Significance: Deutsche Telekom, having been thwarted in its aim of selling its T-Mobile USA arm to AT&T last year, has confirmed a merger with flat rate player MetroPCS. There is logic behind the move, