
4,142 -2,696 -2,696 -568 -878 1,801
4,164 -2,710 -2,710 -624 -830 1,952
1.3 5.4 1.8 13.7 7.6 3.7
1.3 4.7 1.8 10.6 26.2 9.4 4.0
1.3 4.4 1.8 9.4 9.0 10.3 4.3
1.2 4.1 1.8 8.4 5.2 11.3 4.8
Balance sheet summary (GBPm)
3,672 14,856 6,285 2,482 28,680 7,151 12,791 10,309 -2,650 15,180
3,465 14,598 4,489 1,033 24,831 6,759 9,987 8,954 547 14,761
3,465 14,331 5,417 1,911 25,502 6,979 9,987 8,076 1,589 14,323
3,465 14,104 4,556 1,000 24,423 7,120 8,246 7,246 2,753 14,004
Issuer information
Share price (GBPp)
185 Target price (GBPp)
200 Potent'l return (%)
7.9
Reuters (Equity) BT.L Market cap (USDm) 23,013 Free float (%) 100 Country United Kingdom Analyst Stephen Howard
Bloomberg (Equity) BT/A LN Market cap (GBPm) 14,381 Enterprise value (GBPm) 26478 Sector Diversified Telecoms Contact 44 20 7991 6820
196 176 156 136 116 196 176 156 136 116 96 76 56 2010
Rel to FTSE ALL-SHARE

Ratios (%)
-2.5 61.8 605.3 6.0
-3.3 8.9 22.2 69.3 28.7
-0.5 3.4 9.9 20.5 13.5
0.1 2.8 6.5 13.0 12.2
96 76 56 2009
British Telecom
2011
2012
Per share data (GBPp)
1.3 9.9 -83.4 5.8 24.7 10.2 4.5 0.0 2.0 37.7
1.3 12.5 -128.3 7.6 27.9 12.9 6.0 1567.5 1.6 47.3
1.4 14.2 143.3 8.5 29.0 14.2 7.2 500.3 1.4 51.3
1.4 15.5 79.1 9.1 29.7 15.1 8.2 260.7 1.2 57.5
Source: HSBC
13.28 13.52 6.90 -34.24
18.19 17.40 7.32 7.06
20.62 19.75 8.05 20.51
23.09 22.17 8.86 35.52
105
abc
Bezeq (Overweight, TP ILS11.6)

Fixed line business in Israel remains a duopoly with HOT (cable operator, NR) being the only other alternative to the incumbent Bezeq with c70%+ market share. Bezeqs roll-out of NGN means an unmatched broadband offering that could possibly arrest the rate of fixed line erosion. HOTs broadband offering is hampered by service and last mile connectivity reach. Bezeqs 100% last mile connectivity means that competitors use the naked ADSL pipes to offer services to their customers. Absence of regulations on net neutrality is positive for incumbents. Local loop unbundling by altnets is yet to be allowed in Israel. Bezeqs market share in the commercial market has fallen below the 80% threshold (regulatory requirement), meaning that it is eligible for regulatory relief measures such as allowing Bezeq to bundle services; these measures should provide Bezeq with better tools to cope with the expected competition.
We expect Pelephone to follow with a similar data pricing structure. Given the rational pricing environment in Israel, with these tiered-data pricing systems, risks of capacity constraints seem low to us.

OTT players are less likely to pose a threat to the incumbent given the firm access control of Bezeq. Naked DSL pricing (regulated) ensures that Bezeq collects its toll, in the case of alternative VoB operators, taking voice minutes share from Bezeq. Its strong brand name in Israel as the preferred operator at home should help Bezeq offer advanced NGN based services to its large subscriber base in the future.
Mobile applications
Pelephone has been able to maintain a good grip on content, data applications and VAS thanks to a relatively close garden approach, presented by all mobile operators. The ability of Pelephone to successfully charge for VAS/application offerings should be supported by its domestic customisation offering. Still we expect data revenues in the future to gradually move towards tiered pricing for mobile connectivity, maximising data network resources.
Mobile connectivity
Bezeq (via Pelephone) launched its HSPA network (previously CDMA) in January 2009, giving a major push to data applications. Its 3G subscriber base increased from 46% of the total subscriber base in Q1 2009 to 63% of the total subscriber base in Q3 2010. Data revenues as a percentage of total service revenues for Pelephone reached 25%. There are no signs of capacity constraints on the network (given the late introduction of the iPhone in Israel), despite the ongoing move from walled garden to open garden. Israel is fast moving from all you can eat plans to tiered data pricing; which pushes the capacity crunch thesis further away for the present. Cellcom has recently launched speed tiered pricing (rather than capacity related).
Investment thesis
We are bullish on the ability of Bezeq to grow its cash flow by implementing further efficiencies post ownership change (ie new retirement agreement with its union); material cost savings on the back of NGN implementation (ie reducing national switches from c140 to only 30); capex drop following NGN project completion; mobile HSPA network grabbing all new roaming revenues and corporate clients to better handle MTR cut/MVNO threats. A strong macro outlook for Israel should also help Bezeq maximise its ongoing real estate asset sales.
106
abc
Financials & valuation: Bezeq

Financial statements Year to Profit & loss summary (ILSm) 12/2009a 12/2010e 12/2011e 12/2012e Key forecast drivers Year to 12/2009a 12/2010e
Overweight
12/2011e 12/2012e
Cash flow summary (ILSm)
11,519 4,457 -1,485 2,972 31 2,969 2,969 -807 2,162 2,162
11,605 4,722 -1,414 3,308 -94 3,127 3,127 -795 2,332 2,332
11,663 4,741 -1,357 3,384 -62 3,265 3,265 -783 2,481 2,481
11,808 4,844 -1,303 3,541 -41 3,468 3,468 -798 2,670 2,670
Wireline revenues Pelephone revenues Bezeq International Revenues Wireline EBITDA margin Pelephone EBITDA margin
5,303 5,376 1,316 44 33
5,175 5,638 1,355 47 35
4,983 5,804 1,382 46 36
4,903 6,076 1,401 46 36
Balance sheet summary (ILSm)
3,622 -1,622 -1,663 -1,941 -1,704 1,789
3,806 -1,398 -1,183 -3,646 832 2,345
3,916 -1,082 -1,012 -2,407 -498 2,834
3,997 -928 -858 -2,576 -563 3,069
2.5 6.5 3.5 12.6 4.1 7.0 13.3
2.6 6.4 3.7 11.6 5.1 9.1 8.7
2.6 6.3 4.5 11.0 5.1 11.0 9.2
2.5 6.0 4.7 10.3 5.0 11.8 9.8
1,885 5,303 3,659 580 13,941 3,196 4,136 3,556 6,544 8,361
1,856 5,374 3,567 660 13,516 3,149 5,048 4,388 5,252 8,134
1,851 5,034 3,676 768 13,222 3,170 4,658 3,890 5,326 6,622
1,855 4,584 4,223 1,279 13,291 3,197 4,606 3,327 5,421 6,186
Issuer information
Share price (ILS)
10.03 Target price (ILS)
16
Reuters (Equity) BEZQ.TA Market cap (USDm) 7,360 Country Israel Analyst Avshalom Shimei
Bloomberg (Equity) BEZQ IT Market cap (ILSm) 26,958 Sector Diversified Telecoms Contact 972 3 710 1197
Price relative
12 11 10 9 8 7 6 5 4 3 2009
Bezeq Source: HSBC
Ratio, growth and per share analysis Year to Y-o-y % change 12/2009a 12/2010e 12/2011e 12/2012e

Ratios (%)
4.6 8.7 12.6 18.5 27.4
0.8 5.9 11.3 5.3 8.9
0.5 0.4 2.3 4.4 5.6
1.2 2.2 4.6 6.2 6.9
12 11 10 9 8 7 6 5 4 3 2010
Rel to TA-100 INDEX
2011
2012
Per share data (ILS)
1.3 24.3 38.4 17.4 38.7 25.8 54.4 0.8 101.9
1.4 29.9 39.5 18.0 40.7 28.5 50.5 83.6 0.9 86.7
1.6 34.9 46.9 19.8 40.7 29.0 76.3 73.1 0.8 100.7
1.8 42.6 49.7 21.5 41.0 30.0 118.3 61.4 0.7 120.1
1.34 0.80 1.33 2.46
0.87 0.87 0.87 1.96
0.92 0.92 0.92 1.97
0.98 0.98 0.98 1.99
107
abc
Deutsche Telekom (Underweight, TP EUR9.5)

Deutsche Telekom (DT) announced its NGA programme as early as 2006, initially deploying solely VDSL/FTTC. At its 2010 capital markets day, DT announced a further roll-out of NGA, targeting 31% of German homes to be passed with VDSL/FTTC and up to another 10% with FTTH by the end of 2012. Altnets have either been consolidated or are struggling to keep their subscriber bases stable. Even big pocket players like Vodafone and Telefonica are not growing their fixed businesses in Germany and are hesitant about building their own NGA networks. Cable operators, having upgraded faster than DT, are gaining share strongly in an increasingly saturated market. BNetzA, the German regulator, has obliged DT to open its NGA network on a wholesale as well as on an unbundling basis, but conditions are such that even Telefonica and Vodafone find it uneconomical to unbundle at the street cabinet level. Wholesale price points for NGA are at least double the copper ULL fee. On net neutrality in Germany, BNetzA are committed to net neutrality principles but leave themselves a big backdoor by accepting that differentiation of service can be beneficial to society and economically sensible as long as it does not discriminate and limit competition. We view the political climate around net neutrality as sensible overall. Deutsche Telekom entered the CDN market in early 2009, partnering with California-based Edgecast.
300MB for cheaper smartphone tariffs; only price points from around EUR60 include 1GB of data volume and more. In February 2011, DT announced a new initiative with France Telecom to explore areas of cooperation, including radio access network sharing in Europe and improving WiFi user experience while roaming.

DT Triple Play product is slowly gaining traction. With 1m subscribers, DTs market share remains well below 4% of the German TV market, despite featuring Bundesliga and an increasingly wide VideoOnDemand library. OTT platforms have yet to gain traction in Germany and cable operators are just about to introduce VideoOnDemand. Its very early days for OTT in Germany, but DT is sensibly positioning itself in this market.
Mobile applications
DT has attempted to develop a broad service set, going well beyond connectivity into the applications layer. Capabilities include a networkbased address book. But, while a useful feature set, many (though not all) of these capabilities are available via the big internet and technology brands, and customers may well gravitate towards using these providers rather than their telecoms supplier. DT is part of WAC, the Wholesale Application community, but to date little tangible results have come out of this alliance.
Investment thesis
We believe Deutsche Telekom is at risk of continuing to underinvest in what is an aboveaverage competitive domestic fixed and mobile market. We also see no easy solution to the subscale issues of its former growth engine TMobile US. We, thus, remain Underweight on Deutsche Telekom given its unattractive organic growth profile versus peers.
Mobile connectivity
Historically DT has not had tiered data pricing, but has had fairly generous usage policies. With the new tariff scheme introduced before the 2010 Christmas sales season, DT changed its small print, capping the data at a volume as low as
108
abc
Financials & valuation: Deutsche Telekom

Underweight
12/2011e 12/2012e
64,639 19,906 -13,894 6,012 -2,555 2,655 5,723 -1,782 353 3,493
62,317 18,162 -11,015 7,147 -2,489 4,506 4,920 -1,415 2,676 2,991
60,787 18,054 -10,773 7,280 -2,560 5,011 5,011 -1,416 2,945 2,945
60,630 18,093 -10,801 7,292 -2,344 5,334 5,334 -1,484 3,194 3,194
DT Revenues DT EBITDA adjusted DT EBIT adjusted DT Net income adjusted DT FCF definition
64,639 20,668 9,143 5,140 6,969
62,317 19,555 8,414 4,425 6,245
60,787 19,059 8,576 4,279 6,321
60,630 19,098 8,683 4,553 6,414
14,982 -9,202 -8,649 -3,474 2,717 6,174
14,220 -8,927 -10,929 -3,402 1,801 5,438
14,486 -9,151 -8,751 -3,025 -2,309 4,965
14,626 -9,124 -8,724 -2,998 -2,503 5,138
1.4 4.5 0.9 12.4 1.2 12.5 7.9
1.5 5.2 1.0 14.4 1.2 10.6 7.1
1.5 5.1 1.0 14.5 1.2 9.6 7.1
1.5 5.0 1.1 13.2 1.2 9.9 7.1
51,705 45,468 23,012 7,206 127,774 15,967 48,117 40,911 36,354 97,012
52,923 45,141 14,823 4,734 128,216 17,984 47,446 42,712 35,153 90,170
49,829 46,213 15,103 5,243 126,332 18,090 45,646 40,403 34,701 87,812
46,797 47,168 15,883 6,046 124,897 18,409 43,946 37,900 34,504 85,392
Issuer information
Share price (EUR)
-4.0
Reuters (Equity) DTEGn.DE Market cap (USDm) 57,976 Free float (%) 63 Country Germany Analyst Dominik Klarmann
Bloomberg (Equity) DTE GR Market cap (EURm) 42,781 Enterprise value (EURm) 94126 Sector Diversified Telecoms Contact +49 211 910 2769
14 13 12 11 10 9 8 7 6 2009
Deutsche Telekom

Ratios (%)
4.8 10.5 -14.6 -23.1 22.1
-3.6 -8.8 18.9 69.7 -14.2
-2.5 -0.6 1.9 11.2 -0.6
-0.3 0.2 0.2 6.4 9.4
14 13 12 11 10 9 8 7 6 2010
Rel to DAX-100
2011
2012
0.7 8.7 9.2 2.3 30.8 9.3 7.8 97.6 2.1 36.6
0.7 7.7 8.4 3.9 29.1 11.5 7.3 105.4 2.4 33.3
0.7 8.4 8.4 4.3 29.7 12.0 7.1 100.7 2.2 35.9
0.7 8.6 9.2 4.4 29.8 12.0 7.7 94.6 2.1 38.6
Source: HSBC
0.08 0.80 0.78 8.34
0.61 0.69 0.70 8.08
0.68 0.68 0.70 8.05
0.75 0.75 0.70 8.07
109
abc
eAccess (Overweight (V), TP 79,000)

Following the business combination of July 2010, the fixed-assets of eAccess were incorporated into eMobile in a reverse acquisition. The DSL business comprises 42% of overall revenues, but mobile is the growth driver. eAccess primarily provides DSL connectivity on a wholesale basis to ISPs, but also offers bundled wireline products with eMobile wireless broadband packages. We expect DSL revenues and subscribers to decline gradually, as a result of increasing competition (and lower prices) from fibre providers.
42Mbps service in Japan, enabling it to compete with DoCoMos Xi LTE service.

eAccess, as a mostly wholesale DSL supplier, is less involved in fixed line applications. This is largely the preserve of its ISP clients its primary focus is on maintaining EBITDA trends given higher levels of customer churn.
Mobile applications
eMobile has refocused its smartphone strategy, which was initially based around Windows Mobile. It launched HTCs Aria Android smartphone on 17 December 2010, and the Pocket WiFi S from Huawei in January 2011 both feature Android 2.2, and allow full tethering, or portable WiFi service. In conjunction with a flatrate voice promotion, we believe these initiatives will help restore sales that suffered in the second quarter from a lack of new products.
Mobile connectivity
eMobile has been at the vanguard of wireless broadband service provision in Japan. Launched in March 2007, it took advantage of capacity constraints at the larger operators, and slower speeds at PHS operator Willcom, to corner the market in wireless broadband. It reached operating profitability after just 2.5yrs in June 2009, and reached 3m accounts in January 2011. eMobile ARPU has declined as a result of increased competition (from WiMAX operator UQ Communications) and an increasing proportion of MVNO sales (at lower ARPU and acquisition costs). However, we expect it to benefit from the growth in demand for Android smartphones (it offers models from HTC and Huawei) and also mobile WiFi: its Pocket WiFi product remains very popular at c50% of gross additions, fuelled by growing demand for tablet PCs. With its late launch of HSPA services, it was able to take advantage of lower-priced equipment from Ericsson and Huawei, and built out a high-speed network at a fraction of the cost of NTT DoCoMo, which launched services six years earlier. In December 2010, it launched the first
Investment thesis
We believe eMobiles competitiveness has improved considerably as a result of the product launches in the quarter ending December 2010. The upgrade to 42Mbps, and the launch of both high and low-end Android smartphones (Aria and Pocket WiFi S) should enable it to capitalise on growing consumer demand. We also see upside for the company as a result of the growth in demand for tablet PCs: it is the only provider currently to offer tethering or mobile WiFi services without restriction. For the eMobile business on a stand-alone basis, we forecast an EBITDA margin of 28% in FY March 2011 and 30% in March 2012. In our view, this business is a case study in the potential profitability of the stand-alone mobile data business model.
110
abc
Financials & valuation: eAccess Ltd

Financial statements Year to 03/2010a 03/2011e 03/2012e 03/2013e Key forecast drivers Year to 03/2010a
Overweight (V)
03/2011e 03/2012e 03/2013e
Profit & loss summary (JPYbn)
Cash flow summary (JPYbn)
83 27 -7 19 -2 11 11 -7 4 6
184 57 -34 24 -8 13 16 -6 7 9
219 83 -38 45 -7 36 39 -6 30 23
254 91 -35 56 -5 49 51 -8 41 30
ADSL Revenue (JPY bn) Mobile Revenue (JPY bn) ADSL EBITDA (JPY bn) Mobile EBITDA (JPY bn) Total Operating Profit (JPY bn Total Capex (JPY bn)
71 113 26 18 19 3
61 144 23 44 24 39
56 172 21 63 45 51
55 211 20 71 56 42
Balance sheet summary (JPYbn)
15 -3 -4 -4 -5 14
53 -53 -53 -4 150 -9
71 -51 -51 -3 -17 19
78 -42 -42 -3 -33 36
2.5 7.7 9.6 11.4 5.8 8.1 4.7
1.9 6.2 1.4 21.9 2.0 -5.2 1.2
1.5 4.1 1.3 7.7 1.7 10.6 1.6
1.2 3.3 1.2 5.9 1.3 20.6 1.6
3 16 46 26 87 17 55 29 13 21
10 200 178 100 409 43 279 179 75 245
9 214 179 100 423 46 262 162 103 255
9 221 181 100 432 51 229 129 141 260
Issuer information
Share price (JPY)
51400 Target price (JPY)
79000 Potent'l return (%) 54.9
Reuters (Equity) 9427.T Market cap (USDm) 2,135 Free float (%) 54 Country Japan Analyst Neale Anderson
Bloomberg (Equity) 9427 JP Market cap (JPYbn) 178 Enterprise value (JPYbn) 355 Sector Diversified Telecoms Contact +852 2996 6716
80634 75634 70634 65634 60634 55634 50634 45634 40634 2009
eAccess Ltd

Ratios (%)
-12.1 10.6 14.1
120.9 116.0 21.0 19.8 -47.9
19.4 45.0 96.9 176.5 186.6
15.7 9.7 23.0 36.2 29.6
80634 75634 70634 65634 60634 55634 50634 45634 40634 2010
Rel to TOPIX INDEX
2011
2012
Per share data (JPY)
3.4 46.5 57.4 5.0 32.0 22.9 12.5 220.0 1.1 51.2
1.4 10.6 21.3 4.8 31.3 12.6 7.0 238.1 3.1 29.3
0.9 11.0 26.1 8.4 38.0 20.7 11.2 157.6 1.9 43.8
1.0 12.8 24.7 10.4 36.0 22.0 18.6 91.2 1.4 60.8
Source: HSBC
2,864 4,496 2,400 8,929
2,343 2,343 600 25,428
8,732 6,714 800 29,702
11,846 8,704 800 40,748
111
abc
France Telecom (Overweight, TP EUR21

France is the exception rather than the rule in Europe. Regulation has been effective from the beginning of 2000: unbundling of the local loop, together with an ADSL bistream provided by FT gave Iliad and SFR (after consolidating the market) enough scale to afford to build their own access network (FTTH). As a consequence, FT cannot force altnets to migrate from cheap ULL to expensive wholesale offers and regain EBITDA market share as we see in other European countries. On the other hand, FT does not face a strong cable (6% broadband market share). We expect operators to start pushing mass market distribution of FTTH this year. FT plans to pass 10m homes in 2015 and 15m by 2020 (out of 28m) for a total cost of EUR2.5bn. On the positive side for FT, the French regulator (ARCEP) published a list of recommendations to preserve the neutrality of the net, in September 2010. ARCEP insists on granting operators the ability to market managed services due to scarcity alongside internet access, which may create some revenue opportunities with OTTs in particular. FT CEO Stephane Richard has been calling for a fair split of value (and costs) between content providers and telcos (Les Echos, 18 November 2010). This can be achieved by forcing traffic generators to pay for the use of the networks. It is too early to see tangible signs of pricing power in the business to business area, but we are seeing some on the retail side: In Q1, Iliad and SFR increased their pricing for triple play by more than the VAT increase (we estimate that two-thirds was not tax related) thanks to new set-top boxes and new services (unlimited calls to mobile now included). FT still commands a small premium, but we think the trend has been favourable for FT.
Mobile connectivity
FT has been introducing caps in France and in other European countries (UK, Spain) to preserve the quality of the network. Stephane Richard already warned that FT will move towards yield management (pricing depending on the load on the network). We do not believe Iliads entry in the mobile market in 2012 should disrupt this data pricing trend: Iliad still has to secure a data roaming agreement first (time to build its 3G network) and the available wholesale pricing may reflect the congestion issues.

36% of FTs broadband customer base takes IPTV services. IPTVs success was not triggered by troublesome cable competition but by Iliads early and innovative move into IPTV. FT controls its box and offer premium content (Orange own TV channel, but also proposes Canal+ packs). FT bought 49% of the OTT Dailymotion (second largest video sharing site behind YouTube) to build its franchise in aggregation and content broadcasting.
Mobile applications
Orange is promoting its own mobile applications such as Orange TV, map services or music streaming through its partnership with Deezer. These services are part of the largest contracts but can also be added as paid options for smaller packs.
Investment thesis
Despite strong domestic competitors, we think FT has the right network and application strategy. We think FT may benefit from its large scale (43% broadband market share in France) to monetise its network against OTT traffic, but also in mobile, with the strong data growth across its geography. FT is also offering a high level of applications in fixed and mobile. We are Overweight FT, target price EUR21.
112
abc
Financials & valuation: France Telecom

Overweight
12/2011e 12/2012e
45,944 14,794 -6,935 7,859 -2,299 5,560 7,161 -2,295 2,997 4,537
45,341 15,608 -7,132 8,476 -1,828 6,903 7,640 -1,858 5,659 4,604
45,439 15,504 -6,807 8,697 -1,681 7,475 7,945 -2,242 4,634 4,725
45,165 15,460 -6,736 8,724 -1,518 7,789 8,159 -2,337 4,828 4,842
FT revenue FT EBITDA FT EBIT FT Net Income FT FCF
45,944 14,794 7,859 2,997 8,443
45,341 14,867 8,476 5,659 8,055
45,439 14,968 8,697 4,634 7,734
45,165 14,867 8,724 4,828 6,889
13,775 -5,659 -7,031 -3,141 -1,918 5,316
11,727 -5,378 -5,898 -3,706 -3,113 6,521
12,581 -5,725 -6,693 -3,708 -2,179 5,855
11,909 -5,967 -6,379 -3,708 -1,822 5,639
1.6 5.0 1.2 9.5 1.7 13.2 8.6
1.6 4.6 1.4 9.3 1.5 16.0 8.6
1.5 4.5 1.4 9.1 1.4 14.1 8.6
1.5 4.5 1.4 8.9 1.4 13.3 8.6
38,549 24,321 21,956 3,949 92,044 20,517 37,890 33,941 26,021 60,360
39,912 23,153 10,930 2,000 88,975 20,670 32,828 30,828 28,782 51,325
40,780 22,627 11,765 2,835 90,095 22,213 31,484 28,649 29,708 50,124
41,192 22,451 13,587 4,657 92,165 23,150 31,484 26,827 30,827 49,423
Issuer information
Share price (EUR)
Reuters (Equity) FTE.PA Market cap (USDm) 58,293 Free float (%) 73 Country France Analyst Nicolas Cote-Colisson
Bloomberg (Equity) FTE FP Market cap (EURm) 43,015 Enterprise value (EURm) 71521 Sector Diversified Telecoms Contact 44 20 7991 6826
25 23 21 19 17 25 23 21 19 17 15 13 11 2010
Rel to SBF-120

Ratios (%)
-14.1 -19.3 -23.5 -23.7 -2.5
-1.3 5.5 7.8 24.2 1.4
0.2 -0.7 2.6 8.3 2.6
-0.6 -0.3 0.3 4.2 2.5
15 13 11 2009
France Telecom
2011
2012
0.8 9.2 16.9 5.0 32.2 17.1 6.4 118.1 2.3 40.6
0.8 11.1 16.8 7.0 34.4 18.7 8.5 98.1 2.0 38.0
0.9 12.2 16.2 7.1 34.1 19.1 9.2 87.8 1.8 43.9
0.9 12.5 16.0 7.1 34.2 19.3 10.2 78.9 1.7 44.4
Source: HSBC
1.13 1.71 1.40 9.77
2.14 1.74 1.40 10.87
1.75 1.78 1.40 11.22
1.82 1.83 1.40 11.64
113
abc
KPN (OW, TP EUR15)

KPNs current fibre households coverage stands at c15%; it is targeting c60% in the medium term. In the meantime, KPN is relying on VDSL2 (currently 80% IPTV coverage and 70% HDTV coverage) as an interim solution to compete with the cable DOCSIS3.0 alternative. NGA regulation is favourable in the Netherlands: the Dutch regulator OPTA is of the view that NGA investments entail an element of risk and hence regulation should provide room for the incumbent to earn a return adequate to the risk inherent in such investment. OPTA has issued detailed regulation on NGA and laid down clear tariffs regulation for unbundled fibre access, therefore removing most of the regulatory risk on such investment. On the net neutrality side, OPTA has not expressed any views in public, but we generally expect it to follow the European Commission recommendations, which so far have been pragmatic and in general positive for operators.
month data plans cost between EUR45 to EUR110, depending on the chosen allocation of voice minutes and texts. KPNs German operation, E-Plus, has also announced data plans, with initial caps as low as 50MB; beyond this point, customers have the option of either accepting a reduced speed or topping up. It is particularly encouraging to witness a teenager (market share in the teens) operator like E-Plus displaying such a rational stance with tiered pricing plans.

KPN has rapidly expanded its market share to c15% in cable dominated TV market from 10% two years ago. KPN provides TV services through a mix of DVB-T and an IPTV based platform. FTTH, in combination with the VDSL2 upgrade, have enabled KPN to make its IPTV offer a real success, which can be gauged from the fact that IPTV net adds in 2010 have been +155k vs only 17k Digitenne net adds; now IPTV subs constitute one-quarter of TV subscribers, up from 10% a year ago.
Mobile connectivity
KPN has been investing in its networks (fixed and mobile) with domestic capex/sales ratio of c15% in the past two years. Even with rapid growth in data traffic at +130% y-o-y (non-voice revenue as a % of ARPU was c35% in Q4 2010 and 45% of post-paid customers have a data product), KPNs network has not shown sign of congestion. KPN has proactively invested in upgrading the network to HSDPA (speed up to 14.4 Mb/sec), also conducting trials on high speed LTE (up to 100Mbps). Anticipating the pressure on the mobile networks, KPN has already connected close to 60% of its UMTS base station with fibre. On mobile data pricing, management has now delivered on the undertaking that it would migrate to tiered data plans in the Netherlands by the end of 2010. All iPhone plans come with a data bundle of either 500MB or 1GB per month. The 1GB per
Mobile applications
KPN has introduced a few mobile applications and also formed collaboration with some of the key M2M service providers (machine to machine communication) to provide a cost effective comprehensive machine to machine solution. KPN has also reached an agreement with some Dutch banks (Rabobank, ING and ABN Amro) to launch mobile payment services by 2012e.
Investment thesis
Our Overweight rating on KPN is primarily driven by its strong FCF generation as well as its policy of 100% distribution of FCF. KPNs FTTH strategy with VDSL as an a interim solution and its transition to tiered data plans from flat rate plans are steps in the right direction, in our view, to command pricing power and monetise scarcity.
114
abc
Financials & valuation: KPN

Overweight
12/2012e 12/2013e
13,398 5,476 -2,226 3,250 -903 2,316 2,299 -516 1,799 1,703
13,423 5,472 -2,229 3,244 -763 2,457 2,457 -595 1,861 1,824
13,579 5,519 -2,245 3,274 -726 2,531 2,531 -612 1,920 1,882
13,782 5,557 -2,253 3,304 -726 2,567 2,567 -619 1,948 1,909
Netherland International mobile Others Group revenues KPN FCF definition
9,274 4,181 -120 13,335 2,428
9,275 4,265 -117 13,423 2,446
9,339 4,358 -119 13,579 2,635
9,436 4,467 -120 13,782 2,501
3,808 -1,809 -2,149 -1,152 722 2,470
3,902 -1,846 -1,746 -1,277 121 2,256
4,077 -1,887 -1,787 -1,322 0 2,290
4,133 -1,932 -1,882 -1,363 0 2,276
2.3 5.6 2.0 10.9 5.3 13.4 6.8
2.3 5.6 2.1 9.6 5.8 12.2 7.5
2.3 5.5 2.2 8.6 6.2 12.4 8.5
2.2 5.5 2.2 8.0 6.7 12.3 9.1
9,755 7,514 2,870 823 22,737 4,240 12,788 11,965 3,500 15,076
9,332 7,454 3,047 1,000 22,407 4,302 13,086 12,086 3,010 14,531
8,949 7,379 3,071 1,000 21,956 4,362 13,086 12,086 2,599 14,036
8,570 7,388 3,101 1,000 21,604 4,408 13,086 12,086 2,276 13,650
Issuer information
Share price (EUR) Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
11.84 Target price (EUR) KPN.AS 25,233 100 Netherlands Luigi Minerva
Bloomberg (Equity) KPN NA Market cap (EURm) 18,620 Enterprise value (EURm) 30568 Sector Diversified Telecoms Contact 44 20 7991 6928
14 13 12 11 10 9 8 7 6 2009
KPN Rel to AEX

Ratios (%)
-0.8 5.5 14.0 13.8 10.3
0.2 -0.1 -0.2 6.1 13.2
1.2 0.9 0.9 3.0 12.2
1.5 0.7 0.9 1.4 7.3
14 13 12 11 10 9 8 7 6 2010 2011 2012
0.9 20.0 46.4 10.4 40.9 24.3 6.1 341.9 2.2 31.8
0.9 20.2 56.0 10.8 40.8 24.2 7.2 401.5 2.2 32.3
1.0 21.2 67.1 11.2 40.6 24.1 7.6 465.0 2.2 33.7
1.0 22.0 78.3 11.5 40.3 24.0 7.7 531.0 2.2 34.2
Source: HSBC
1.15 1.09 0.80 2.24
1.26 1.23 0.89 2.03
1.41 1.38 1.00 1.91
1.51 1.48 1.07 1.77
115
abc
KT Corp (Overweight, TP KRW60,000)

KT Corp is unusual in that regulatory risk to its wireline business is relatively low. The government has no stake in KT (having sold its final 28% holding in May 2008), and facilitiesbased competition is very strong. Total broadband penetration is approaching 100%, and KT has c43% of this market. The prevalence of highspeed broadband infrastructure from the three main operators means there is little need for KT to offer unbundled services. Each operator has deployed a blend of FTTH and FTTx technology, with KT tending towards fibre to the home. A blend of FTTx and LAN/VDSL within the user premises is also common.
The size and location of KTs WiFi accesspoints means that it is very advanced in off-loading data from the wireless to the wireline network: in September 2010, 67% of the its mobile data traffic around 2,500 Terabytes per month was offloaded on to WiFi.

KT has been the most aggressive telco operator developing an IPTV product. It saw subscriptions increase sharply after adding with satellite content in 2010, and targets 3m subscribers by end 2011 (from 2m at end 2010). It currently offers 120,000 channels and 90,000 VoD files, and is in negotiations with key content companies to further improve its content.
Mobile applications
As for the other Korean and Japanese operators, KT has focused mainly on creating a local app store that offers users a familiar, Korean-language portal to download mobile content. It is not as aggressive as SK Telecom in sourcing or developing its own in-house content.
Mobile connectivity
Like SoftBank in Japan, KT was the first operator in Korea to tap into nascent demand for smartphones when it launched the iPhone in late November 2009. Since then, 2.7m subscribers, or 17.5% of its subscriber base, have migrated to a smartphone. In July 2010, it announced it would replicate the All-in-One tariffs offering unlimited data from SK Telecom. KT has done this reluctantly, but ultimately could not afford not to match this tariff from the largest operator in the market. Despite our preference for tiered data plans that link usage to revenue, we see substantial upside in the migration of users from cKRW36,000 ARPU levels to smartphone tariffs of KRW50,000 and above. KT continues to leverage the strength of its network: we see greater upside in 2011 from its under-utilised WiBro network, which can be used to create mobile WiFi hotspots for tablet PC users.
Investment thesis
We believe KT Corp has one of the best mobile networks in Asia. It is an integrated operator with an extensive fibre deployment. It has both a CDMA and WCDMA 3G network, and a largely unused wireless broadband network (using WiBro, the Korean variant of WiMAX). Crucially, compared with SK Telecom, it has a massive network of WiFi hotspots KT targets 100,000 by end 2010. This is a legacy of its heavy deployment in 2000-05, and means that it is able to offload wireless network traffic in many public places, relieving the capacity on its wide-area cellular network. Management has been active in seeking to leverage this network, which has helped it gain revenue market share with the iPhone. It targets 25-30 smartphone launches in 2011.
116
abc
Financials & valuation: KT Corp

Financial statements Year to 12/2010a 12/2011e 12/2012e 12/2013e Valuation data Year to 12/2010a 12/2011e
Overweight
12/2012e 12/2013e
Profit & loss summary (KRWb)
Cash flow summary (KRWb)
20,234 4,968 -2,915 2,053 -378 1,517 1,742 -345 1,172 1,396
20,648 5,429 -2,895 2,534 -362 2,221 2,261 -466 1,756 1,796
21,508 6,043 -2,966 3,077 -235 2,911 2,931 -625 2,287 2,307
21,844 6,513 -2,915 3,598 -108 3,560 3,580 -784 2,776 2,796
0.9 3.5 1.2 7.0 0.9 7.6 5.8
0.8 3.0 1.1 5.4 0.8 15.0 6.3
0.7 2.5 1.0 4.2 0.7 19.8 8.7
0.6 2.0 0.9 3.5 0.6 23.3 11.1
Issuer information
Share price (KRW) 41,200 Target price (KRW) 60,000 Potent'l return (%) 3,870 -3,057 -3,082 -298 310 813 4,606 -2,992 -3,092 -569 -946 1,615 5,200 -3,068 -3,178 -613 -1,409 2,132 5,629 -3,121 -3,121 -845 -1,663 2,508 Reuters (Equity) 030200.KS Market cap (USDm) 9,545 Free float (%) 78 Country Korea Analyst Neale Anderson
52
Bloomberg (Equity) 030200 KS Market cap (KRWb) 10,758 Enterprise value (KRWb) 16374 Sector Diversified Telecoms Contact +852 2996 6716
Balance sheet summary (KRWb)
Price relative
1,138 13,948 6,112 935 24,101 5,563 7,497 6,562 11,041 14,700
1,138 14,044 7,490 2,324 25,575 5,407 7,940 5,616 12,183 14,941
1,138 14,146 8,896 3,733 27,083 5,242 7,940 4,207 13,625 15,205
1,138 14,352 10,551 5,396 28,945 5,171 7,940 2,544 15,321 15,474
53945 48945 43945 38945 33945 28945 23945 18945 2009

KT Corp
53945 48945 43945 38945 33945 28945 23945 18945 2010

Rel to KOSPI INDEX
2011
2012
Source: HSBC

Ratios (%)
6.2 19.6 98.7 90.3 136.9
2.0 9.3 23.4 46.4 28.6
4.2 11.3 21.4 31.1 28.5
1.6 7.8 16.9 22.3 21.2
Per share data (KRW)
1.4 10.9 13.0 6.4 24.6 10.1 13.2 59.4 1.3 59.0
1.4 13.5 15.5 8.6 26.3 12.3 15.0 46.1 1.0 82.0
1.4 16.0 17.9 10.2 28.1 14.3 25.7 30.9 0.7 123.6
1.4 18.3 19.3 11.3 29.8 16.5 60.5 16.6 0.4 221.3
4,966 5,918 2,410 46,795
7,441 7,611 2,600 51,636
9,692 9,776 3,580 57,747
11,768 11,852 4,580 64,935
117
abc
Mobily (Overweight, TP SAR71)

Mobile connectivity
We view Mobily as the best placed operator in the MENA region in terms of pricing power for mobile data. Mobily offers a range of tiered data plans and an unlimited plan. Fixed line broadband coverage in Saudi Arabia is limited to a few areas in big cities like Riyadh, Jeddah and Makkah. There are no cable operators in Saudi Arabia; hence, we believe mobile operators will continue to benefit from the uptake of broadband services. There is not much price differentiation within fixed and mobile broadband services, which should continue to drive the demand for mobile broadband services in the kingdom. Mobily is the market leader in the mobile broadband segment with 60% market share. Hence, it has been able to command a pricing premium, which is reflected in its high data ARPU of cUSD55. Pricing has remained stable in Saudi Arabia for broadband services unlike other markets like Egypt. Mobilys unlimited plan sells at SAR350 per month (USD96); the 5GB plan sells at SAR 200 per month (USD55), while 1GB plan sells at SAR100 per month (USD 27). This hasnt changed since late 2007 when broadband was introduced in Saudi Arabia. We expect the contribution of data to Mobilys revenue to grow from 18.3% in 2010 to c29% by 2012e, driven by strong subscriber growth along with limited ARPU dilution due to its strong pricing power. Net neutrality or any other form of intrusive regulation is not a concern for Saudi telcos. This situation is unlikely to change significantly in the medium term, in our opinion.
Mobile applications
Services based on news, entertainment and religious content offer good growth prospects in Saudi Arabia if mobile operators can develop the right products at the right price. Also, with the increase in network coverage and online Arabic content, wealthier nations like Saudi Arabia should demonstrate strong broadband uptake in the coming years, in our view. This has resulted in high data transmitted over Mobilys network, which has increased by 70% y-o-y, registering 85 terabytes/day in 2010 (50TB in 2009 and 19TB in 2008). The growing sales of smart phones and tablets along with falling PC prices and high disposable income should play a significant role in boosting data revenue. Mobily owns Bayanat Al Oula, which has a licence to offer data services in the kingdom over fixed line. We see this as an important platform to offload some network capacity in dense and high usage area like Riyadh.
Investment thesis
Mobily is a growth story and a good way to gain exposure to MENA telcos growth profile, in our view. We like Mobily for its broadband-focused strategy: it is the market leader in this segment, with a market share of 60% in 2010, despite a market share of only 39% of overall subs. Managements focus is on growth via increases in broadband utilisation rates and the achievement of operational efficiency through infrastructure sharing. We believe Mobily is best positioned to capture the growing demand for broadband services, given its growing 3.5G coverage (currently at 92%). We believe there is a demand supply gap in Saudi Arabia, which is being served by Mobily with mobile data card offerings. We expect the data contribution to revenue to increase by 1100bp to 29% of revenue by 2012 as it already has 2.3m broadband subscribers, a figure we expect to grow to 4.1m by 2012e.
118
abc
Financials & valuation: Etihad Etisalat(Mobily)

Overweight
12/2012e 12/2013e
Profit & loss summary (SARm)
Cash flow summary (SARm)
16,013 6,165 -1,810 4,355 -76 4,279 4,279 -67 4,211 4,211
17,930 7,148 -2,049 5,099 -209 4,890 4,890 -122 4,768 4,768
18,645 7,420 -2,221 5,199 -161 5,038 5,038 -126 4,912 4,912
19,116 7,599 -2,385 5,214 -100 5,114 5,114 -128 4,986 4,986
2.8 7.2 2.1 9.1 2.5 5.5 3.7
2.4 5.9 1.9 8.0 2.1 8.8 4.6
2.2 5.4 1.7 7.8 1.8 9.2 5.5
2.0 5.1 1.6 7.7 1.6 10.6 6.4
Issuer information
Share price (SAR) 5,470 -3,288 -3,227 -875 -1,802 2,115 6,839 -3,454 -3,454 -1,400 -1,985 3,362 7,122 -3,580 -3,580 -1,750 -1,792 3,538 7,371 -3,319 -3,319 -2,100 -1,952 4,052 Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
54.75 Target price (SAR) 7020.SE 10,234 40 Saudi Arabia Kunal Bajaj
Bloomberg (Equity) EEC AB Market cap (SARm) 38,325 Enterprise value (SARm) 42200 Sector Wireless Telecoms Contact 9714 5077200
Balance sheet summary (SARm)
Price relative
11,558 12,457 9,415 2,111 33,430 9,814 7,972 5,860 15,580 21,505
11,032 14,388 9,519 2,190 34,939 10,212 6,065 3,875 18,597 22,538
10,506 16,273 9,512 2,075 36,291 10,658 4,158 2,083 21,409 23,558
9,981 17,732 10,394 2,886 38,106 11,078 3,017 131 23,945 24,142
60 55 50 45 40 35 30 25 2009
Etihad Etisalat(Mobily)
60 55 50 45 40 35 30 25 2010 2011 2012

Rel to TADAWUL ALL SHARE INDEX
Source: HSBC

Ratios (%)
22.6 27.5 35.8 40.5 39.7
12.0 16.0 17.1 14.3 13.2
4.0 3.8 1.9 3.0 3.0
2.5 2.4 0.3 1.5 1.5
Per share data (SAR)
0.8 23.2 30.3 13.5 38.5 27.2 81.1 37.6 1.0 93.3
0.8 24.9 27.9 14.7 39.9 28.4 34.1 20.8 0.5 176.5
0.8 24.2 24.6 14.3 39.8 27.9 46.2 9.7 0.3 341.9
0.8 23.5 22.0 13.8 39.7 27.3 76.1 0.5 0.0 5615.1
6.02 6.02 2.00 22.26
6.81 6.81 2.50 26.57
7.02 7.02 3.00 30.58
7.12 7.12 3.50 34.21
119
abc
MTN (Overweight, TP ZAR148)

Mobile connectivity
MTN is well placed in terms of pricing power on mobile data. All the mobile operators in Africa currently offer tiered data plans. Even if a data plan is marketed as an unlimited plan, in reality the plan is not truly unlimited. There is no all you can eat data plan currently, which gives significant pricing power to the mobile operators. The lack of fixed line alternatives necessitates the use of mobile phones and USB modems to access data. Satellite-based broadband is very expensive and is unlikely to be a threat in the African markets. Competition from fixed line operators is limited as these suffer from a lack of network coverage and proper distribution systems. MTN is making attempts to monetise the data growth potential in Africa. Data usage in South Africa has increased 49% since December 2009, with its data revenue increasing 42% y-o-y in H1 2010. Even more impressive is the 58% y-o-y growth in non-SMS data revenue, which accounted for 11% of its South African service revenue. Data usage is increasing across most of its operations and has started contributing to top-line growth. Among its larger operations, Nigeria currently has the lowest proportion of revenue (4.2%) coming from data services, because of a lack of required bandwidth. With the landing of various submarine cables in Africa, we expect the capacity bottleneck to be removed soon. This should, in our view, allow MTN to begin to capture the vast growth potential for data usage in Nigeria. We believe the risks to MTNs pricing power are quite limited and unlikely to be significant in the near to medium term. However, the risks from the entry of a credible new entrant (particularly Bharti) cannot be ignored. As we have witnessed in the Kenyan voice market, Bharti can be quite
aggressive in pricing. If a similar sort of risk materialises in the data services, it would hurt MTN. This is, however, unlikely in the near term, particularly in data segment, as Bharti will need large capex to build sufficient capacity to meet the growing demand.
Mobile applications
So far, MTN has been able to maintain a relatively good grip on content, data applications and VAS thanks to its advantages in terms of familiarity with the local culture. We expect low-cost smartphones based on the Android platform to drive internet usage in Africa. However, language could present a barrier to would-be developed world competitors. Hence, we believe MTN would not be disintermediated from the applications layer. Mobile payment services have achieved considerable success in Africa as most of the countries here have a large unbanked population. MTNs mobile payment service, Mobile Money, is showing a good take-up in countries where it has been launched. For instance, MTN Uganda has almost 16% of its subscriber base already using the service. MTN has c2.2m mobile payment subs across its operations. We reiterate our thematic view (Assessing new growth opportunities, 25 January 2010) that, over the medium term, mobile payments will contribute approximately 6% of revenue at a margin of around 20% for African mobile operators.
Investment thesis
MTNs is a strong growth story in the CEEMEA telecoms universe, as one of the fastest-growing telcos in the region, in our view. We are also bullish on the data usage growth potential in Africa as the affordability barrier and capacity bottlenecks are removed. Despite the increasing competition across its markets, we find it well positioned to compete within the changing competitive environment with a focus on cost management.
120
abc
Financials & valuation: MTN

Overweight
12/2011e 12/2012e
Profit & loss summary (ZARm)
Cash flow summary (ZARm)
111,947 46,063 -14,475 31,588 -2,201 25,773 32,050 -8,612 14,650 22,122
114,097 48,225 -15,332 32,894 -1,993 29,906 33,404 -10,589 16,881 21,297
125,761 53,469 -17,025 36,444 -1,700 34,862 37,247 -11,679 20,417 23,202
134,884 57,624 -18,514 39,110 -1,161 38,067 40,452 -12,677 22,393 25,178
Nigeria EBITDA (ZARm) South Africa EBITDA (ZARm) Ghana EBITDA (ZARm) Iran EBITDA (ZARm)
19,746 10,410 2,566 2,664
19,889 12,063 2,359 3,829
21,838 12,921 2,433 4,426
23,055 13,688 2,464 4,754
39,664 -27,720 -33,192 -3,382 -1,711 7,701
36,778 -19,558 -18,953 -6,313 -10,546 15,980
40,647 -19,871 -19,871 -7,617 -13,159 19,686
44,384 -20,101 -20,101 -9,682 -14,601 23,285
2.4 5.9 3.1 10.5 3.4 3.0 1.5
2.3 5.4 3.0 10.9 3.1 6.2 2.7
1.9 4.6 2.7 10.0 2.7 7.7 3.5
1.7 4.0 2.6 9.2 2.4 9.2 4.8
Balance sheet summary (ZARm)
Issuer information
37,526 67,541 46,024 23,999 156,237 39,094 36,917 12,918 67,866 87,998
35,100 73,784 51,258 34,062 167,432 38,522 36,434 2,372 75,228 87,558
32,715 79,015 66,167 47,221 185,304 41,379 36,434 -10,787 87,479 89,297
30,330 82,988 82,137 61,822 202,980 44,861 36,434 -25,388 98,675 88,772
Share price (ZAR) 125.99 Target price (ZAR) 148.00 Potent'l return (%) 17.5 Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst MTNJ.J 32,520 75 South Africa Herve Drouet Bloomberg (Equity) MTN SJ Market cap (ZARm) 237,432 Enterprise value (ZARm) 258603 Sector Wireless Telecoms Contact 44 20 7991 6827
Price relative
143 143 133 123 113 103 93 83 73 2010
Rel to JSE ALL SHARE
133 123 113 103 93 83 73 2009

MTN Source: HSBC

Ratios (%)
9.2 6.7 3.9 -9.5 8.8
1.9 4.7 4.1 16.0 -3.2
10.2 10.9 10.8 16.6 8.9
7.3 7.8 7.3 9.2 8.5
2011
2012
Per share data (ZAR)
1.2 28.7 30.7 12.6 41.1 28.2 20.9 17.7 0.3 307.0
1.3 28.6 29.8 13.4 42.3 28.8 24.2 2.8 0.0 1550.5
1.4 30.6 28.5 14.4 42.5 29.0 31.4 -11.0 -0.2
1.5 32.5 27.1 14.2 42.7 29.0 49.6 -22.5 -0.4
7.91 11.95 1.92 36.66
9.17 11.57 3.43 40.87
11.09 12.61 4.44 47.53
12.17 13.68 6.08 53.61
121
abc
Mobile Telesystems (Overweight, TP USD29)

MTS recent acquisition of a controlling stake in Comstar has opened up new horizons in the fixed line space. Comstar has been seen historically as a local player in Moscow, but MTSs nationwide presence gives access to the regions of Russia. We expect Comstar to benefit from rolling out its broadband offering to existing MTS mobile subscribers. It is well positioned in terms of pricing power on data services as tariffs are data tiered. There are no all you can eat data packages. There are some substitutes to deliver broadband but those are relatively limited. There are some small regional cable operators but these have limited scale and financial power to upgrade their cable network. Substitution from fixed regional operators is also limited at the moment. Absence of regulations on net neutrality is positive for incumbents. Local loop unbundling by altnets is yet to be allowed in Russia. There have also been cases of private reselling of connectivity and illegal sharing. The impact, while potentially significant, would be mitigated as long as data tariffs remain tiered. Here we would expect the potential issues with illegal sharing to gradually fade as the demand for higher speed takes hold.
In Q3, VAS (SMS included) amounted to 20.5% of revenue in Russia (increased 2.1ppt y-o-y), helped by the expansion of its 3G network and fast growing data traffic. We believe pricing power is unlikely to be altered significantly in the medium term as the competitive environment has remained quite rational in Russia.

It is plausible that certain OTT competitors will pose a threat to incumbent IPTV services and could also challenge well-established brands in the pay-TV market. However, factors like the complexities of rights arrangements, power of its existing brands and absence of net neutrality regulations would provide significant pricing power and advantage to incumbents like MTS.
Mobile applications
MTS has been able to maintain a relatively good grip on content, data applications and VAS thanks to its advantages in terms of familiarity with the local culture. Data content already represents c5% of mobile revenue in Russia, almost matching the revenue from data traffic excluding SMS. We expect low-cost smartphones based on the Android platform to drive internet usage in Africa. However, language could present a barrier to would-be developed world competitors. Hence, we believe MTS would not be dis-intermediated from the applications layer.
Mobile connectivity
Growing data demand is pushing mobile connectivity with the number of 3G USB modems having now overtaken fixed line broadband subscribers. MTS is well positioned in terms of pricing power on mobile data services. In Russia and CIS countries, tariffs are data tiered. Some packages are marketed as unlimited but in reality are capacity-limited. In local currency terms, data traffic revenue jumped 69% y-o-y in 9M 2010, while total VAS revenue increased c26% y-o-y (in local currency).
Investment thesis
We are bullish on the data usage growth potential in Russia. Usage trends are reassuring, with ARPU growth in local currency terms, driven by higher voice and data usage. The mobile operators in Ukraine have shifted their focus to profitability, while making upward adjustments to tariffs. The recent trends indicate an improving macroeconomic outlook in Russia and CIS, increased usage and recovery in Ukraine.
122
abc
Financials & valuation: Mobile Telesystems

Overweight
12/2011e 12/2012e
9,867 4,473 -1,930 2,543 -467 1,486 2,162 -505 1,001 1,685
11,170 4,932 -1,964 2,968 -722 2,312 2,271 -517 1,627 1,580
12,155 5,402 -2,110 3,292 -572 2,720 2,720 -626 1,981 1,981
13,109 5,851 -2,201 3,650 -528 3,122 3,122 -718 2,312 2,312
Russia subscribers (000s) Russia ARPU (USD) Ukraine subscribers (000s) Ukraine ARPU (USD)
69,342 7.8 17,564 4.7
70,192 8.4 18,313 4.9
72,214 8.9 19,206 5.2
73,982 9.4 19,979 5.5
3,596 -1,942 -3,718 -1,262 1,739 1,578
3,697 -1,989 -2,042 -496 -622 1,698
4,269 -2,674 -2,841 -958 -469 1,486
4,644 -2,687 -2,687 -1,188 -768 1,864
2.5 5.6 2.7 11.1 5.6 8.2 5.1
2.2 4.9 2.5 12.0 5.3 8.8 5.1
2.0 4.4 2.3 9.6 4.3 7.7 6.3
1.8 4.0 2.2 8.2 3.6 9.6 7.3
Issuer information
2,236 7,749 4,390 2,728 15,749 2,256 8,349 5,621 3,330 9,391
2,401 8,441 3,055 960 14,390 3,011 5,959 4,999 3,601 9,926
2,401 9,172 3,243 1,000 15,309 3,452 5,530 4,530 4,394 10,363
2,401 9,658 3,386 1,000 15,938 3,833 4,762 3,762 5,318 10,611
Share price (USD)
19.77 Target price (USD)
Reuters (Equity) MBT.N Market cap (USDm) 19,704 Free float (%) 44 Country Russian Federation Analyst Herve Drouet
Bloomberg (Equity) MBT US Market cap (USDm) 19,704 Enterprise value (USDm) 24327 Sector Diversified Telecoms Contact 44 20 7991 6827
Price relative
29 29 24 19 14 9 4 2010
Rel to RTS INDEX
24 19 14

Ratios (%)
-17.1 -23.5 -30.3 -49.3 -32.2
13.2 10.3 16.7 55.6 -7.7
8.8 9.5 10.9 17.6 25.4
7.8 8.3 10.9 14.8 16.7
9 4 2009
Mobile Telesystems Source: HSBC
2011
2012
1.1 21.9 40.9 9.3 45.3 25.8 9.6 129.2 1.3 64.0
1.2 23.7 45.6 16.0 44.2 26.6 6.8 108.3 1.0 74.0
1.2 25.0 49.6 17.2 44.4 27.1 9.4 82.1 0.8 94.2
1.2 26.8 47.6 18.1 44.6 27.8 11.1 57.5 0.6 123.5
1.06 1.79 1.00 3.53
1.70 1.65 1.00 3.76
2.07 2.07 1.24 4.58
2.41 2.41 1.45 5.55
123
abc
NTT DoCoMo (Overweight, TP JPY172,000)

Mobile connectivity
We forecast smartphones as a proportion of DoCoMo handset users to grow from 2.3% at end 2010 to 14% by end 2011, with these users tending to spend more as they begin to measure data usage in megabytes, rather than kilobytes. A DoCoMo study of Sony Ericsson Xperia users found they spent JPY2,100 more per month after upgrading while this amount will decline, the trend is indicative. A further positive for DoCoMo is its network strength. It has invested consistently in its network, and its strength here is in strong contrast to SoftBank, which is struggling to manage strong data demand from its iPhone users. Expectations regarding data tariff levels are ingrained: unlike in the US and Europe, c40% of Japanese users already subscribe to some form of tiered data plan. This makes it difficult for the Japanese operators to shift to a pay-per-usage model at the high-end at least for 3G. However, all operators have flagged the migration to LTE as a chance to change this. Subscribers to DoCoMos Xi (pronounced Crossy) LTE service launched in December 2010 will pay JPY2,500 (USD30) for each 2GB of usage beyond the 5GB tier (which costs a hefty JPY8,700, or USD104).
(with c1m titles) and a book store (which had recorded 1.5m downloads by December 2010). Part of the delay in launching Android-based phones for both DoCoMo and KDDI has been a result of their desire to integrate native Japanese applications (such as e-wallet) services, with the Android platform as shipped by the vendor.
Investment thesis
In our view, NTT DoCoMo is well-placed to benefit from surging demand for smartphones in Japan. This is coming from a very low base: just c2.3% of DoCoMo users had smartphones at end December 2010 a result of the dominance of domestic vendors, who (as in Korea) have been late to produce smartphones. As a result, DoCoMo performed poorly relative to SoftBank in 2009 and 2010: as the sole supplier of the iPhone this operator had c80% of market smartphone subscribers at September 2010. With the launch of Android-based smartphones in the last quarter of 2010, NTT DoCoMo has seen strong sales. It raised its smartphone sales target to 2.5m for FY March 2011 (originally set at 1.3m), and saw strong demand in December and January. We see its network strength as a strong positive: users are increasingly aware of the disparity between different networks, a strong benefit to DoCoMo as the breadth and depth of its smartphone portfolio improves. While we see little strategic benefit from DoCoMos early launch of LTE, we see minimal downside: unlike its early move to 3G, DoCoMo will spend a relatively low proportion of its capex on LTE. We project capex as a proportion of sales to remain constant at c15%. NTT is our top pick in the Japanese telecoms sector.
Mobile applications
Each of the Japanese operators is focusing on building a local suite of applications and services that ensure continuity in the smartphone era with services their customers were using before, on more proprietary platforms such as DoCoMos iMode. Within the DoCoMo market subscribers can choose between an app store, a music store
124
abc
Financials & valuation: NTT DoCoMo Inc.

Financial statements Year to Profit & loss summary (JPYb) 03/2010a 03/2011e 03/2012e 03/2013e Valuation data Year to 03/2010a 03/2011e
Overweight
03/2012e 03/2013e
Cash flow summary (JPYb)
4,284 1,573 -735 838 -4 839 863 -338 499 507
4,249 1,614 -731 883 -4 874 931 -355 517 548
4,354 1,652 -716 937 -4 929 969 -379 548 570
4,595 1,730 -707 1,023 -3 1,016 1,056 -416 597 621
1.4 3.9 1.6 12.6 1.4 8.0 3.4
1.4 3.8 1.6 12.2 1.3 8.9 3.4
1.3 3.5 1.5 11.4 1.2 9.9 3.6
1.2 3.1 1.4 10.5 1.1 9.8 3.7
Issuer information
Share price (JPY) 151100 Target price (JPY) 172000 Potent'l return (%) 17.2 1,183 -726 -1,164 -209 -36 488 990 -715 -722 -216 45 539 1,251 -665 -665 -220 -354 601 1,295 -700 -700 -224 -372 598 Reuters (Equity) 9437.T Market cap (USDm) 79,358 Free float (%) 35 Country Japan Analyst Neale Anderson Bloomberg (Equity) 9437 JP Market cap (JPYb) 6,617 Enterprise value (JPYb) 6121 Sector Wireless Telecoms Contact +852 2996 6716
Balance sheet summary (JPYb)
Price relative
198 3,236 2,061 761 6,757 1,007 762 1 4,636 3,728
209 3,214 2,220 773 6,895 944 820 47 4,801 3,926
209 3,204 2,597 1,142 7,258 969 836 -307 5,113 3,899
209 3,237 3,005 1,514 7,694 1,036 836 -679 5,477 3,900
181668 171668 161668 151668 141668 131668 121668 111668 2009

NTT DoCoMo Inc.
181668 171668 161668 151668 141668 131668 121668 111668 2010

Rel to TOPIX INDEX
2011
2012
Source: HSBC

Ratios (%)
-3.7 -6.3 0.9 7.6 7.2
-0.8 2.6 5.4 4.1 4.0
2.5 2.4 6.1 6.3 6.2
5.5 4.7 9.2 9.4 9.0
Per share data (JPY)
1.2 14.2 11.3 7.6 36.7 19.6 417.0 0.0 0.0 94625.4
1.1 14.4 11.6 7.6 38.0 20.8 438.6 1.0 0.0 2121.9
1.1 14.7 11.5 7.8 38.0 21.5 429.3 -6.0 -0.2
1.2 16.1 11.7 8.1 37.7 22.3 507.4 -12.3 -0.4
11,947 11,947 5,200 111,063
12,422 12,422 5,200 115,404
13,197 13,197 5,400 123,118
14,383 14,383 5,600 131,902
125
abc
Portugal Telecom (Neutral, TP EUR9)

PT currently covers 1m homes (c20%) with its FTTH and has plans to raise this to 1.6m homes in the medium term. In our view, the regulators stance on fibre is beneficial for PT as it aims to ensure transparent legislation that promotes investment in NGA, while at the same time safeguarding a return for operators and conditions for sustainable competition. On the net neutrality side, regulator Anacom has not expressed any views in public, but we believe it is likely to follow the European Union framework; considering the EUs pragmatic stance this should be favourable for PT.

PT has rapidly expanded its market share to currently c27% of the Portuguese pay-TV market from a greenfield operation started in H2 2007. PT provides TV services through a mix of IPTV and DTH based platforms. PTs TV customers constitute almost 80% of its broadband subscriber base, which is one of the highest among the European incumbents. Success of its TV strategy is also reflected in the top-line performance of its fixed division; which is likely to post positive revenue growth two years in a row.
Mobile applications
PT has made some attempts to enter the application space. First, TMN has launched Meo mobile, which makes available 40 TV channels with innovative features such as wireless remote recording and sms alerts. Second, TMN has launched its own application store with sports, news, entertainment, games, books and utility applications available to its customers. This application store leverages on PTs portal Sapo, the most popular portal in Portugal. Third, PT has introduced an aggregation service that enables access to multiple personal accounts, aggregation of social network accounts in a single place, simultaneous posts in multiple accounts and sharing of photos and videos.
Mobile connectivity
PT has been investing in its networks with a mobile capex/sales ratio of c12% over the past four years and has the best 3G coverage of 93% in Portugal vs 92% and 86% for the second and third operators, respectively. TMN experienced a relatively rapid take-up of data services, which can be gauged from the fact that TMNs mobile broadband now has 3x the number of broadband customers it had two years back. Anticipating the rise of data traffic, PT is conducting trials on high speed LTE and has connected 85% of its cell sites with fibre. On mobile data pricing, management has migrated to tiered data plans. TMN has some basic data plans with data cap at 600MB per month but it has continued with its unlimited data plans priced at EUR100/month (it comes with a voice and text bundle with a fair usage policy of 5,000 voice minutes, 5,000 text message, unlimited WiFi and 5GB internet). On the iPad it has an unlimited data plan for a price of EUR29.9.
Investment thesis
PTs investment in FTTH, investment in fibre backhaul to support mobile traffic and its successful IPTV strategy remain the key positives, but, in our opinion, are fully priced in at the current trading price. PT is currently trading at a 10% premium to the 2011e sector EV/EBITDA multiple. Hence, we are Neutral on Portugal Telecom with a EUR9 price target (including EUR0.65 of expected special dividend).
126
abc
Financials & valuation: Portugal Telecom

Neutral
12/2011e 12/2012e
6,785 2,502 -1,527 975 -302 1,021 951 -233 685 624
5,585 2,051 -1,254 797 -245 5,993 797 -148 5,699 480
5,933 2,275 -1,071 1,204 -186 1,175 1,267 -299 750 842
6,766 2,559 -1,128 1,431 -244 1,353 1,431 -343 865 942
Domestic fixed Domestic mobile Vivo Telemar Others Group revenues
1,948 1,518 3,138 0 181 6,785
1,949 1,399 1,885 0 352 5,585
1,969 1,337 0 2,234 392 5,933
1,979 1,316 0 3,067 404 6,766
1,612 -1,268 -871 -504 -25 706
1,548 -967 4,462 -1,380 -5,157 700
1,719 -989 -2,664 -1,139 4,132 801
1,678 -1,051 -1,051 -569 -58 797
2.3 6.3 1.9 12.0 5.7 7.0 6.7
1.9 5.1 1.8 15.7 2.1 7.0 26.8
2.5 6.4 1.5 8.9 2.0 8.0 7.6
2.1 5.6 1.5 8.0 1.9 8.0 7.8
4,047 4,862 3,699 1,500 14,831 2,904 7,046 5,547 1,318 8,204
1,080 3,635 4,896 1,500 11,103 2,237 1,890 390 3,594 5,874
4,436 5,641 4,043 1,500 16,393 2,751 6,022 4,522 3,775 9,869
4,436 5,624 4,168 1,500 16,556 2,812 5,964 4,464 4,053 9,916
Issuer information
8.58 Target price (EUR) PTC.LS 10,423 90 Portugal Luigi Minerva
4.9
Bloomberg (Equity) PTC PL Market cap (EURm) 7,691 Enterprise value (EURm) 10464 Sector Diversified Telecoms Contact 44 20 7991 6928
11 10 9 8 7 11 10 9 8 7 6 5 4 2010
Rel to PSI 20

Ratios (%)
1.0 0.9 -16.5 10.0 -7.3
-17.7 -18.0 -18.2 487.2 -23.1
6.2 10.9 51.0 -80.4 75.5
14.0 12.5 18.8 15.2 11.9
6 5 4 2009
Portugal Telecom
2011
2012
0.9 10.8 80.4 7.3 36.9 14.4 8.3 232.6 2.2 29.1
0.8 9.7 19.5 47.1 36.7 14.3 8.4 10.4 0.2 397.2
0.8 12.4 22.9 7.7 38.4 20.3 12.2 113.2 2.0 38.0
0.7 11.5 24.1 7.5 37.8 21.2 10.5 103.1 1.7 37.6
Source: HSBC
0.78 0.71 0.57 1.50
6.51 0.55 2.30 4.10
0.86 0.96 0.65 4.31
0.99 1.08 0.67 4.63
127
abc
Saudi Telecom Company (Overweight, TP SAR48)

STC is the incumbent telecom service provider in Saudi Arabia, currently operating in a duopoly environment in the fixed line space. However, while the second fixed line operator, Etihad Etheeb, has recently started operations, it has not invested in its network because of balance sheet constraints, giving STC still 98% market share in fixed voice. There are no substitutes to deliver broadband in Saudi Arabia on fixed line ( no cable, altnets etc) and mobile broadband (dongles, USBs) are the only alternative to fixed line broadband for the customers. VOIP is banned in Saudi Arabia, meaning internet is not a direct substitute for voice services. Absence of regulations on net neutrality is positive for incumbents. Local loop unbundling by altnets is yet to be allowed in Saudi Arabia. STC currently has around 85,000 km of fibre cable in the country and the DSL penetration is only 32% of households. Pricing is not a differentiating factor, with a 2Mbps fixed broadband for unlimited data plan cost of around SAR221, which is 10% higher than mobile broadband for 2GB usage cap.
already started to bundle its services through triple play offerings of landline, DSL and IPTV.
Mobile applications
STC has been able to maintain a relatively good grip on content, data applications and VAS thanks to its advantages in terms of familiarity with the local culture .While there has been no separate disclosures from Saudi companies on data content revenues, STC has started to invest in contents and applications for its 3.5G services on mobile. In October 2008, STC, in association with Astro All Asia Networks of Malaysia and Saudi Research and Marketing Group (SRMG), set up a new mobile content services company with capital of USD74.8m, with 51% ownership by STC.
Investment thesis
We are bullish on broadband prospects in Saudi Arabia as long-term healthy demand for broadband services is be supported by Saudis attractive demographics, favourable regulatory environment, lack of entertainment options and absence of alternative technologies (like cable). We expect STCs data revenue as a percentage of domestic revenues to grow from 14.8% currently to 19.4% within the next 5 years. We have little doubt that the NGN will create new growth opportunities for STC from about 2011 onwards. Fixed-line technology is much better suited than mobile technology for the transmission of data at very high speeds, and hence it will always be superior to mobile for high-bandwidth applications such as video streaming and television. STC is the only integrated operator in Saudi Arabia and thus should be able to offer such applications in bulk; this is one reason why STC should remain the telecoms provider of choice for large business and public-sector customers and for very high end consumers.
Mobile connectivity
There are around 4m mobile broadband users in Saudi Arabia compared with 1.5m fixed broadband users. STC is well positioned in terms of pricing power on mobile data services, as is Mobily. Pricing has remained stable in Saudi Arabia for broadband services, unlike other markets like Egypt.

The Saudi Arabian media market is unregulated and the mass medium of content distribution is satellite. Around 52% of households in Saudi with a television are on the satellite platform, with the rest on terrestrial DTT. STC formally launched IPTV in August 2010 as InVision and has
128
abc
Financials & valuation: Saudi Telecom Company

Overweight
12/2012e 12/2013e
Profit & loss summary (SARm)
Cash flow summary (SARm)
51,787 19,625 -8,645 10,981 -1,789 10,983 9,708 -884 9,496 8,323
54,477 20,631 -9,137 11,494 -1,009 10,486 10,486 -844 9,157 9,157
54,975 21,104 -9,239 11,865 -954 10,911 10,911 -941 9,484 9,484
55,796 21,479 -9,510 11,969 -892 11,077 11,077 -966 9,626 9,626
2.1 5.5 1.3 9.7 1.8 7.3 7.4
2.0 5.2 1.3 8.8 1.7 9.4 7.4
1.9 5.0 1.3 8.5 1.6 10.8 9.4
1.8 4.7 1.2 8.4 1.6 12.4 9.6
Issuer information
Share price (SAR) 18,780 -10,674 -13,046 -6,109 704 6,185 18,757 -10,791 -10,791 -6,000 -1,966 7,976 19,260 -10,043 -10,043 -7,587 -1,629 9,119 19,630 -9,178 -9,178 -7,701 -2,752 10,427 Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
40.30 Target price (SAR) 7010.SE 21,523 30 Saudi Arabia Kunal Bajaj
Bloomberg (Equity) STC AB Market cap (SARm) 80,600 Enterprise value (SARm) 106771 Sector DIVERSIFIED TELECOMS Contact 9714 5077200
Balance sheet summary (SARm)
Price relative
31,806 55,135 18,666 5,904 110,709 18,190 30,188 24,284 44,998 81,513
30,676 57,919 18,716 5,904 112,412 18,218 28,222 22,318 48,156 83,189
29,546 59,854 18,766 5,904 113,267 18,319 26,593 20,688 50,052 83,942
28,416 60,652 18,848 5,904 113,017 18,411 23,841 17,937 51,978 83,601
61 56 51 46 41 36 31 26 2009 2010 2011

Rel to TADAWUL ALL SHARE INDEX Saudi Telecom Company
61 56 51 46 41 36 31 26 2012
Source: HSBC

Ratios (%)
2.0 -4.8 -14.3 -10.2 -20.6
5.2 5.1 4.7 -4.5 10.0
0.9 2.3 3.2 4.1 3.6
1.5 1.8 0.9 1.5 1.5
Per share data (SAR)
0.7 14.1 19.1 10.1 37.9 21.2 11.0 45.4 1.2 77.3
0.7 14.1 19.7 9.6 37.9 21.1 20.5 39.1 1.1 84.0
0.7 14.2 19.3 9.7 38.4 21.6 22.1 34.8 1.0 93.1
0.7 14.3 18.9 9.8 38.5 21.5 24.1 29.0 0.8 109.4
4.75 4.16 3.00 22.50
4.58 4.58 3.00 24.08
4.74 4.74 3.79 25.03
4.81 4.81 3.85 25.99
129
abc
Sprint Nextel (Neutral (V), TP USD5)

Sprint Nextels only fixed-line asset is its longdistance network, having spun off its local access division in 2006 (as Embarq) which is now owned by CenturyLink. Understandably therefore Sprint Nextel has not been particularly vocal on the issue of net neutrality, releasing only a brief statement regarding the FCCs new policy agreed in December 2010 which maintained strict net neutrality principles for fixed networks while permitting reasonable traffic management on mobile networks. Sprint Nextel commented that the FCC had recognized the differences between fixed and mobile networks and had resolved a fair and balanced approach to a difficult issue.
We expect the 3GPPs LTE technology to form the core of Network Vision. As a challenger for market share and with few capacity pressures on a series of relatively underoccupied networks, Sprint Nextel does not have the same imperative as AT&T to ration consumption through tiered data pricing although periodically the firm notes that it does not rule out this option in future. Sprint Nextel does not have a public WiFi network (indeed some might quip this is one of the few network technologies it doesnt operate), although via its 54% stake in Clearwire, Sprint does have substantial amounts of spare WiMAX capacity: Clearwire has nearly 100MHz of spectrum in its major markets and only around one million customers. Such abundance of capacity does not encourage pricing power, we believe. Indeed, Sprint Nextel, having ample capacity as well as a desire to regain market share lost over the past four years, makes for a potentially disruptive combination, which detracts from the appeal of the US wireless market, in our view.
Mobile connectivity
Sprint Nextels network is under considerably less pressure than that of AT&T, and potentially that of Verizon Wireless too (as the latter expects to roughly double its smartphone base this year on the back of new CDMA iPhone sales). The reason for Sprints greater breathing room is hardly positive though as the company has been losing mobile subscribers consistently for more than four years now: Q4 2010 saw the first positive direct postpaid net additions (+58,000) since Q2 2006. A dwindling subscriber base has enabled the company to reduce mobile capex to very low levels this averaged 5% of divisional sales 2008-2010. Part of the companys problems may be ascribed, we believe, to its running multiple network technologies CDMA, iDEN and (via Clearwire) WiMAX in parallel. Although questions still remain on further funding of Clearwire, Sprint Nextel has announced Network Vision, a network modernization frame agreement with Alcatel-Lucent, Ericsson and Samsung to rationalise the companys technology roadmap.

As Sprint Nextel lacks a fixed-line local access presence the firm does not have plans for IPTV.
Mobile applications
We see little prospect of Sprint Nextel forcing change to the status quo in the mobile applications value chain (given we believe even its greatly more powerful rivals, Verizon Wireless and AT&T, are equally unable to resist the dominance in this space from Apple and Google).
Investment thesis
We have a Neutral (V) rating on Sprint Nextel. Although the company has reported improving metrics over the past couple of quarters, we believe it still remains vulnerable to increased competitive intensity from larger rivals, which underpins our cautious stance on the stock.
130
abc
Financials & valuation: Sprint Nextel Corp

Neutral (V)
12/2012e 12/2013e
32,563 5,633 -6,537 -904 -1,367 -2,271 -2,276 -166 -2,437 -2,276
32,348 5,917 -5,548 370 -1,439 -1,069 -1,069 99 -970 -1,069
32,822 6,188 -4,842 1,347 -1,360 -14 -14 5 -8 -14
33,553 6,367 -4,365 2,002 -1,209 793 793 -301 492 492
Reported revenues Reported EBITDA Reported EBIT Reported PBT Reported EPS
32,563 5,633 -821 -2,188 -1
32,348 5,917 370 -1,069 0
32,822 6,188 1,347 -14 0
33,553 6,367 2,002 793 0
4,815 -1,935 -2,556 0 2,237 2,145
5,035 -2,927 -2,927 0 -2,157 1,591
4,781 -3,221 -3,221 0 -1,559 1,465
5,064 -3,332 -3,332 0 -1,732 1,426
1.0 5.8 0.8 NA 0.9 16.0 0.0
0.9 5.2 0.8 NA 0.9 11.8 0.0
0.9 4.7 0.8 NA 0.9 10.9 0.0
0.8 4.3 0.8 26.7 0.9 10.6 0.0
22,704 15,214 9,880 817 51,654 6,235 20,191 19,374 14,546 40,746
22,465 13,121 13,488 4,432 52,931 6,095 21,649 17,217 14,384 38,548
22,182 11,784 12,369 3,241 50,191 6,113 18,899 15,658 14,376 36,981
21,937 10,997 12,737 3,500 49,527 6,431 17,426 13,926 14,867 35,740
Issuer information
4.60 Target price (USD) S.N 13,579 100 United States Richard Dineen
8.7
Bloomberg (Equity) S US Market cap (USDm) 13,579 Enterprise value (USDm) 30661 Sector Diversified Telecoms Contact 1 212 525 6707
7 6 5 4 7 6 5 4 3 2 1 2010
Rel to S&P 500

Ratios (%)
0.9 -6.4 NA NA NA
-0.7 5.0 NA NA NA
1.5 4.6 264.2 NA NA
2.2 2.9 48.7 NA NA
3 2 1 2009
Sprint Nextel Corp
2011
2012
0.8 1.4 -13.9 -1.7 17.3 -2.8 4.1 133.2 3.4 24.9
0.8 2.0 -7.4 1.0 18.3 1.1 4.1 119.7 2.9 29.2
0.9 4.3 -0.1 2.9 18.9 4.1 4.5 108.9 2.5 30.5
0.9 3.8 3.4 2.6 19.0 6.0 5.3 93.7 2.2 36.4
Source: HSBC
-0.85 -0.80 0.00 5.10
-0.34 -0.37 0.00 5.04
0.00 0.00 0.00 5.04
0.17 0.17 0.00 5.21
131
abc
Swisscom (Overweight, TP CHF470)

Swissscom, with VDSL coverage of 78% and ongoing investments in FTTH, has a strong domestic position in fixed-line services. That means alternative operators competing on ULL are at a disadvantage. The regulatory environment in Switzerland is favourable to Swisscom, because fibre networks (VDSL or FTTH) are not included in the current regulatory network and are not expected to be regulated anytime soon. In 2009, the Swiss Parliament asked the Federal Council to assess the level of competition in the telecom market. In a report published on 17 September 2010, the Council answered that there was no urgent need to change the current telecoms law as it would discourage ongoing investments. Also, Swisscom is laying multi-fibre rather than single fibre, which reduces the chances of regulation.

Swisscom started providing IPTV services in 2007 to fight back against Cablecom. It had secured around 10% of the TV market by the end of Q3 2010 by which point, 23% of its broadband lines were taking an IPTV service. Swisscoms triple-play package starts at CHF99 per month and includes line rental, free voice calls to the Swisscom fixed-line and mobile networks, 10Mbps broadband and a TV service with over 160 channels. The incumbent is competing against cable operator Cablecoms triple-play package (CHF75 per month for fixed-line calls, 20Mbps broadband and more than 120 TV channels). Swisscom has an agreement with OTT provider Zattoo that lets subscribers of Swisscom VDSL watch some extra channels in the player. Those channels have a high-quality signal that shows a sharp picture even if one enlarges the player window. The benefit to Swisscom is that it encourages people to upgrade to fibre-based broadband.
Mobile connectivity
Swisscom has been continually investing in its networks (fixed and mobile) with average domestic capex/sales ratio of 14% over 2006-10e. Hence, even with rapid growth in data revenues in 2010 (in 9M 2010 non-messaging data revenues grew by 36% y-o-y), Swisscoms networks do not face congestion. Swisscom has proactively upgraded its network to HSDPA and has complementary WiFi coverage that helps to offload traffic in high-density areas. In its Q2 2010 conference call, the company said it would ensure that all Apple devices log in automatically to wireless hotspots when in theyre in a WiFi zone. Swisscom has three sets of data plans catering to the different class of data users its basic plan is 250MB for CHF35, then it has a 1GB plan for CHF55 and an unlimited voice and data plan (speeds reduced after 2GB of usage) for CHF169.
Mobile applications
Swisscom promotes the use of apps with its Swiscom Appvisor to help clients select applications. Swisscom also offers live TV (Swisscom TV air), news (Swisscoms own portal) and location services. As the application layer is properly served, clients are encouraged to use their mobile more, enabling Swisscom to monetise it.
Investment thesis
Our Overweight rating on Swisscom is primarily driven by its strong domestic position in both fixed and mobile services, which we believe the company can maintain over the long term as it is continually investing in its networks. The regulatory environment is also favourable for Swisscom. On the mobile side, Swisscom has the right connectivity and application strategy to monetise the capacity scarcity.
132
abc
Financials & valuation: Swisscom

Financial statements Year to 12/2009a 12/2010e 12/2011e 12/2012e Key forecast drivers Year to
Overweight
12/2009a 12/2010e 12/2011e 12/2012e
Profit & loss summary (CHFm)
Cash flow summary (CHFm)
12,001 4,666 -1,988 2,678 -336 2,385 2,385 -460 1,928 1,896
12,008 4,696 -1,977 2,719 -346 2,407 2,509 -519 1,898 1,999
11,951 4,741 -1,949 2,792 -272 2,553 2,557 -530 2,018 2,022
12,243 4,840 -1,972 2,867 -255 2,646 2,646 -549 2,093 2,093
Swisscom Revenues Swisscom EBITDA adjusted Swisscom EBIT adjusted Swisscom Net income adjusted Swisscom FCF definition (OpFCF)
12,001 4,666 2,678 1,928 2,669
12,008 4,696 2,719 1,898 2,611
11,951 4,741 2,792 2,022 2,844
12,243 4,840 2,867 2,093 2,919
4,195 -1,987 -1,189 -984 -928 1,892
3,737 -1,876 -2,148 -1,036 -230 1,946
3,935 -1,834 -1,834 -1,243 -857 2,110
4,034 -1,849 -1,849 -1,347 -838 2,183
2.6 6.7 1.9 11.8 3.5 8.5 4.6
2.6 6.6 2.0 11.2 4.2 8.8 5.6
2.5 6.3 2.0 11.0 3.8 9.5 6.0
2.4 6.0 2.0 10.7 3.4 9.9 6.5
Balance sheet summary (CHFm)
8,979 8,044 4,154 1,078 21,960 3,371 10,010 8,932 6,409 16,728
8,581 8,035 4,361 1,116 21,856 4,587 9,818 8,702 5,270 15,274
8,429 8,072 5,216 1,974 22,578 4,698 9,818 7,844 5,945 15,045
8,277 8,101 4,057 800 21,279 4,830 7,807 7,007 6,588 14,804
Issuer information
Share price (CHF) 430.90 Target price (CHF) 470.00 Potent'l return (%) Reuters (Equity) SCMN.VX Market cap (USDm) 22,966 Free float (%) 42 Country Switzerland Analyst Nicolas Cote-Colisson
9.1
Bloomberg (Equity) SCMN VX Market cap (CHFm) 22,321 Enterprise value (CHFm) 30837 Sector Diversified Telecoms Contact 44 20 7991 6826
436 416 396 376 356 336 316 296 276 2009
Swisscom

Ratios (%)
-1.6 -2.6 1.4 8.5 -1.3
0.1 0.7 1.5 0.9 5.4
-0.5 1.0 2.7 6.1 1.1
2.4 2.1 2.7 3.7 3.5
436 416 396 376 356 336 316 296 276 2010
Rel to SMI
2011
2012
Per share data (CHF)
0.7 13.7 32.1 10.3 38.9 22.3 13.9 132.8 1.9 47.0
0.8 14.2 34.2 10.0 39.1 22.6 13.6 162.9 1.9 43.0
0.8 15.4 36.1 10.2 39.7 23.4 17.4 130.5 1.7 50.2
0.8 16.0 33.4 10.6 39.5 23.4 19.0 105.3 1.4 57.6
Source: HSBC
37.22 36.61 20.00 123.74
36.65 38.60 24.00 101.75
38.96 39.04 26.00 114.78
40.40 40.40 28.00 127.19
133
abc
TDC (Overweight, TP DKK58)

TDC owns Denmarks biggest cable network, covering about 57% of Danish homes under the YouSee brand. It has invested substantially in recent years in NGA infrastructure (14% capex/sales on average in 2007-10) both on the cable as well as on the telecom side. It also acquired fibre network operator DONG in the Copenhagen area. Its main competition arises from utilities building fibre networks, which currently cover almost a quarter of Danish homes. Those utility fibre networks are technologically competitive but typically lack scale and knowhow in increasingly complex multi-play telco products. The Danish regulators key line of action at the moment is to force TDC to open its YouSee cable network on a wholesale bitstream basis to competitors.

Given its cable ownership, TDC is the leading pay-TV operator in Denmark. Sales of triple-play to its telecom subscriber base have been modest. It is primarily used as a retention tool under the pre-requisite not to cannibalise its YouSee cable subscriber base. Video on demand is offered on both platforms, and only towards the end of 2010 started to gain momentum, with VoD sessions doubling in Q4 sequentially.
Mobile applications
TDC is not part of WAC, the Wholesale Application Community. Given the small size if the Danish market, it appears sensible not to try and compete in the application market with global internet giants, but rather to focus on providing and monetising quality connectivity.
Investment thesis
We like TDC as a pure play on the stable Danish market with no M&A risk, high visibility on cash generation and use. We believe TDC has invested sensibly in past years. TDC also enjoys the advantage of owning an upgraded cable network. TDC compares favourably to its peers in diversifying its brand portfolio, having acquired and established several no-frill, discount and premium brands around its core TDC brand. TDC has also reorganised into segments grouped by customers rather than products. We have an Overweight rating on TDC, with a target price of DKK58. A policy of 80% to 85% payout of FCFE leaves little room for valuedestructive use of cash and the historically high capex/sales ratio provides flexibility to protect short-term FCF should the economy slow.
Mobile connectivity
We see TDC as well-positioned in data pricing strategy with tiered data tariffs in place, albeit in an especially challenging market with contract periods limited to six months, which stimulates churn, and aggressive MVNO activity. TDC is well-advanced in rolling out HSPA+ covering 15% of its network. After securing 2.5GHz of spectrum, TDC in April 2010 started to roll out LTE, with Ericsson as primary vendor. TDC offers mobile data in tiered tariffs with 1GB priced at DKK99 (EUR14) and 1GB at DKK199 (EUR28).
134
abc
Financials & valuation: TDC

Overweight
12/2012e 12/2013e
Profit & loss summary (DKKm)
Cash flow summary (DKKm)
26,167 9,425 -5,356 4,069 -1,591 2,586 2,759 -782 1,804 2,069
26,143 9,794 -5,222 4,572 -1,040 3,539 3,632 -885 2,654 2,724
26,200 9,873 -5,077 4,796 -1,035 3,767 3,860 -942 2,825 2,895
26,307 9,945 -4,945 5,000 -1,033 3,973 4,066 -993 2,980 3,049
EBITDA company EBITDA HSBC EBITDA reported FCF HSBC TMT FCFE HSBC FCFE company
10,772 9,600 9,425 2,897 3,977 4,515
10,966 9,794 9,794 3,787 4,500 4,562
11,045 9,873 9,873 4,389 4,486 4,581
11,117 9,945 9,945 4,386 4,462 4,590
7,265 -3,646 -3,923 -70 -10,822 3,977
7,408 -3,464 -3,464 -1,882 -2,063 4,500
7,445 -3,485 -3,485 -3,771 -189 4,486
7,475 -3,509 -3,509 -3,786 -183 4,410
2.1 5.8 1.2 18.2 1.8 12.4 0.0
2.0 5.4 1.2 13.8 1.6 14.0 10.0
2.0 5.3 1.2 13.0 1.7 14.0 10.0
2.0 5.3 1.2 12.4 1.7 13.7 10.0
Balance sheet summary (DKKm)
34,799 15,531 6,248 831 64,786 11,783 23,644 22,813 20,855 43,964
33,239 15,426 8,316 2,894 65,195 9,383 23,644 20,750 23,509 44,704
31,808 15,358 8,531 3,083 63,917 8,904 23,644 20,561 22,555 43,710
30,501 15,322 5,996 500 60,045 8,449 20,879 20,379 21,748 42,870
Issuer information
Share price (DKK)
46.31 Target price (DKK)
Reuters (Equity) TDC.CO Market cap (USDm) 6,851 Free float (%) 33 Country Denmark Analyst Stephen Howard
Bloomberg (Equity) TDC DC Market cap (DKKm) 37,698 Enterprise value (DKKm) 52850 Sector Diversified Telecoms Contact 44 20 7991 6820
60 55 50 45 40 35 30 25 20 2009
TDC Rel to KFX

Ratios (%)
0.3 0.1 -14.5 -6.6 -21.6
-0.1 3.9 12.4 36.8 31.6
0.2 0.8 4.9 6.5 6.3
0.4 0.7 4.2 5.5 5.3
60 55 50 45 40 35 30 25 20 2010 2011 2012
Per share data (DKK)
0.5 5.9 8.6 4.1 36.0 15.6 5.9 109.4 2.4 31.8
0.6 7.9 12.3 5.3 37.5 17.5 9.4 88.3 2.1 35.7
0.6 8.3 12.6 5.6 37.7 18.3 9.5 91.2 2.1 36.2
0.6 8.8 13.8 6.1 37.8 19.0 9.6 93.7 2.0 36.7
Source: HSBC
2.22 2.54 0.00 25.62
3.26 3.35 4.62 28.88
3.47 3.56 4.64 27.71
3.66 3.75 4.65 26.72
135
abc
Tele2 (Neutral, TP SEK160)

Tele2s focus is its mobile business; it has limited fixed-line presence. Tele2s cable TV network in Sweden passes through about 210k households, or c5% of total Swedish households. Although Tele2 is currently building out a fibre backhaul network in Sweden to support its mobile broadband business, it has not disclosed its coverage plans.
regulator seems to feel charging for packet prioritisation is perfectly reasonable. The regulatory view on net neutrality certainly is more relaxed at an EU level than is the case in the US. But it is plainly helpful that the national regulator has clear and sensible views on the subject as well.

Tele2 serves about 5% of Swedish homes with its cable TV network, competing against ComHem (cable), ViaSat (satellite) and other TV players.
Mobile connectivity
Tele2 has predominantly been a price challenger in the Swedish market and has continued to be aggressive in its offering. But with the rapid growth in smartphone penetration, the company has been increasing its focus on the post-paid segment of the market to take advantage of data growth. Tele2 launched 4G services in five cities in Sweden late last year. It plans to cover 100 cities by the end of FY 2011e and 99% of the population by the end of FY 2012e. The Swedish market still has high end flat-fee offers. Telenor, Tele2 and Three continue to have some of their plans under flat-fee offers. Telenor, which has about 18% market share, falls under our teenage operator definition and could disrupt the market in its effort to gain share and scale. Encouragingly, Telenors management recently commented that it sees flat-rate data tariffs as unsustainable. We note that even though flat-rate tariffs still exist in Sweden, the market already enjoys high single-digit mobile revenue growth rates. On net neutrality, the current thinking of the Swedish telecom regulator, PTS, is that: Traffic prioritisation does not need to be socio-economically inefficient as long as it takes place on a market where there is competition; rather it may often promote consumer benefits instead. Charging for better performance or access to desirable content services is a completely natural phenomenon in a market subject to competition. In other words, the
Mobile applications
Tele2 is not part of WAC (Wholesale Application Community) or OMA (Open Mobile Alliance).
Investment thesis
Tele2s focus continues to be infrastructure-led mobile businesses in Sweden, Russia, the Baltic countries and Kazakhstan. Tele2s operations in Russia and Sweden have continued to deliver good results and we expect positive growth momentum to continue. However, we have been sceptical on Tele2 arbitrage (ULL/MVNO) businesses, which contribute c25% to revenues. We expect these businesses to face significant pressure as incumbents (through fibre) and cable operators (through DOCSIS) upgrade their network capabilities, which in our view will put them in a strong position to take away market share from ULL operators. We also believe Tele2s lack of 3G spectrum could impede its growth prospects in Russia. We have a Neutral rating on the stock with a target price of SEK160 (increased from SEK155, post change to our WACC assumption) and believe Tele2s low debt levels (1.1x FY 2011e net debt/EBITDA post accounting for FY 2010e dividends) should continue to provide support for competitive yields.
136
abc
Financials & valuation: Tele2

Neutral
12/2012e 12/2013e
Profit & loss summary (SEKm)
Cash flow summary (SEKm)
40,197 10,284 -3,596 6,688 -497 6,412 6,268 -283 6,574 4,656
40,769 10,672 -3,618 7,054 -439 6,693 6,693 -1,720 4,965 4,965
42,043 11,653 -3,855 7,799 -536 7,340 7,340 -1,888 5,444 5,444
42,700 12,321 -4,058 8,263 -500 7,840 7,840 -2,018 5,822 5,814
Tele2 Revenues Tele2 EBITDA adjusted Tele2 PBT reported Tele2 Net income reported
40,197 10,284 6,735 6,926
40,769 10,672 6,693 4,965
42,043 11,653 7,340 5,444
42,700 12,321 7,840 5,814
9,610 -3,603 -5,260 -2,580 -203 5,801
9,178 -5,247 -5,247 -11,944 8,013 3,211
9,286 -4,816 -4,816 -3,760 -710 4,302
9,863 -4,202 -4,294 -4,123 -1,446 5,531
1.5 6.0 2.2 14.1 2.3 9.7 18.2
1.7 6.5 1.9 13.2 2.2 5.4 5.7
1.6 5.9 1.9 12.1 2.1 7.2 6.3
1.6 5.4 1.8 11.3 2.0 9.3 6.7
Balance sheet summary (SEKm)
13,201 15,130 7,697 946 40,369 7,695 2,948 2,002 28,872 27,387
13,201 16,760 7,837 1,000 42,216 262 11,015 10,015 30,077 36,535
13,201 17,721 8,028 1,000 43,445 873 10,305 9,305 31,398 37,077
13,293 17,865 8,126 1,000 43,857 1,312 8,859 7,859 32,808 36,973
Issuer information
Share price (SEK) 148.70 Target price (SEK) 160.00 Potent'l return (%) Reuters (Equity) TEL2b.ST Market cap (USDm) 9,638 Free float (%) 72 Country Sweden Analyst Dominik Klarmann
7.6
Bloomberg (Equity) TEL2B SS Market cap (SEKm) 62,449 Enterprise value (SEKm) 69438 Sector Diversified Telecoms Contact +49 211 910 2769
170 150 130 110 170 150 130 110 90 70 50 2010
Rel to OMX

Ratios (%)
0.7 11.3 17.8 27.7 23.7
1.4 3.8 5.5 4.4 6.6
3.1 9.2 10.6 9.7 9.6
1.6 5.7 6.0 6.8 6.8
90 70 50 2009
Tele2
2011
2012
Per share data (SEK)
1.5 18.6 16.3 16.1 25.6 16.6 20.7 6.9 0.2 480.0
1.3 16.4 16.8 12.9 26.2 17.3 24.3 33.3 0.9 91.7
1.1 15.7 17.7 13.7 27.7 18.5 21.8 29.6 0.8 99.8
1.2 16.6 18.1 14.2 28.9 19.4 24.6 23.9 0.6 125.5
Source: HSBC
14.86 10.53 27.00 65.27
11.22 11.22 8.50 67.99
12.31 12.31 9.32 70.98
13.16 13.14 9.95 74.17
137
abc
Telecom Italia (Underweight, TP EUR1.05)

Telecom Italia is lagging behind other European incumbents in terms of NGA deployment, currently running trials in Milan and Rome. The company has a target of reaching 1.3m households (about 5.5% of total) covered with FTTN/VDSL by the end of 2012 and a longerterm goal of 10m households (about 42% of total). Despite such unambitious plans, the lack of a cable infrastructure in Italy has in part preserved TIs position. However, it has not protected it from pricing pressure from altnets on an undifferentiated voice and broadband product. The national regulator AGCOM has granted TI an increase in ULL copper fees, which will gradually rise from EUR8.49 in early 2010 to EUR9.28 in 2012. While this has brought TIs ULL fees closer to the EU average, it is in counter trend with the other European markets. This, together with wholesale broadband access fees that most likely will be determined on a LRIC basis, does not create strong regulatory incentives for TI to accelerate its NGA deployment. AGCOMs position on net neutrality is no different from the other European regulators and is likely, in our view, to follow the EU recommendations, which currently appear pragmatic.
not be surprised if TI decided to adopt tiered data plans during the course of 2011.

Given the limited NGA deployment and the inferior contents available compared with the satellite alternative Sky Italia, IPTV has not so far had a significant impact: as of 2010, TIs IPTV customers are equivalent to only 5% of its broadband lines. Moreover, Fastweb and Sky have recently strengthened their commercial relationship: customers can now choose services from either within a joint plan and then receive a unified bill with dedicated call centre support. This represents an additional challenge to TI.
Mobile applications
TIs efforts in the applications space have so far had a limited impact: Tim store has around 1,000 applications available, grouped in categories such as social network, sport, games, travels and others. TI has reached an agreement with the main Italian publishers to offer a virtual bookshop; however, this is available on TIs dedicated device (the biblet) rather than as an app available on iPads or Android tablets. Although we see the soft SIM threat as overstated in most European markets, in Italy, given the history of traditionally low or no subsidies, a vendor like Apple would take a lower risk by taking the soft SIM route.
Investment thesis
We expect domestic mobile to continue its underperformance in Q4 2010 (results due on 24 February 2011) with double-digit service revenue decline. Increasingly tough competition in fixed line is a source of concern and may put additional pressure on the Q4 results. However, Brazil should outperform and dividend growth should resume. We are Underweight Telecom Italia with a target price of EUR1.05.
Mobile connectivity
TIs approach to tiered data pricing continues to make it quite an exception among peers. While most peers have shifted to tiered data plans, TI continues to stick to flat rate tariffs up to 2GB per month. As a comparison, Vodafone Italy has reduced its cap to 1GB per month. However, we do not think TIs case is any different from other operators: mobile capacity is an intrinsically scarce resource also for the company. We would
138
abc
Financials & valuation: Telecom Italia

Underweight
12/2011e 12/2012e
26,894 11,115 -5,852 5,263 -2,227 2,481 3,622 -1,121 1,345 2,256
27,462 11,391 -5,685 5,706 -2,071 3,684 4,109 -1,291 2,193 2,368
29,695 12,221 -5,850 6,371 -2,078 4,308 4,308 -1,632 2,227 2,227
29,927 12,159 -5,498 6,661 -1,979 4,696 4,696 -1,779 2,433 2,433
Domestic business Brazilian mobile TI Media Olivetti Others/Elimination Group Revenues
21,663 4,753 230 350 -102 26,894
20,073 6,181 246 377 585 27,462
19,494 6,824 251 377 2,748 29,695
19,213 7,223 256 377 2,857 29,927
6,094 -4,543 -5,188 -1,050 -90 2,535
6,326 -4,593 -4,411 -1,033 -2,020 2,417
7,596 -4,751 -4,751 -1,134 -1,712 3,725
7,960 -4,988 -4,988 -1,231 -1,741 3,369
2.1 5.0 0.9 9.0 0.8 11.8 4.8
1.9 4.7 0.8 8.6 0.8 11.3 5.2
1.7 4.2 0.8 9.1 0.7 17.2 5.7
1.7 4.1 0.8 8.3 0.7 15.4 6.2
49,909 14,902 17,683 6,619 86,181 11,302 43,738 33,949 25,952 64,573
49,837 15,808 16,573 6,346 86,886 11,108 42,790 31,929 26,754 64,764
49,754 14,792 16,562 6,000 85,779 11,503 40,733 30,217 27,750 63,604
49,933 14,103 16,597 6,000 85,308 11,592 38,992 28,476 28,856 63,040
Issuer information
1.05 Target price (EUR) TLIT.MI 26,218 81 Italy Luigi Minerva
-0.1
Bloomberg (Equity) TIT IM Market cap (EURm) 19,347 Enterprise value (EURm) 53344 Sector Diversified Telecoms Contact 44 20 7991 6928
2.5 2 1.5 2.5 2 1.5 1 0.5 0 2010
Rel to BCI ALL-SHARE INDEX

Ratios (%)
-7.3 0.2 1.1 -4.7 12.4
2.1 2.5 8.4 48.5 4.9
8.1 7.3 11.7 16.9 -6.0
0.8 -0.5 4.5 9.0 9.3
1 0.5 0 2009
Telecom Italia
2011
2012
0.4 7.6 8.7 4.8 41.3 19.6 5.0 125.2 3.1 18.0
0.4 7.7 9.0 4.4 41.5 20.8 5.5 109.5 2.8 19.8
0.5 8.3 8.2 4.7 41.2 21.5 5.9 99.5 2.5 25.1
0.5 8.6 8.6 5.0 40.6 22.3 6.1 89.8 2.3 28.0
Source: HSBC
0.07 0.12 0.05 1.35
0.11 0.12 0.06 1.39
0.12 0.12 0.06 1.44
0.13 0.13 0.07 1.50
139
abc
Telefonica (Overweight, TP EUR22)

The economic recession and the consequent slowdown of the broadband market have led Telefonica to shelve its NGA plans announced in October 2007 (aim of achieving 25% FTTH coverage). Now TEF targets a more focused VDSL in those areas where cable operator ONO is already present with a DOCSIS3.0 offer. The regulatory environment remains benign in Spain and this is likely to benefit Telefonica. First, national regulator CMT in October 2008 said that Telefonica would have to share its infrastructure as well as provide a wholesale service but only up to speeds of 30Mbps. Clearly, if this regulation were implemented, TEF would have an upper edge in competition considering that most of the fibre products would have speed greater than 30Mbps. Second, the CMT has indicated that in-building fibre should be shared, but the terms should be established by commercial negotiations (based on reasonable requests). On the net neutrality side, we believe that in Europe the regulators (both at the EC and national level) are likely to adopt a pragmatic approach that is unlikely to hurt operators. CMT has stated that traffic prioritisation could be allowed as long as it does not have a negative impact on competition.
all of its key markets, promoting WiFi usage and has been deploying selective femtocells. Telefonica has already departed from unsustainable all-you-can-eat tariffs by introducing tiering pricing. O2 UK is offering a maximum monthly allowance of 1GB and customers exceeding this are required to purchase an additional allowance, at GBP5 for 500MB and GBP10 for 1GB. In Spain, TEF has introduced a milder form of tiering plans. A data package of 1GB is 2.5x more expensive than a data package of 100MB; however, customers exceeding their data limit would see the download speed reduced significantly to 128Kbps.

Telefonica launched its IPTV service in Spain under the brand name Imagenio in 2004 and has now 773k customers with 18% market share of the pay-TV market. Telefonicas current IPTV offer is based on ADSL2+, which provides limited options. Despite the relatively early launch; Spain has an IPTV household penetration of only 5% (as of Q3 2010).
Mobile applications
Telefonica is one of the key promoters of the Wholesale Application Community (WAC) initiative, which in our view comes as too little, too late given the strong position that Apple and Google have built in this space. In February 2011, Telefonica announced a new multi-platform service (to be launched by Q3 2011 in among others Brazil, Mexico, Spain, UK and Germany) that allows customers to use applications on mobile, tablets, netbooks, and set-top-box TV.
Mobile connectivity
TEF has good 3G coverage of +90% in Europe and has proactively invested in HSDPA and currently has almost 80% coverage. TEF has faced some network outrage due to the exponential data traffic growth. For instance, O2 UK has faced some wellpublicised strains to its network in London and Bristol. TEF has consequently deployed additional 3G cell sites in major towns and cities, has moved to tiered data plans, has been conducting LTE trials in
Investment thesis
Telefonica offers a strong combination of value (with growing DPS visibility until 2012 and a very attractive 2010e yield of 7.6%) and growth (thanks to LatAm exposure and a strong mobile performance in the UK, which should offset a slower than we expected recovery in Spain.
140
abc
Financials & valuation: Telefonica

Overweight
12/2011e 12/2012e
56,731 22,603 -8,956 13,647 -3,307 10,387 9,949 -2,450 7,776 6,818
61,425 26,765 -9,282 17,484 -2,734 14,836 11,114 -3,550 11,074 7,896
65,342 24,353 -10,159 14,194 -3,086 11,199 11,199 -3,055 7,718 7,718
66,575 25,031 -9,825 15,206 -3,221 12,079 12,079 -3,296 8,297 8,297
Spain LatAm O2 Others Group revenues TEF FCF definition
19,432 22,984 13,532 783 56,731 9,098
18,775 26,336 15,320 994 61,425 7,283
18,539 30,357 15,327 1,118 65,342 7,984
18,368 31,715 15,219 1,273 66,575 8,750
15,897 -7,592 -8,770 -4,557 778 8,873
16,325 -8,208 -15,009 -5,877 12,236 11,871
16,240 -9,053 -10,559 -6,326 2,449 8,550
17,542 -9,589 -9,589 -7,074 -879 8,689
2.4 6.0 2.1 12.2 3.8 9.7 6.3
2.4 5.5 1.8 10.5 3.9 12.8 7.6
2.3 6.2 1.8 10.5 4.0 9.3 8.7
2.2 5.9 1.9 9.8 3.9 9.5 9.5
35,412 32,003 23,830 9,122 108,141 17,752 56,791 43,551 21,734 64,371
54,495 35,077 19,870 2,276 128,540 26,504 60,078 55,787 21,445 80,662
56,001 33,971 21,073 2,696 130,234 27,285 62,947 58,235 20,285 81,064
56,001 33,735 23,124 4,500 132,142 28,506 63,871 57,356 20,845 79,854
Issuer information
18.32 Target price (EUR) TEF.MC 113,341 100 Spain Luigi Minerva
20.1
Bloomberg (Equity) TEF SM Market cap (EURm) 83,635 Enterprise value (EURm) 148181 Sector Diversified Telecoms Contact 44 20 7991 6928
21 20 19 18 17 16 15 14 13 12 2009
Telefonica

Ratios (%)
-2.1 -1.4 -1.6 -4.8 -3.9
8.3 18.4 28.1 42.8 16.7
6.4 -9.0 -18.8 -24.5 -0.1
1.9 2.8 7.1 7.9 7.5
21 20 19 18 17 16 15 14 13 12 2010
Rel to MADRID SE
2011
2012
0.9 15.2 35.0 10.0 39.8 24.1 6.8 179.4 1.9 36.5
0.8 17.6 36.6 11.4 43.6 28.5 9.8 181.3 2.1 29.3
0.8 12.8 37.0 8.1 37.3 21.7 7.9 196.7 2.4 27.9
0.8 13.7 40.3 8.5 37.6 22.8 7.8 190.1 2.3 30.6
Source: HSBC
1.71 1.50 1.15 4.77
2.45 1.75 1.40 4.75
1.75 1.75 1.60 4.59
1.88 1.88 1.75 4.71
141
abc
Telekom Austria (Neutral, TP EUR11)

Telekom Austrias update on its NGA strategy at its capital markets day on 17 December 2010 is encouraging. At the end of December 2010 the companys VDSL/ Fibre to the Exchange network reached 38% of households, and it plans to have 45% of Austrian households by the end of 2011. Additionally, 5% of homes will be covered with either FTTB/FTTC or FTTH by end of 2011. While TA has long been the laggard in Europe in terms of NGA roll-out, the new management team seems to be much more aware of the fixed line opportunity; the fibre to the exchange technology combined with relatively short local loop length should allow TA a relatively cheap and fast NGA roll-out and TAs guidance of fixed access line growth is both exceptional in Europe and encouraging.
consider a positive for the incumbent. Historically, the Austrian regulator has closely followed EU recommendations, which appear much more relaxed on the issue than the US. We therefore consider this aspect as neutral to slightly positive for TA.

Telekom Austria offers IPTV service with 90-plus channels. It had 134,000 subscribers at the end of Q3 2010. It also offers 1,000 video-on-demand films and series.
Mobile applications
Telekom Austria is a member of the Open Mobile alliance and Wholesale Applications Community but to date little tangible results have come out of this initiative.
Investment thesis
Recent improvements in fixed-line trends and the progress in NGA roll-out have been encouraging, in our view, but we believe it is too early to call an economic recovery in the CEE markets (37% of group EBITDA). The company has set a floor of EUR0.76 dividend per share over 2011-12, which translates into a yield of 7% currently. During the capital markets day in December 2010, the company revised its payout ratio to 55% of free cash flow from 65% of net income earlier and increased its leverage target from 1.8-2.0x to 2.0-2.5x also flagging general interest in inorganic expansion within its region. We have a Neutral rating on the stock with a target price of EUR11 (increased from EUR10; we have raised slightly our revenue and EBITDA estimates to reflect stabilisation in fixed-line losses and stabilisation in CEE. Our target-price revision reflects the positive revision to our estimates and rolling forward of our valuation to year-end 2011.
Mobile connectivity
The Austrian market remains highly competitive in mobile broadband and the migration from fixed to mobile broadband has been quite high. Telekom Austria has been continually upgrading its mobile network, and now more than 1,000 base stations have been connected via fibre backbone. The company is also adopting dual cell roll-out, where in a second carrier per cell is installed. 25% of base stations were dual-cell by the end of 2010. Telekom Austria also became the first Austrian mobile operator to launch LTE services and has already upgraded 50 base stations. The Austrian market has largely moved away from all-you-can-use offers. Telekom Austria, T-mobile and Orange all offer data packages with download caps, except Three Austria, which continues to offer flat fee packages albeit at a higher monthly commitment. There is no active discussion about net neutrality in Austria at the moment, which we generally
142
abc
Financials & valuation: Telekom Austria

Neutral
12/2011e 12/2012e
4,802 1,794 -1,450 344 -220 106 530 -11 95 398
4,618 1,641 -1,068 573 -205 365 408 -82 283 307
4,570 1,586 -1,016 569 -223 346 346 -87 260 260
4,600 1,598 -949 649 -226 424 424 -106 318 318
TA Revenues (EURm) TA EBITDA Adj.(EURm) TA EBIT Adj.(EURm) TA Net Income Adj.(EURm) TA FCF Definition (EURm)
4,802 1,794 696 358 674
4,618 1,680 611 307 655
4,570 1,586 569 260 495
4,600 1,598 649 318 474
1,385 -711 -930 -332 -378 667
1,355 -699 -728 -332 -81 557
1,279 -784 -784 -332 -163 488
1,244 -770 -835 -336 -72 474
1.7 4.7 1.4 11.7 2.9 14.1 7.1
1.8 5.0 1.6 15.2 3.9 11.8 7.1
1.8 5.1 1.7 17.9 4.1 10.3 7.2
1.7 5.0 1.7 14.7 4.2 10.0 7.2
3,393 2,675 2,024 1,122 8,499 1,178 4,736 3,615 1,611 5,793
3,185 2,501 1,522 622 7,836 1,484 4,155 3,534 1,209 5,102
3,069 2,385 1,574 676 7,656 1,489 4,047 3,371 1,132 4,863
2,979 2,296 1,340 431 7,309 1,479 3,729 3,298 1,113 4,706
Issuer information
Share price (EUR)
4.6
Reuters (Equity) TELA.VI Market cap (USDm) 6,316 Free float (%) 73 Country Austria Analyst Dominik Klarmann
Bloomberg (Equity) TKA AV Market cap (EURm) 4,660 Enterprise value (EURm) 8273 Sector Diversified Telecoms Contact +49 211 910 2769
15 14 13 12 11 10 9 8 7 6 5 2009
Telekom Austria

Ratios (%)
-7.1 38.5 153.8 -9.1
-3.8 -8.5 66.6 243.5 -23.0
-1.0 -3.4 -0.6 -5.2 -15.3
0.7 0.8 14.1 22.4 22.4
15 14 13 12 11 10 9 8 7 6 5 2010
Rel to ATX
2011
2012
0.8 9.6 21.1 3.2 37.4 7.2 8.2 224.0 2.0 38.3
0.8 9.5 21.7 5.5 35.5 12.4 8.0 291.6 2.2 38.3
0.9 10.3 22.2 5.7 34.7 12.5 7.1 297.0 2.1 37.9
1.0 11.6 28.3 6.6 34.7 14.1 7.1 295.5 2.1 37.7
Source: HSBC
0.21 0.90 0.75 3.64
0.64 0.69 0.75 2.73
0.59 0.59 0.76 2.56
0.72 0.72 0.76 2.52
143
abc
Telenor (Overweight, TP NOK111)

Telenor enjoys the advantage of owning the cable network covering about a third of Norwegian homes. Its main competition arises from utilities building fibre networks that are technologically competitive but typically lack scale and knowhow in increasingly complex multi-play telco products. The Norwegian regulator laid out its position on net neutrality in consultation with operators in February 2009. On the issue of traffic management, the view is that internet users are entitled to an internet connection that is free of discrimination with regard to type of application, service or content or based on the sender or receiver address. However, it does not prohibit traffic management measures on an operators network if they are intended to ensure the quality of service for specific applications that require this, or to deal with situations of temporary network overload. We believe this could leave the door open to prioritising data streams of timesensitive material, such as voice and video. It would appear that Norway is at the tougher end of the spectrum in Europe. Telenor has some low-profile CDN activities in Norway competing with independent CDN players.
tiering volumes, speeds and time. In Norway, all the operators across the various subscription plans have usage caps, which range from 200MB to 6GB. In Sweden, Telenor maintains unlimited volume tariffs but rationalises usage by speed caps rather than volume caps. That has not harmed the Swedish mobile market, which is growing in high single digits.

Given its cable and satellite ownership, Telenor only introduced IPTV in 2010 and it has yet to gain traction. Overall, including satellite, cable and IPTV, Telenor enjoys a TV market share of 48%. Video on demand is in its infancy in Norway, a function of the small size of the market and potentially the monopolistic market structure.
Mobile applications
Telenor is especially active in financial services and mobile banking in emerging markets. Easypais in Pakistan aims for 7m active users and PKR10bn (USD120m) revenues in 2013. Telenor is also part of WAC, the Wholesale Application Community, but so far few tangible results have come out of this alliance.
Investment thesis
We believe Telenor has an attractive portfolio of assets that offer some of the best and most sustainable growth potential among Western European incumbents. In addition to the exposure to high-growth, low-penetration emerging economies, we believe Telenors multiple platform holdings in Norway (mobile, cable, satellite and DSL) make it well-placed as the industries converge and limit competitive pressures. We have an Overweight rating on Telenor, with a target price of NOK111. Telenor in our view is an attractive combination of sustainable growth and a competitive and growing distribution.
Mobile connectivity
We see Telenor as well-positioned in its data-plan structure; the company has introduced speed caps beyond a usage limit of 5GB per month and in its offer states that this policy is to ensure good service to all customers. Telenor management, during its capital-markets day presentation held in September 2010, highlighted that flat-rate data plans are not sustainable. Telenor is working on a new pricing structure for individual customer segments by
144
abc
Financials & valuation: Telenor

Overweight
12/2012e 12/2013e
Profit & loss summary (NOKm)
Cash flow summary (NOKm)
98,083 28,687 -16,349 12,338 -1,140 19,960 15,111 -4,975 14,332 9,974
100,544 30,510 -16,017 14,493 -1,077 17,851 17,851 -5,058 11,952 11,952
105,789 33,580 -15,769 17,811 -1,050 22,723 22,723 -5,365 16,113 16,113
109,174 35,526 -15,158 20,368 -836 26,487 26,487 -5,830 18,988 18,988
Telenor revenue Telenor EBITDA adjusted Telenor PBT adjusted Telenor Net income adjusted Telenor OCF definition
98,083 29,262 15,111 10,475 17,732
100,544 30,510 17,851 11,952 18,650
105,789 33,580 22,723 16,113 21,441
109,174 35,526 26,487 18,988 23,400
24,381 -11,863 -15,588 -4,141 -6,621 10,728
24,626 -11,860 -11,860 -6,198 -4,068 12,401
27,870 -12,138 -12,138 -2,116 -13,616 14,528
30,433 -12,127 -12,127 -8,862 -9,444 16,350
1.8 6.1 2.0 15.1 1.7 7.0 4.2
1.7 5.5 2.0 12.4 1.6 8.2 4.6
1.4 4.5 2.1 9.1 1.5 9.8 6.0
1.3 3.9 2.0 7.7 1.4 11.4 7.1
Balance sheet summary (NOKm)
51,479 52,963 32,209 14,552 172,731 34,212 35,577 21,025 87,867 87,887
47,198 53,087 29,596 11,794 168,012 34,811 28,751 16,957 90,617 83,276
43,174 53,481 43,999 25,410 181,766 42,154 28,751 3,341 97,868 73,088
39,808 53,815 53,951 34,854 192,066 44,324 28,751 -6,103 106,413 68,395
Issuer information
Share price (NOK)
91.45 Target price (NOK) 111.00 Potent'l return (%) 21.4 Bloomberg (Equity) TEL NO Market cap (NOKm) 151,614 Enterprise value (NOKm) 168311 Sector Diversified Telecoms Contact +49 211 910 2769
Reuters (Equity) TEL.OL Market cap (USDm) 25,924 Free float (%) 46 Country Norway Analyst Dominik Klarmann
105 95 85 75 65 55 45 35 25 2009
Telenor

Ratios (%)
0.3 -7.1 -7.0 31.5 -31.2
2.5 6.4 17.5 -10.6 21.8
5.2 10.1 22.9 27.3 36.2
3.2 5.8 14.4 16.6 17.8
105 95 85 75 65 55 45 35 25 2010
Rel to OBX INDEX
2011
2012
Per share data (NOK)
1.1 9.2 12.2 9.6 29.2 12.6 25.2 21.9 0.7 116.0
1.2 12.1 13.4 8.1 30.3 14.4 28.3 17.4 0.6 145.2
1.4 17.4 17.1 10.5 31.7 16.8 32.0 3.2 0.1 834.2
1.5 22.5 18.6 11.6 32.5 18.7 42.5 -5.4 -0.2
Source: HSBC
8.71 6.06 3.80 53.38
7.38 7.38 4.18 55.94
10.05 10.05 5.53 61.04
11.84 11.84 6.51 66.37
145
abc
TeliaSonera (Neutral, TP SEK58)

In March 2008, TLSN announced it would upgrade 1.5m to 2.0m households, out of 4.4m Swedish homes and enterprises, with a combination of FTTH and VDSL technology over a period of five years. Key competitors are Com Hem, the Swedish cable operator covering roughly 40% of Swedish homes, and fibre operator Bredbandbolaget. At the end of 2010, TeliaSonera covered 10% of Swedish homes with NGA, with a take-up rate of c32%. Uncertainty about regulation remains. The Swedish regulator, PTS, has proposed forcing TLSN to unbundle its fibre-optic broadband access network and offer a full range of wholesale fibre services to competitors, including fibre and ducts. PTS is expected to announce its decision on fibre wholesale prices sometime in Q2 2011.
On net neutrality, the current thinking of PTS is that Traffic prioritisation does not need to be socioeconomically inefficient as long as it takes place on a market where there is competition; rather it may often promote consumer benefits instead. Charging for better performance or access to desirable content services is a completely natural phenomenon in a market subject to competition. In other words, the regulator seems to understand that charging for packet prioritisation is perfectly reasonable. The regulatory view on net neutrality certainly is more relaxed at an EU level than it is in the US. But it is plainly helpful that the national regulator has clear and sensible views on the subject as well.

TeliaSonera serves about 15% of Swedish homes with IPTV, including video on demand, competing against ComHem (cable), ViaSat (satellite), Tele2Vision (IPTV) and other TV players. TeliaSonera has about 450,000 IPTV subscribers or roughly 10% of Swedish homes.
Mobile connectivity
We see TeliaSoneras management as frontrunners in developing sophisticated pricing plans to monetise the data growth opportunity. TeliaSonera introduced tiered data pricing plans in mid 2009. The Swedish market still has high flat-fee offers. Telenor, Tele2 and Three provide some of their plans under flat-fee offers. Telenor, which has about 18% market share, falls under our teenage operator definition and could disrupt the market in its quest to increase share and scale. Encouragingly, even Telenors management recently commented that it sees flat-rate data tariffs as unsustainable. We would note that although flat-rate tariffs still exist in Sweden, mobile revenue is already growing at healthy high-single-digit rates.
Mobile applications
TeliaSoneras management is a firm believer in telecoms core strength being connectivity and providing super-fast broadband rather than in services and applications. Hence, TeliaSonera is not part of WAC (Wholesale Application Community) or OMA (Open Mobile Alliance).
Investment thesis
Our view is that TeliaSonera has been taking the right steps by investing in mobile and fixed business infrastructure and its reasonable exposure to Eurasian markets should help to drive top-line growth. However, uncertainty over its Russian and Turkey assets, along with increased M&A risk and competition expected in Kazakhstan, make us prefer Telenor. We have a Neutral rating on the stock with a target price of SEK58. TeliaSoneras low debt levels should provide support for competitive dividend yields.
146
abc
Financials & valuation: TeliaSonera

Neutral
12/2012e 12/2013e
Profit & loss summary (SEKm)
Cash flow summary (SEKm)
106,582 37,741 -13,479 24,262 -1,863 30,001 30,217 -6,374 21,322 22,040
106,041 37,242 -12,598 24,644 -1,728 30,875 30,900 -6,416 21,852 21,877
109,481 38,409 -12,643 25,766 -1,630 32,499 32,499 -6,758 22,980 22,980
110,850 38,941 -13,068 25,873 -1,716 33,097 33,097 -6,764 23,461 23,461
TLSN Revenues TLSN EBITDA adjusted TLSN PBT adjusted TLSN Net income adjusted TLSN FCF definition
106,582 36,977 30,217 21,538 12,901
106,041 37,242 30,900 21,877 17,331
109,481 38,409 32,499 22,980 17,890
110,850 38,941 33,097 23,461 18,705
27,434 -14,934 -16,476 -10,104 -1,726 14,570
31,767 -14,450 -14,450 -12,349 5,031 14,634
32,388 -14,666 -14,666 -12,828 -4,894 15,187
33,263 -14,574 -14,574 -13,470 -5,220 15,871
2.1 5.9 1.6 11.1 1.9 8.5 5.0
2.1 6.0 1.6 11.0 1.9 8.6 5.5
2.0 5.6 1.5 10.2 1.7 9.1 5.7
1.9 5.3 1.4 10.0 1.6 9.8 6.0
Balance sheet summary (SEKm)
90,531 58,353 39,209 17,821 250,551 27,627 67,029 49,208 125,907 142,645
90,531 60,205 24,916 3,500 244,349 29,084 57,739 54,239 124,931 143,068
90,531 62,228 29,432 7,500 257,729 31,087 56,846 49,346 134,441 143,605
90,531 63,734 29,637 7,500 266,577 32,964 51,626 44,126 143,759 143,439
Issuer information
Share price (SEK)
54.50 Target price (SEK)
6.4
Reuters (Equity) TLSN.ST Market cap (USDm) 37,772 Free float (%) 36 Country Sweden Analyst Dominik Klarmann
Bloomberg (Equity) TLSN SS Market cap (SEKm) 244,730 Enterprise value (SEKm) 224545 Sector Diversified Telecoms Contact +49 211 910 2769
58 53 48 43 58 53 48 43 38 33 28 2010
Rel to OMX

Ratios (%)
-2.4 7.1 8.8 8.6 3.7
-0.5 -1.3 1.6 2.9 1.3
3.2 3.1 4.6 5.3 7.4
1.3 1.4 0.4 1.8 2.1
38 33 28 2009
TeliaSonera
2011
2012
Per share data (SEK)
0.7 13.4 16.9 9.7 35.4 22.8 20.3 37.1 1.3 55.8
0.7 13.7 17.4 10.4 35.1 23.2 21.6 40.4 1.5 58.6
0.8 14.2 17.7 10.8 35.1 23.5 23.6 33.7 1.3 65.6
0.8 14.3 16.9 10.6 35.1 23.3 22.7 27.8 1.1 75.4
Source: HSBC
4.75 4.91 2.75 28.04
4.97 4.97 2.98 28.39
5.34 5.34 3.13 31.23
5.45 5.45 3.29 33.39
147
abc
Telstra (Overweight, TP AUD3.40)

Telstras wireline outlook is dominated by the current negotiations over the National Broadband Network (NBN). If its deal with the government and the NBN Co (with a post-tax NPV to Telstra of AUD11bn) is approved, it will lease its fixed network assets to the NBN Co, which will become the sole wholesale supplier of high-speed broadband to businesses and consumers in Australia. The government-funded venture aims to provide high-speed broadband to households and businesses nationwide (using FTTH and wireless), with a price tag of up to cAUD43bn. Our scenario analysis indicates margin compression over time as a result of Telstras move away from vertical integration, but we expect it to be able to port many of its existing client relationships into the new structure.
but Telstra erred in allowing tariff premiums to escalate as competitors cut prices in 2009 and early 2010. It has now revised its tariffs and allowances, increasing the appeal to users frustrated by coverage and capacity shortcomings at VHA. This operator faces a class action as a result of a series of network problems over the past year the result of issues subsequent to the Vodafone and Three Australia merger, as well as network overloading. The basic pricing model for mobile data is a tiered structure, with Telstra more aggressive than peers in charging for usage outside these caps.

Telstras recent fixed line application strategy has centred on its T-Hub (VoIP) and T-Box (IPTV) products which had 128,000 and 107,000 users at end December 2010, from launch in early 2010. It is focused on adding bundled service customers (804k at end 2010) to reduce customer churn. These are powered by its Next IP backbone network.
Mobile connectivity
Telstra results for the half ending December 2010 indicated that its investment in opex (cAUD1bn this FY) to rebuild market share is paying off. It added over 1m accounts in the half (a record), including 917,000 mobile accounts. The company is benefiting from strong organic demand for smartphones, in addition to network problems at no.3 operator Vodafone Hutchison Australia (VHA). We believe the company is growing revenue market share as a result. Telstra is increasing subsidy levels (particularly in smartphones) to rebuild market share, rather than cut prices. However, its shift to capped plans (allowing larger amounts of usage for a specific cap) is limiting ARPU gains postpaid handset ARPU in the half ending Dec 2010 was down 0.9% y-o-y. Telstra continues to leverage its Next G wireless network. This has the best coverage in Australia,
Mobile applications
Telstras mobile application strategy remains a mix: after a period of focusing on vendors such as HTC for smartphones, it fully embraced the iPhone 4. It is also working to integrate closer with Sensis, its advertising business, as more traffic and activity moves towards mobile.
Investment thesis
Our Overweight rating for Telstra is predicated on its high dividend yield (c9.6%) and operational improvement as a result of this years strategic investment in EBITDA. We expect Telstras NPV of AUD11bn from the NBN negotiations to remain secure, and believe its early investment in the Next G wireless and Next IP wireline networks has created a durable competitive advantage relative to its peers.
148
abc
Financials & valuation: Telstra Corp

Overweight
06/2012e 06/2013e
Profit & loss summary (AUDm)
Cash flow summary (AUDm)
25,029 10,847 -4,346 6,501 -963 5,538 5,538 -1,598 3,883 3,877
25,284 10,093 -4,388 5,705 -938 4,767 4,193 -1,290 3,441 2,935
25,176 10,532 -4,232 6,300 -922 5,378 5,378 -1,645 3,694 3,765
25,120 10,584 -4,111 6,474 -842 5,631 5,631 -1,721 3,869 3,942
2.0 4.6 1.7 9.3 2.9 15.1 9.6
2.0 5.0 1.7 12.3 3.0 11.7 9.6
1.9 4.6 1.7 9.6 3.0 12.1 9.7
1.9 4.5 1.7 9.2 2.9 12.5 9.8
Issuer information
Share price (AUD) 8,629 -3,595 -3,466 -3,474 -2,820 5,434 6,801 -3,762 -3,643 -3,484 28 4,186 8,158 -3,450 -3,450 -3,484 -1,224 4,352 8,069 -3,507 -3,507 -3,519 -1,043 4,486
2.91 Target price (AUD)
Reuters (Equity) TLS.AX Market cap (USDm) 36,284 Free float (%) 50 Country Australia Analyst Neale Anderson
Bloomberg (Equity) TLS AU Market cap (AUDm) 36,209 Enterprise value (AUDm) 50007 Sector Diversified Telecoms Contact +852 2996 6716
Balance sheet summary (AUDm)
Price relative
1,802 29,120 7,419 2,688 39,282 5,807 16,763 14,075 12,696 29,846
1,802 28,109 6,561 1,000 37,138 5,614 15,103 14,103 11,916 29,858
1,802 27,327 6,548 1,000 36,343 5,829 13,879 12,879 12,092 28,848
1,802 26,723 6,541 1,000 35,732 5,905 12,835 11,835 12,407 28,161
5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 2009

Telstra Corp
5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 2010 2011 2012

Rel to AUSTRALIAN ALL ORDINARIES
Source: HSBC

Ratios (%)
-2.0 -0.1 0.5 -0.6 0.1
1.0 -6.9 -12.2 -13.9 -24.3
-0.4 4.3 10.4 12.8 28.3
-0.2 0.5 2.8 4.7 4.7
Per share data (AUD)
0.8 14.8 30.9 11.8 43.3 26.0 11.3 108.2 1.3 61.3
0.8 13.4 23.9 11.1 39.9 22.6 10.8 115.9 1.4 48.2
0.9 15.0 31.4 12.0 41.8 25.0 11.4 104.0 1.2 63.3
0.9 15.9 32.2 12.6 42.1 25.8 12.6 92.9 1.1 68.2
0.31 0.31 0.28 1.02
0.28 0.24 0.28 0.96
0.30 0.30 0.28 0.97
0.31 0.32 0.29 1.00
149
abc
TIM Participaes (Overweight, TP BRL7.50)

TIM Participaes is effectively a mobile pure-play, having relatively little revenue exposure to fixed line via its 2009 acquisition of Brazilian long-distance altnet Intelig. Unbundling is not a major problem for fixed-line players in Brazil as most local access competition is facilities-based (although players like fibre-based altnet GVT and dominant cable player NET Servios offer enough competition for the incumbents, we would argue). Neither is net neutrality a hot issue for the Brazilian regulator Anatel at the moment it seems, generating little public comment on the matter thus far.
Mobile connectivity
Although smartphone penetration in Brazil is low excluding QWERTY devices like the BlackBerry, we estimate penetration of true smartphones is in the low single digits we expect it to rise steadily in coming years, driven by continued steep price erosion of Android-based smartphones, primarily produced by Asian vendors. TIM has thus far not fully exploited the mobile data opportunity in Brazil we would argue, having instead focused up to this point mainly on voice services. This is changing though as the firm re-vamped its data offering in H2 and placed greater emphasis on selling smartphones in the fourth quarter, we believe. Data as a percentage of mobile service revenues is in the lowteens we estimate, which compares with more than 20% at data leader Vivo, highlighting the potential growth upside. Given relatively low levels of data traffic on its network, TIM has not experienced capacity issues, and continues to invest primarily in increasing its 3G footprint, which is now around 60% of the urban population. Although the data pricing environment in Brazil is less aggressive than the voice market, TIM is offering some of the more competitive tariffs
in the market at the moment, including a BRL0.50 per day flat rate for a 300kbps service (but crucially with bandwidth throttled to a maximum of 50kbps after a threshold of 10MB per day or 300MB per month has been reached). Unlike Vivo and America Movils Claro unit, which focus primarily on the corporate and high-end consumer segments of the mobile data market, TIM is focused firmly on the large, but lower-income C-class (c60% of Brazils population), targeting users who typically require only email, instant messaging and social networking rather than video or other high-bandwidth services. We believe that lacking the massive capital resources of its rivals (which are controlled by the two dominant Latin American telecoms groups, America Movil and Telefonica), it is sensible for TIM to pursue a differentiated strategy rather than try to compete head-to-head.

With no local access wireline presence TIM does not have plans to enter the pay-TV market.
Mobile applications
TIM lacks the scale of regional giants America Movil and Telefonica. As we do not expect these players to stand much chance of securing a more central role in the applications value chain, we see even less chance for TIM. Note that unlike the revenue share model for the Apple App Store (whereby operators generally receive nothing), Googles Android Market often has provisions for revenue shares with operators (with Google instead taking little or no revenue itself).
Investment thesis
We rate TIM Participaes Overweight. Following the fixed-mobile merger of America Movil and Telmex Internacional and the pending fixed-mobile merger of Vivo and Telesp (the Sao Paulo wireline incumbent), we believe TIM is the clearest way to play the mobile data scarcity theme in Latin America.
150
abc
Financials & valuation: TIM Participaes

Overweight
12/2011e 12/2012e
Profit & loss summary (BRLm)
Cash flow summary (BRLm)
13,744 3,469 -2,693 776 253 1,030 1,094 -6 1,024 710
14,404 4,164 -2,701 1,464 -258 1,205 1,205 -255 951 795
15,907 4,668 -2,765 1,903 -174 1,729 1,729 -415 1,314 1,141
16,836 5,032 -2,882 2,150 -117 2,033 2,033 -488 1,545 1,342
Subscribers ARPU Service revenues Handset revenue Total revenues
41,102 27 12,806 938 13,744
51,028 24 13,652 752 14,404
56,066 22 15,117 789 15,907
59,899 22 16,098 739 16,836
2,331 -2,690 -2,690 0 -394 1,015
3,400 -2,935 -2,935 -201 -279 712
4,075 -2,964 -2,964 -225 -886 1,110
4,402 -3,620 -3,620 -409 -374 783
1.3 5.1 1.8 20.2 1.7 6.3 2.1
1.2 4.2 1.6 18.7 1.7 4.5 2.3
1.0 3.5 1.5 13.1 1.6 6.9 2.7
1.0 3.2 1.4 11.1 1.4 4.9 5.0
Balance sheet summary (BRLm)
4,494 5,323 6,767 2,559 17,450 4,320 4,160 1,601 8,323 9,705
3,930 5,866 5,621 1,404 16,402 3,307 2,726 1,322 8,678 10,705
2,977 7,018 6,538 2,290 17,519 3,519 2,726 436 9,583 10,725
2,251 8,482 7,005 2,664 18,723 3,931 2,726 62 10,376 11,144
Issuer information
Share price (BRL) Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
6.02 Target price (BRL) TCSL4.SA 9,568 30 Brazil Richard Dineen
Bloomberg (Equity) TCSL4 BZ Market cap (BRLm) 15,983 Enterprise value (BRLm) 17,305 Sector Wireless Telecoms Contact 1 212 525 6707
8 7 6 5 4 3 2 1 2009
TIM Participacoes
8 7 6 5 4 3 2 1 2010
Rel to BOVESPA INDEX

Ratios (%)
5.1 19.6 58.1 788.5 273.2
4.8 20.0 88.5 17.0 7.7
10.4 12.1 30.0 43.4 43.4
5.8 7.8 13.0 17.6 17.6
2011
2012
Per share data (BRL)
1.4 5.2 8.8 8.3 25.2 5.6 19.2 0.5 145.6
1.4 9.5 9.4 9.0 28.9 10.2 16.1 15.2 0.3 257.2
1.5 11.7 12.5 9.0 29.3 12.0 26.8 4.6 0.1 933.9
1.5 13.0 13.4 9.6 29.9 12.8 43.1 0.6 0.0 7047.8
Source: HSBC
0.43 0.30 0.13 3.50
0.38 0.32 0.14 3.51
0.53 0.46 0.17 3.87
0.62 0.54 0.30 4.19
151
abc
TPSA (Underweight, TP PLN15.3)

TPSA has an ADSL-based network to provide high-speed broadband. In 2009, it entered into an agreement with the regulator that requires TPSA to provide high-speed broadband to 1.2m lines, which will need an investment of cPLN3bn over three years. That has helped to ease the regulatory pressure somewhat. We expect the regulator to intervene less as the network improvement continues. However, convergence with European regulations may increase net neutrality threat. Polish subscribers spend relatively little on broadband. TPSA is working to increase its market share using triple-play services. In 2010, it reduced fixed-line broadband pricing, through which it intends to re-engage with the core market. TPSA also continues to roll out TV services to its broadband subscribers. Competing offerings, however, are strong in Poland, mainly from cable operators, which are upgrading their networks to DOCSIS3.0. Overall, we consider TPSA vulnerable to the cable threat.
pricing power may change significantly if data pricing becomes irrational, like the reductions in voice tariffs after MTR cuts in 2009.
IPTV is considered an add-on in Poland, with users reluctant to pay unless real premium content is provided. In October 2010, TPSA signed a 10year content agreement with TVN group to crosssell each others services, with a special focus on multi-play bundles. TPSAs own portal, Wirtualna Polska, is one of the most frequently visited websites in Poland. TPSA is also trying to leverage its subsidiaries to venture into adjacent sectors like payment and e-commerce. Competing offerings on content are strong. UPC has had experience providing media content in CE3 with cable since the late 1990s. It is plausible that certain OTT competitors may challenge IPTV services and well-established brands in the pay-TV market.
Mobile applications
TPSA has no substantial advantage over global players in content and data applications, despite its familiarity with the local culture. We expect low-cost smartphones based on the Android platform to drive internet usage in Poland. However, language may present a barrier to would-be developed-world competitors, so TPSA may not be completely dis-intermediated from the applications layer.
Mobile connectivity
In Poland, tariffs are data tiered. Some packages are marketed as unlimited, but in fact are capacity-limited. There is no all-you-can-eat data plan currently, which gives pricing power to the mobile operators. However, new entrants like P4 look to undercut the larger operators on pricing, thereby providing tough competition. TPSAs mobile broadband subscribers make up less than 4% of its mobile subscribers, which seems to reflect weaker demand for broadband services and increased competition in mobile data from P4. Non-voice revenue has not increased significantly over the past two years, remaining around 24% of revenue for TPSA. We believe
Investment thesis
TPSA is trading at the top end of its five-year PE band of 10x to 16x. We believe TPSA does not warrant a premium over its CE3 peers, because of the uncertainties involving DPTG claims. If the recent claim by DPTG is taken into account, it could almost halve EPS for 2011. Furthermore, its dividend yield offers a much smaller spread over the sovereign bond yield compared with TelefonicaO2 CZ.
152
abc
Financials & valuation: TPSA

Underweight
12/2011e 12/2012e
Profit & loss summary (PLNm)
Cash flow summary (PLNm)
16,560 6,280 -4,183 2,097 -363 1,598 1,823 -315 1,281 1,456
15,631 4,742 -3,801 941 -389 508 1,639 -321 184 1,309
15,375 5,786 -3,564 2,222 -327 1,895 1,895 -379 1,514 1,514
15,095 5,712 -3,364 2,348 -336 2,012 2,012 -402 1,608 1,608
Reported revenue (PLNm) Reported EBITDA (PLNm) Reported EBIT (PLNm) Reported PBT (PLNm) Reported EPS (PLN)
16,560 6,280 1,961 1,598 0.96
15,631 4,742 896 508 0.14
15,375 5,786 2,222 1,895 1.13
15,095 5,712 2,348 2,012 1.20
5,541 -2,207 -2,281 -2,004 -1,268 3,337
4,720 -2,516 -2,632 -2,004 -229 1,508
5,070 -2,760 -2,760 -2,003 -307 2,299
4,931 -2,717 -2,717 -2,003 -211 2,191
1.6 4.2 0.8 15.2 1.3 15.1 9.1
1.7 5.5 0.8 16.9 1.7 6.8 9.1
1.7 4.5 0.8 14.6 1.8 10.4 9.1
1.7 4.5 0.8 13.8 1.9 9.9 9.1
Balance sheet summary (PLNm)
7,298 17,743 4,189 2,218 29,365 -4,756 6,499 4,281 16,539 31,768
7,236 16,456 3,610 1,539 27,416 -7,505 5,591 4,052 12,694 33,268
7,236 15,652 3,902 1,846 26,903 -7,480 5,591 3,745 12,205 32,424
7,236 15,005 4,057 2,057 26,411 -7,381 5,591 3,534 11,809 31,622
Issuer information
Share price (PLN) Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
16.57 Target price (PLN) TPSA.WA 7,657 50 Poland Herve Drouet
-7.7
Bloomberg (Equity) TPS PW Market cap (PLNm) 22,132 Enterprise value (PLNm) 26086 Sector Diversified Telecoms Contact 44 20 7991 6827
26 24 22 20 18 16 14 12 10 8 2009
TPSA Rel to WIG 20

Ratios (%)
-8.8 -16.5 -36.7 -38.4 -33.1
-5.6 -24.5 -55.1 -68.2 -10.1
-1.6 22.0 136.1 273.5 15.7
-1.8 -1.3 5.7 6.2 6.2
26 24 22 20 18 16 14 12 10 8 2010 2011 2012
Per share data (PLN)
0.5 5.1 8.6 5.3 37.9 12.7 17.3 25.9 0.7 129.4
0.5 2.3 9.0 2.0 30.3 6.0 12.2 31.9 0.9 116.5
0.5 5.4 12.2 6.7 37.6 14.5 17.7 30.6 0.6 135.4
0.5 5.9 13.4 7.2 37.8 15.6 17.0 29.9 0.6 139.5
Source: HSBC
Stated accounts as of 31 Dec 2005 are IFRS compliant
0.96 1.09 1.50 12.38
0.14 0.98 1.50 9.50
1.13 1.13 1.50 9.14
1.20 1.20 1.50 8.84
153
abc
Turk Telekom (Underweight, TP TRY6.80)

Broadband is the growth driver for the Turkish telecom market, but fixed broadband growth in Turkey is hampered by low PC penetration. Subscriber growth will be hard to come by in the near future, and we are sceptical about the short-term growth potential of fixed broadband. We believe it will be constrained by the small gap between PC penetration and broadband penetration (households with PC but no broadband), as well as Turkcells aggressive promotion of 3G. Average revenue per line (ARPL) is increasing, however, mainly owing to the companys focus on up-selling/upgrading services to existing clients. With more than 70% of its users already subscribed to its highest-speed package (up to 8Mbps), and more than 50% using unlimited packages, the scope for further increase in ARPL looks quite limited. We believe that Turk Telekoms attempt to increase prices for unlimited packages in 2011 and impose fair usage quotas could adversely affect its subscriber and ARPL growth in the near term. Furthermore, the nominal price increase would not translate into pricing power, in real terms, owing to the c7% inflation in Turkey in 2011 forecast by HSBC economists. Positives for Turk Telekom include the absence of regulations on net neutrality, and the fact that local loop unbundling remains unattractive for altnets.
believe pricing power may diminish significantly if data pricing becomes irrational similar to the competition seen in the voice segment post introduction of mobile number portability in November 2008.

It is plausible that certain OTT competitors could pose a threat to the established brands in the pay-TV market in Turkey. However, factors such as the complexities of rights arrangements, the power of existing brands and the absence of net neutrality regulations could give incumbents significant pricing power and advantages.
Mobile applications
Mobile operators in Turkey particularly Turkcell have focused on capturing the applications layer. Even Avea (TurkTelekom mobile subsidiary) has attempted to develop innovative applications aimed at certain segments. We expect Turkish mobile operators to keep their edge over the global players on data content as they have the advantage of being familiar with the local culture. We expect low-cost smartphones based on the Android platform to drive internet usage. However, language may present a barrier to would-be developed world competitors. As a result, Avea may not be completely disintermediated from the applications layer.
Investment thesis
We forecast 2011 pro forma revenue growth of 4.8% and an EBITDA margin of 45.3%. HSBC economists expect 7% inflation in Turkey in 2011, implying a revenue decline, in real terms, for Turk Telekom. In our view, wage inflation could be a serious threat to its EBITDA margin as personnel costs still account for c20% of its fixed-line revenue. There is also not much scope for a significant reduction in the employee headcount. It is trading at a premium to its peer group, which we believe is unwarranted. Its dividend yield implies a negative spread to the countrys long-term bond yield.
Mobile connectivity
3G services, heavily promoted by Turkcell, are driving broadband growth in Turkey. Tariffs in Turkey are data tiered and there is currently no all you can eat data plan, which gives mobile operators the pricing power. Turkcell has recently launched competitive tariff plans in the post-paid segment, while Vodafone Turkey is increasingly pushing wireless data with increased 3G coverage. We
154
abc
Financials & valuation: Turk Telekom

Underweight
12/2012e 12/2013e
Profit & loss summary (TRYm)
Cash flow summary (TRYm)
10,852 4,835 -1,524 3,311 -97 3,127 2,750 -799 2,451 2,107
11,570 5,245 -1,620 3,626 -259 3,366 3,366 -673 2,703 2,703
12,201 5,542 -1,640 3,902 -243 3,658 3,658 -732 2,927 2,927
12,782 5,707 -1,651 4,056 -334 3,722 3,722 -744 2,978 2,978
Mobile penetration (%) Mobile revenue (TRYm) Mobile EBITDA (TRYm) Broadband HH penetration (%) Fixed line revenue (TRYm) Fixed line EBITDA (TRYm)
85.6 2,646 332 38.0 8,511 4,507
91.8 2,958 629 40.9 8,938 4,616
97.6 3,453 898 43.6 9,091 4,644
102.9 3,949 1,058 46.2 9,192 4,649
3,844 -1,733 -1,523 -1,590 -382 2,165
4,266 -1,998 -1,998 -2,244 -24 2,289
4,630 -1,931 -2,949 -2,459 778 2,641
4,677 -1,951 -1,951 -2,780 55 2,713
2.5 5.7 1.8 11.8 4.0 8.6 9.0
2.4 5.3 1.7 9.2 3.9 9.1 9.9
2.3 5.1 1.6 8.5 3.8 10.4 11.2
2.2 5.0 1.6 8.4 3.7 10.7 11.4
Balance sheet summary (TRYm)
3,570 7,161 3,712 1,875 15,100 -2,957 4,164 2,289 6,175 15,525
3,570 7,539 2,712 847 14,478 -3,153 3,112 2,265 6,418 16,127
4,071 7,830 3,466 1,500 16,023 -3,639 4,542 3,042 6,564 17,505
4,071 8,130 3,559 1,500 16,416 -3,828 4,597 3,097 6,713 18,087
Issuer information
Share price (TRY) Reuters (Equity) Market cap (USDm) Free float (%) Country Analyst
7.12 Target price (TRY) TTKOM.IS 15,702 13 Turkey Herve Drouet
-4.5
Bloomberg (Equity) TTKOM TI Market cap (TRYm) 24,920 Enterprise value (TRYm) 27545 Sector Diversified Telecoms Contact 44 20 7991 6827
9 8 7 6 5 4 3 2 1 2009
Turk Telekom

Ratios (%)
2.7 11.0 18.3 32.5 23.5
6.6 8.5 9.5 7.7 28.3
5.4 5.7 7.6 8.7 8.3
4.8 3.0 4.0 1.7 1.7
9 8 7 6 5 4 3 2 1 2010
Rel to ISTANBUL COMP
2011
2012
Per share data (TRY)
0.7 16.2 36.3 18.0 44.5 30.5 50.1 34.2 0.5 167.9
0.7 18.3 42.9 20.0 45.3 31.3 20.2 32.7 0.4 188.4
0.7 18.6 45.1 20.8 45.4 32.0 22.8 46.3 0.5 152.2
0.7 18.2 44.9 20.6 44.7 31.7 17.1 46.1 0.5 151.0
Source: HSBC
Stated accounts as of 31 Dec 2006 are IFRS compliant
0.70 0.60 0.64 1.76
0.77 0.77 0.70 1.83
0.84 0.84 0.79 1.88
0.85 0.85 0.81 1.92
155
abc
Verizon (Overweight, TP USD41)

The FCCs Triennial Review in 2003 largely withdrew support for unbundling (known in the US as UNE-P, Unbundled Network Elements Platform) of incumbent local access fibre investments. This shift almost immediately prompted Verizon and AT&T to initiate extensive fibre projects: FiOS and U-verse respectively. Over the past five years unbundled providers in fibred areas in the US have all but disappeared. While unbundling regulation has favoured the incumbents, the FCC has taken a hard line on net neutrality. This is understandable given the strategic importance of US Internet companies (whose products and services may be exported, unlike those of the domestically-focused RBOCs). In a December 2010 decision, the FCC agreed a new policy that preserves strict net neutrality principles for fixed-line services while permitting some traffic management on mobile networks.
Verizon Wireless has resisted following AT&T in moving to tiered data plans and still offers smartphone customers a single USD30 flat-rate plan, although it has not ruled out moving to tiers longerterm. We believe that sticking with unlimited for the time being is primarily a tactical move designed to maximise initial sales of the iPhone. Verizon Wireless generally enjoys stronger pricing power than rivals, which stems from the broad customer perception that it leads the market in terms of network quality. Unlike AT&T, Verizon Wireless does not currently or plan to use public WiFi as a means of offloading traffic. However, we expect Verizons attitude here to be flexible, particularly as the impact of doubling its smartphone base on the network is unknown at this stage.

Verizons FiOS TV competes head-to-head with cable in areas where FiOS is deployed and selectively against satellite players, as the firm also re-sells DirecTV satellite services for customers outside the FiOS footprint. FiOS TV has garnered 3.5m customers thus far averaging 28% penetration in areas available for sale.
Mobile connectivity
The US is among the most advanced mobile data markets in the world. Verizon reports that around 26% of its base currently has a smartphone a figure which it expects to double to around 50% by end-2011, driven primarily by sales of the first CDMA iPhone4, available since early February. We expect a significant portion of Verizons existing base to migrate to the iPhone and for it to gain an increasing share of postpaid net additions over the next two years as customers on other networks switch as their contracts expire, attracted by Verizons reputation for superior network coverage and quality. Until this point Verizon has not experienced any major network outages as a result of smartphone demand, although this could clearly change as smartphone penetration is expected to double this year.
Mobile applications
Being the home of Apple and Google, the RBOCs are strongly challenged in the mobile applications space. Initial attempts by operators to promote the LIMO Foundation (Linux Mobile) as a more equitable partnership with application developers have largely failed, crushed by the momentum of the Apple App Store and Googles Android Market. We see little chance of operators re-gaining lost ground.
Investment thesis
We rate Verizon Overweight and increase our target price to USD41 (from USD37). The firm is well-placed to exploit mobile capacity scarcity, we believe, and we expect particularly strong data revenue momentum this year driven by the iPhone.
156
abc
Financials & valuation: Verizon Communications

Overweight
12/2012e 12/2013e
106,565 33,154 -16,349 16,805 -2,523 14,790 17,343 -3,189 3,933 6,479
110,599 36,422 -16,505 19,917 -2,681 17,759 18,034 -3,175 6,300 6,539
118,385 39,640 -16,487 23,153 -2,507 21,185 21,185 -4,079 7,791 7,791
123,006 41,235 -16,633 24,601 -2,051 23,105 23,105 -4,560 8,662 8,662
Verizon revenues Verizon EBITDA adjusted Verizon EBIT adjusted Verizon PBT adjusted Verizon EPS adjusted
106,565 35,911 19,866 17,343 2.28
110,599 36,697 20,716 18,034 2.30
118,385 39,640 23,692 21,185 3.05
123,006 41,235 25,156 23,105 3.32
34,767 -16,458 -11,971 -5,412 -14,179 10,953
30,595 -16,792 -16,792 -5,395 -7,926 13,645
33,514 -16,767 -16,767 -5,348 -11,398 15,843
34,887 -17,074 -17,074 -6,029 -11,783 17,549
1.7 5.6 1.0 16.0 2.5 7.9 5.3
1.6 4.8 1.0 15.8 2.4 10.0 5.6
1.4 4.0 0.9 13.3 2.3 11.8 5.8
1.2 3.5 0.8 11.9 2.2 13.5 6.1
100,814 87,711 22,348 7,216 220,005 23,055 52,794 45,578 41,660 180,602
99,359 89,453 25,319 9,890 223,786 22,853 47,542 37,652 42,612 181,388
97,947 91,145 37,885 21,288 237,171 25,162 47,542 26,254 44,374 180,527
97,007 92,526 50,361 33,071 250,643 26,419 47,542 14,471 46,704 180,403
Issuer information
36.39 Target price (USD) VZ.N 101,892 100 United States Richard Dineen
Bloomberg (Equity) VZ US Market cap (USDm) 101,892 Enterprise value (USDm) 174671 Sector Diversified Telecoms Contact 1 212 525 6707
43 38 33 43 38 33 28 23 18 2010
Rel to S&P 500

Ratios (%)
-1.2 8.5 16.8 24.3 -4.4
3.8 9.9 18.5 20.1 0.9
7.0 8.8 16.2 19.3 19.1
3.9 4.0 6.3 9.1 11.2
28 23 18 2009
Verizon Communications
2011
2012
0.6 8.2 15.6 6.1 31.1 15.8 13.1 52.4 1.4 76.3
0.6 9.7 15.5 7.7 32.9 18.0 13.6 39.2 1.0 81.3
0.7 11.0 17.9 8.4 33.5 19.6 15.8 24.5 0.7 127.7
0.7 11.4 19.0 8.6 33.5 20.0 20.1 12.1 0.4 241.1
Source: HSBC
1.38 2.28 1.93 14.66
2.22 2.30 2.02 15.00
2.74 2.74 2.12 15.62
3.05 3.05 2.23 16.44
157
abc
Vodafone (Overweight, TP 230p)

Mobile connectivity
Vodafone was one of the first operators globally to push for more sophisticated data charging structures. At the companys Q3 2010/11 results, management highlighted that tiered data tariffs had now been introduced in eight of its European markets, and that the remainder were to follow during the course of the present quarter. Additionally, the UK network had undertaken a successful experiment with what it has termed, smart notification. This involves contacting those customers whose data consumption is approaching the top end of their data cap, and either offering to sell them an incremental block of usage or suggesting that they take a plan with a greater allocation. This approach is shortly to be extended to the Dutch and German markets, and then rolled out across the rest of the companys European footprint. However, we also remain of the view that the tremendous growth in data volumes will require capex to support. Indeed, it has always been one of Vodafones strengths that it has generally outinvested its rivals to maintain network quality leadership. At Vodafones interim 2010/11 results in November, management said that capex was likely to rise, albeit modestly. We think that it will become easier to justify the incremental investment to the markets as the appeal of the data growth story becomes progressively clearer. Meanwhile, though, the experience of Vodafones Australian JV with Three exposes the danger of running an infrastructure inadequate to users demands. There is even talk (see ABC News, 27 December 2010) about the potential of a class action lawsuit from disgruntled customers frustrated with the inadequate coverage and bandwidth. (We should stress at this point that the operation here is too small a part of the overall group to undermine our positive stance on the stock).
Mobile applications
Vodafone has made some efforts in the mobile applications layer to counter the strides taken by Apple and Google, but initiatives like its 360 platform and participation in the Wholesale Application Community (WAC) look at best more like bargaining chips aimed primarily at keeping the technology giants as co-operative as possible. Nor does Vodafone (in common with most other operators) seem to have much enthusiasm for the GSMAs RCS initiative. It is therefore difficult to avoid the conclusion that the applications layer is moving beyond the companys reach. Clearly, though, it is in Vodafones interest to ensure that the smartphone platform arena is as competitive as possible. This may mean supporting players that have lagged in this space, such as Nokia/Microsoft and RIM. We remain of the view that operators provision of handset subsidies is actually a strategic strength, as it makes it hazardous for vendors to pursue wholesale or soft SIM strategies.
Investment thesis
We believe that Vodafone should be a substantial beneficiary as capacity constraints bestow pricing power on the operators. However, much of the recent appreciation in the share price is simply due to signs of a more pragmatic approach from management on the groups portfolio of assets. Vodafones stake in China Mobile has now been sold, and funds used to buy back shares. Following statements from Verizons management, a resumption of the dividend from the US now looks very likely. In line with the company comments, there is also the prospect that Vivendi may buy full ownership of SFR. Potential triggers such as these in combination with our strategic view relating to mobile data pricing power underpin our Overweight stance. We note that consensus revenue forecasts remain well short of company guidance for growth of up to 4%.
158
abc
Financials & valuation: Vodafone Group

Overweight
03/2012e 03/2013e
Profit & loss summary (GBPm)
Cash flow summary (GBPm)
44,472 14,735 -9,997 4,738 -796 8,674 10,564 -56 8,645 7,836
45,555 14,659 -8,870 5,789 138 13,776 11,165 -1,962 11,858 8,293
46,476 14,825 -8,118 6,707 -929 11,531 11,531 -2,398 8,954 8,341
47,648 15,168 -8,060 7,108 -944 12,348 12,348 -2,632 9,513 8,913
Europe AMAP Other Revenue HSBC Adjusted operating profit
32,833 11,089 550 44,472 11,466
31,693 13,331 531 45,555 12,105
31,555 14,380 542 46,476 12,460
31,871 15,221 555 47,648 13,292
13,179 -4,841 -9,209 -4,139 -1,511 9,042
12,212 -4,684 -2,826 -4,476 -4,016 8,176
12,483 -5,391 -8,391 -4,647 490 6,098
15,456 -5,956 -5,456 -4,894 -5,781 5,576
2.0 6.0 1.0 12.1 1.0 17.0 4.6
1.9 5.8 1.0 11.3 1.1 15.0 4.9
1.8 5.8 1.0 11.0 1.0 11.2 5.3
1.7 5.4 1.0 10.3 0.9 9.9 5.7
Balance sheet summary (GBPm)
74,258 20,642 14,219 4,811 156,985 17,453 39,795 34,984 90,381 86,855
72,987 19,051 20,797 11,157 158,960 19,676 42,125 30,968 89,468 82,002
73,489 18,822 13,343 3,565 155,675 20,173 35,023 31,458 92,928 81,916
71,991 19,216 14,129 4,175 155,358 20,889 29,852 25,677 97,180 80,271
Issuer information
Share price (GBPp)
180 Target price (GBPp)
230 Potent'l return (%) 27.8
Reuters (Equity) VOD.L Market cap (USDm) 149,443 Free float (%) 100 Country United Kingdom Analyst Stephen Howard
Bloomberg (Equity) VOD LN Market cap (GBPm) 93,390 Enterprise value (GBPm) 85413 Sector Wireless Telecoms Contact 44 20 7991 6820
187 177 167 157 147 137 127 117 107 97 2009
Vodafone Group

Ratios (%)
8.4 1.7 168.3 107.1 2.3
2.4 -0.5 22.2 58.8 6.5
2.0 1.1 15.9 -16.3 2.6
2.5 2.3 6.0 7.1 6.9
187 177 167 157 147 137 127 117 107 97 2010
Rel to FTSE ALL-SHARE
2011
2012
Per share data (GBPp)
0.5 8.5 8.9 6.0 33.1 10.7 18.5 38.5 2.4 37.7
0.5 8.5 9.2 7.4 32.2 12.7 34.6 2.1 39.4
0.6 8.9 9.1 6.3 31.9 14.4 16.0 33.9 2.1 39.7
0.6 9.3 9.4 6.7 31.8 14.9 16.1 26.5 1.7 60.2
Source: HSBC
16.44 14.90 8.31 171.88
22.69 15.87 8.89 171.20
17.49 16.29 9.56 181.50
18.58 17.41 10.28 189.80
159
abc
Notes
160
abc
Disclosure appendix
Analyst Certification
The following analyst(s), economist(s), and/or strategist(s) who is(are) primarily responsible for this report, certifies(y) that the opinion(s) on the subject security(ies) or issuer(s) and/or any other views or forecasts expressed herein accurately reflect their personal view(s) and that no part of their compensation was, is or will be directly or indirectly related to the specific recommendation(s) or views contained in this research report: Stephen Howard, Nicolas Cote-Colisson, Richard Dineen, Luigi Minerva, Herve Drouet, Dominik Klarmann, Neale Anderson, Tucker Grinnan, Kunal Bajaj, Luis Hilado and Steve Scruton.
Important disclosures
Stock ratings and basis for financial analysis
HSBC believes that investors utilise various disciplines and investment horizons when making investment decisions, which depend largely on individual circumstances such as the investor's existing holdings, risk tolerance and other considerations. Given these differences, HSBC has two principal aims in its equity research: 1) to identify long-term investment opportunities based on particular themes or ideas that may affect the future earnings or cash flows of companies on a 12 month time horizon; and 2) from time to time to identify short-term investment opportunities that are derived from fundamental, quantitative, technical or event-driven techniques on a 0-3 month time horizon and which may differ from our long-term investment rating. HSBC has assigned ratings for its long-term investment opportunities as described below. This report addresses only the long-term investment opportunities of the companies referred to in the report. As and when HSBC publishes a short-term trading idea the stocks to which these relate are identified on the website at www.hsbcnet.com/research. Details of these short-term investment opportunities can be found under the Reports section of this website. HSBC believes an investor's decision to buy or sell a stock should depend on individual circumstances such as the investor's existing holdings and other considerations. Different securities firms use a variety of ratings terms as well as different rating systems to describe their recommendations. Investors should carefully read the definitions of the ratings used in each research report. In addition, because research reports contain more complete information concerning the analysts' views, investors should carefully read the entire research report and should not infer its contents from the rating. In any case, ratings should not be used or relied on in isolation as investment advice.
Rating definitions for long-term investment opportunities

Stock ratings
HSBC assigns ratings to its stocks in this sector on the following basis: For each stock we set a required rate of return calculated from the risk free rate for that stock's domestic, or as appropriate, regional market and the relevant equity risk premium established by our strategy team. The price target for a stock represents the value the analyst expects the stock to reach over our performance horizon. The performance horizon is 12 months. For a stock to be classified as Overweight, the implied return must exceed the required return by at least 5 percentage points over the next 12 months (or 10 percentage points for a stock classified as Volatile*). For a stock to be classified as Underweight, the stock must be expected to underperform its required return by at least 5 percentage points over the next 12 months (or 10 percentage points for a stock classified as Volatile*). Stocks between these bands are classified as Neutral. Our ratings are re-calibrated against these bands at the time of any 'material change' (initiation of coverage, change of volatility status or change in price target). Notwithstanding this, and although ratings are subject to ongoing management review, expected returns will be permitted to move outside the bands as a result of normal share price fluctuations without necessarily triggering a rating change.
161
abc
*A stock will be classified as volatile if its historical volatility has exceeded 40%, if the stock has been listed for less than 12 months (unless it is in an industry or sector where volatility is low) or if the analyst expects significant volatility. However, stocks which we do not consider volatile may in fact also behave in such a way. Historical volatility is defined as the past month's average of the daily 365-day moving average volatilities. In order to avoid misleadingly frequent changes in rating, however, volatility has to move 2.5 percentage points past the 40% benchmark in either direction for a stock's status to change.
Rating distribution for long-term investment opportunities

As of 15 February 2011, the distribution of all ratings published is as follows: Overweight (Buy) 49% (23% of these provided with Investment Banking Services) Neutral (Hold) Underweight (Sell) 36% 15% (20% of these provided with Investment Banking Services) (21% of these provided with Investment Banking Services)
HSBC & Analyst disclosures

Disclosure checklist Company Ticker Recent price Price Date Disclosure
AMERICA MOVIL AT&T BELGACOM BRITISH TELECOM DEUTSCHE TELEKOM ERICSSON ETIHAD ETISALAT (MOBILY) FRANCE TELECOM KPN KT CORP MOBILE TELESYSTEMS MTN GROUP NTT DOCOMO INC. PORTUGAL TELECOM SAUDI TELECOM COMPANY SPRINT NEXTEL CORP SWISSCOM TDC TELECOM ITALIA TELEFONICA TELENOR TELIASONERA TELSTRA CORP VERIZON COMMUNICATIONS VODAFONE GROUP
Source: HSBC
AMX.N T.N BCOM.BR BT.L DTEGn.DE ERICb.ST 7020.SE FTE.PA KPN.AS 030200.KS MBT.N MTNJ.J 9437.T PTC.LS 7010.SE S.N SCMN.VX TDC.CO TLIT.MI TEF.MC TEL.OL TLSN.ST TLS.AX VZ.N VOD.L
56.44 28.46 27.14 1.84 9.89 81.55 54.00 16.16 11.74 40750.00 19.90 127.10 154200.00 8.48 40.20 4.55 427.90 46.25 1.04 18.28 91.30 54.40 2.92 35.90 1.80
15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011 15-Feb-2011
1, 2, 5, 6, 7, 11 7, 11 7, 11 2, 4, 6, 7, 11 6, 7, 11 2, 7, 11 2, 5, 7 1, 2, 4, 5, 6, 7, 11 6 6, 7, 11 2, 5, 7, 11 6, 7 6 6, 11 2, 6, 7 6 6, 7 1, 5, 6, 11 6, 7, 11 1, 2, 4, 5, 6, 7, 11 6, 7, 11 6, 11 1, 2, 5, 7 6, 11 1, 2, 4, 5, 6, 7, 11
1 2 3 4 5 6 7 8 9
HSBC* has managed or co-managed a public offering of securities for this company within the past 12 months. HSBC expects to receive or intends to seek compensation for investment banking services from this company in the next 3 months. At the time of publication of this report, HSBC Securities (USA) Inc. is a Market Maker in securities issued by this company. As of 31 January 2011 HSBC beneficially owned 1% or more of a class of common equity securities of this company. As of 31 December 2010, this company was a client of HSBC or had during the preceding 12 month period been a client of and/or paid compensation to HSBC in respect of investment banking services. As of 31 December 2010, this company was a client of HSBC or had during the preceding 12 month period been a client of and/or paid compensation to HSBC in respect of non-investment banking-securities related services. As of 31 December 2010, this company was a client of HSBC or had during the preceding 12 month period been a client of and/or paid compensation to HSBC in respect of non-securities services. A covering analyst/s has received compensation from this company in the past 12 months. A covering analyst/s or a member of his/her household has a financial interest in the securities of this company, as detailed below.
162
abc
10 11
A covering analyst/s or a member of his/her household is an officer, director or supervisory board member of this company, as detailed below. At the time of publication of this report, HSBC is a non-US Market Maker in securities issued by this company and/or in securities in respect of this company
Analysts, economists, and strategists are paid in part by reference to the profitability of HSBC which includes investment banking revenues. For disclosures in respect of any company mentioned in this report, please see the most recently published report on that company available at www.hsbcnet.com/research.
* HSBC Legal Entities are listed in the Disclaimer below.
Additional disclosures
1 2 3 This report is dated as at 16 February 2011. All market data included in this report are dated as at close 14 February 2011, unless otherwise indicated in the report. HSBC has procedures in place to identify and manage any potential conflicts of interest that arise in connection with its Research business. HSBC's analysts and its other staff who are involved in the preparation and dissemination of Research operate and have a management reporting line independent of HSBC's Investment Banking business. Information Barrier procedures are in place between the Investment Banking and Research businesses to ensure that any confidential and/or price sensitive information is handled in an appropriate manner. As of 31 January 2011, HSBC and/or its affiliates (including the funds, portfolios and investment clubs in securities managed by such entities) either, directly or indirectly, own or are involved in the acquisition, sale or intermediation of, 1% or more of the total capital of the subject companies securities in the market for the following Company(ies): FRANCE TELECOM , BRITISH TELECOM , TELEFONICA , VODAFONE GROUP As of 21 January 2011, HSBC owned a significant interest in the debt securities of the following company(ies): FRANCE TELECOM , TELEFONICA , KT CORP , TELECOM ITALIA , DEUTSCHE TELEKOM
163
abc
Disclaimer
* Legal entities as at 31 January 2010 Issuer of report 'UAE' HSBC Bank Middle East Limited, Dubai; 'HK' The Hongkong and Shanghai Banking Corporation HSBC Bank plc Limited, Hong Kong; 'TW' HSBC Securities (Taiwan) Corporation Limited; 'CA' HSBC Securities (Canada) Inc, Toronto; HSBC Bank, Paris branch; HSBC France; 'DE' HSBC Trinkaus & Burkhardt AG, 8 Canada Square Dusseldorf; 000 HSBC Bank (RR), Moscow; 'IN' HSBC Securities and Capital Markets (India) Private London, E14 5HQ, United Kingdom Limited, Mumbai; 'JP' HSBC Securities (Japan) Limited, Tokyo; 'EG' HSBC Securities Egypt S.A.E., Cairo; Telephone: +44 20 7991 8888 'CN' HSBC Investment Bank Asia Limited, Beijing Representative Office; The Hongkong and Shanghai Banking Corporation Limited, Singapore branch; The Hongkong and Shanghai Banking Corporation Fax: +44 20 7992 4880 Limited, Seoul Securities Branch; The Hongkong and Shanghai Banking Corporation Limited, Seoul Website: www.research.hsbc.com Branch; HSBC Securities (South Africa) (Pty) Ltd, Johannesburg; 'GR' HSBC Pantelakis Securities S.A., Athens; HSBC Bank plc, London, Madrid, Milan, Stockholm, Tel Aviv, 'US' HSBC Securities (USA) Inc, New York; HSBC Yatirim Menkul Degerler A.S., Istanbul; HSBC Mxico, S.A., Institucin de Banca Mltiple, Grupo Financiero HSBC, HSBC Bank Brasil S.A. - Banco Mltiplo, HSBC Bank Australia Limited, HSBC Bank Argentina S.A., HSBC Saudi Arabia Limited., The Hongkong and Shanghai Banking Corporation Limited, New Zealand Branch. In the UK this document has been issued and approved by HSBC Bank plc (HSBC) for the information of its Clients (as defined in the Rules of FSA) and those of its affiliates only. It is not intended for Retail Clients in the UK. If this research is received by a customer of an affiliate of HSBC, its provision to the recipient is subject to the terms of business in place between the recipient and such affiliate. HSBC Securities (USA) Inc. accepts responsibility for the content of this research report prepared by its non-US foreign affiliate. All U.S. persons receiving and/or accessing this report and wishing to effect transactions in any security discussed herein should do so with HSBC Securities (USA) Inc. in the United States and not with its non-US foreign affiliate, the issuer of this report. In Singapore, this publication is distributed by The Hongkong and Shanghai Banking Corporation Limited, Singapore Branch for the general information of institutional investors or other persons specified in Sections 274 and 304 of the Securities and Futures Act (Chapter 289) (SFA) and accredited investors and other persons in accordance with the conditions specified in Sections 275 and 305 of the SFA. This publication is not a prospectus as defined in the SFA. It may not be further distributed in whole or in part for any purpose. The Hongkong and Shanghai Banking Corporation Limited Singapore Branch is regulated by the Monetary Authority of Singapore. Recipients in Singapore should contact a "Hongkong and Shanghai Banking Corporation Limited, Singapore Branch" representative in respect of any matters arising from, or in connection with this report. In Australia, this publication has been distributed by The Hongkong and Shanghai Banking Corporation Limited (ABN 65 117 925 970, AFSL 301737) for the general information of its wholesale customers (as defined in the Corporations Act 2001). Where distributed to retail customers, this research is distributed by HSBC Bank Australia Limited (AFSL No. 232595). These respective entities make no representations that the products or services mentioned in this document are available to persons in Australia or are necessarily suitable for any particular person or appropriate in accordance with local law. No consideration has been given to the particular investment objectives, financial situation or particular needs of any recipient. This publication has been distributed in Japan by HSBC Securities (Japan) Limited. It may not be further distributed, in whole or in part, for any purpose. In Hong Kong, this document has been distributed by The Hongkong and Shanghai Banking Corporation Limited in the conduct of its Hong Kong regulated business for the information of its institutional and professional customers; it is not intended for and should not be distributed to retail customers in Hong Kong. The Hongkong and Shanghai Banking Corporation Limited makes no representations that the products or services mentioned in this document are available to persons in Hong Kong or are necessarily suitable for any particular person or appropriate in accordance with local law. All inquiries by such recipients must be directed to The Hongkong and Shanghai Banking Corporation Limited. In Korea, this publication is distributed by The Hongkong and Shanghai Banking Corporation Limited, Seoul Securities Branch ("HBAP SLS") for the general information of professional investors specified in Article 9 of the Financial Investment Services and Capital Markets Act (FSCMA). This publication is not a prospectus as defined in the FSCMA. It may not be further distributed in whole or in part for any purpose. HBAP SLS is regulated by the Financial Services Commission and the Financial Supervisory Service of Korea. This publication is distributed in New Zealand by The Hongkong and Shanghai Banking Corporation Limited, New Zealand Branch. This document is not and should not be construed as an offer to sell or the solicitation of an offer to purchase or subscribe for any investment. HSBC has based this document on information obtained from sources it believes to be reliable but which it has not independently verified; HSBC makes no guarantee, representation or warranty and accepts no responsibility or liability as to its accuracy or completeness. The opinions contained within the report are based upon publicly available information at the time of publication and are subject to change without notice. Nothing herein excludes or restricts any duty or liability to a customer which HSBC has under the Financial Services and Markets Act 2000 or under the Rules of FSA. A recipient who chooses to deal with any person who is not a representative of HSBC in the UK will not enjoy the protections afforded by the UK regulatory regime. Past performance is not necessarily a guide to future performance. The value of any investment or income may go down as well as up and you may not get back the full amount invested. Where an investment is denominated in a currency other than the local currency of the recipient of the research report, changes in the exchange rates may have an adverse effect on the value, price or income of that investment. In case of investments for which there is no recognised market it may be difficult for investors to sell their investments or to obtain reliable information about its value or the extent of the risk to which it is exposed. HSBC Bank plc is registered in England No 14259, is authorised and regulated by the Financial Services Authority and is a member of the London Stock Exchange. Copyright. HSBC Bank plc 2011, ALL RIGHTS RESERVED. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, on any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of HSBC Bank plc. MICA (P) 142/06/2010 and MICA (P) 193/04/2010
290898
164
Main Contributors
Tucker Grinnan* Analyst, Global Telecoms, Media & Technology Research The Hongkong and Shanghai Banking Corporation Limited, Hong Kong +852 2822 4686 tuckergrinnan@hsbc.com.hk Tucker joined HSBC in November 2005 and has 15 years of experience as an analyst in the telecommunications, media and technology industries. Prior to joining HSBC, he spent five years as a head of regional telecoms for Asia and Latin America. Tucker also spent five years with a management consulting firm servicing clients in telecommunications and media. He holds degrees from the University of Virginia and George Washington University. Luis A Hilado* Analyst, Global Telecoms, Media & Technology Research The Hongkong and Shanghai Banking Corporation Limited, Singapore +65 6239 0656 luishilado@hsbc.com.sg Luis Hilado joined HSBC in 2010 from a US investment bank, where he covered the SE Asia telecom sector. He has over 15 years of equity research experience primarily covering SE Asia telecoms. Luis held a number of positions on the sell-side including Head of Research, Philippines for a European stock brokerage. He holds a BA in Economics and a BS in Commerce and Business Management from De La Salle University. Stephen Howard* Head, Global Telecoms, Media & Technology Research HSBC Bank plc +44 20 7991 6820 stephen.howard@hsbcib.com Stephen Howard is Head of the Global Telecoms, Media & Technology Research team. He has covered the telecoms sector since joining HSBC in 1996. He also brings experience in the technology industry, having worked previously with IBM. Dominik Klarmann*, CFA Analyst, Global Telecoms, Media & Technology Research HSBC Trinkaus & Burkhardt AG, Dusseldorf +49 211 910 2769 dominik.klarmann@hsbctrinkaus.de Dominik Klarmann has worked as a telecoms analyst since 2007. He obtained a degree in management at Bamberg and Madrid University in 2004. He has been with HSBC since 2007, having previously worked in management consulting and investor relations at Deutsche Telekom AG. Dominik is a CFA charterholder. Luigi Minerva* Analyst, Global Telecoms, Media & Technology Research HSBC Bank plc +44 20 7991 6928 luigi.minerva@hsbcib.com Luigi joined HSBC in 2005 as a telecoms analyst in the Global Telecoms Research team. Before this, he was a buy-side analyst at a fund manager, having previously worked in the Telecoms Practice of McKinsey & Co based in Milan. Luigi holds a Masters in Finance from the London Business School and an M.Sc. in Economics and Econometrics from the University of Southampton. Kunal Bajaj* Analyst, Global Telecoms, Media & Technology Research HSBC Bank Middle East Limited, Dubai +9714 507 7200 kunalbajaj@hsbc.com Kunal joined HSBC in 2005. Before covering Middle Eastern telecoms, he was a member of HSBCs EMEA telcos team, working on Eastern European and South African telecoms companies. Kunal is a Chartered Accountant and has an MBA in Finance. Neale Anderson* Analyst, Global Telecoms, Media & Technology Research The Hongkong and Shanghai Banking Corporation Limited, Hong Kong +852 2996 6716 neale.anderson@hsbc.com.hk Neale Anderson joined HSBC in March 2007. Previously he spent seven years at the specialist consultancy Ovum, where he was Research Director for Asia-Pacific telecommunications markets. He holds a BA from Oxford University and an MA in Advanced Japanese Studies from Sheffield University. Nicolas Cote-Colisson* Head of European Telecoms, Media & Technology Research HSBC Bank plc +44 20 7991 6826 nicolas.cote-colisson@hsbcib.com Nicolas joined HSBC in 2000 as a telecoms analyst in the Global Telecoms research team. Prior to that, he worked as an economist with CCF in Paris for five years and also with the French Ministry of Finance. Nicolas holds a DEA in Econometrics from Paris la Sorbonne. Richard Dineen Analyst, Global Telecoms, Media & Technology Research HSBC Securities (USA) Inc, New York +1 212 525 6707 richard.dineen@us.hsbc.com Richard joined HSBC in 2004 to work on the Global Telecoms Research team, with a particular emphasis on technology strategy. Prior to this, he was research director for Mobile Telecoms at Ovum, a highly respected industry analyst firm, where he spent seven years. Herv Drouet* Analyst, Global Telecoms, Media & Technology Research HSBC Bank plc +44 20 7991 6827 herve.drouet@hsbcib.com Herv has been covering GEMs/CEEMEA Telecoms research for more than 9 years and has been ranked highly and regularly across numerous external surveys. Prior to this, he worked as a senior management consultant in the TMT practice at Deloitte Consulting. He has 15 years experience in the Media, Telecoms and Technology sectors, having worked previously as a project manager for Schlumberger Technologies. He holds a Full time MBA from London Business School and graduated from Ecole Suprieure d'Ingnieurs en Electrotechnique et Electronique in France. Steve Scruton* Head of Equity Research, CEEMEA HSBC Bank plc +44 20 7992 3602 steve.scruton@hsbcib.com Steve Scruton is the Head of Equity Research, CEEMEA. He has been with HSBC since 1997, and was previously the co-head of Global TMT in London and Head of Research, Bangalore, a team that provides support to Global Research across countries, sectors, products and services. Prior to joining HSBC, Steve worked with British Petroleum, Cable & Wireless and a banking house in London.
*Employed by a non-US affiliate of HSBC Securities (USA) Inc, and is not registered/qualified pursuant to FINRA regulations.

HSBC - Abundant Scarcity

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

HSBC - Abundant Scarcity

Transféré par

Droits d'auteur :

Formats disponibles

Global Telecoms, Media & Technology Equity February 2011

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Applications layer Connectivity layer

Eg IPTV Eg superfast broadband Fixed-line network

Eg mobile apps Eg tiered data plans Mobile network

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Winners and losers

Telecoms, Media & Technology Equity 16 February 2011

Global Telecoms: Coverage universe

TPSA Turk Tel Zain Group*

PLN TRY KWD

15.3 6.8 1.2

16.6 7.1 1.4

AMX AT&T Inc. Tele Norte Telesp Sprint Nextel*

USD USD BRL BRL USD

64.0 30.0 31.0 41.0 5.0

56.8 28.5 26.1 39.5 4.6

Source: HSBC estimates, priced as at 11 February 2011

Net Serv Telmex

Telecoms, Media & Technology Equity 16 February 2011

Disclosure appendix Disclaimer

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

No cable operator in Italy 2015e

Cable (key competitor)

Source: company data, HSBC estimates

Telecoms, Media & Technology Equity 16 February 2011

German broadband market in 2007

TEF O2 1% Cable 4% Versatel 3% freenet 7% Alice 12%

Vodafone 14% TEF O2 8% Cable 13%

Source: company data, HSBC

Telecoms, Media & Technology Equity 16 February 2011

NGA builds encourage shift from unbundled products to wholesale

Status quo Status quo

NGA builds NGA builds

Telecoms, Media & Technology Equity 16 February 2011

Incipient fixed-line pricing power

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

End user building

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Could try harder

Deep packet inspection

Policy Engine (PE)

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Non-discrimination and transparency

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Rise of the CDN

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Telecoms, Media & Technology Equity 16 February 2011

Phantom of the copper-a

Telecoms, Media & Technology Equity 16 February 2011

Vectoring mitigates FEXT between signal wires

Signal wire Phantom cir cuit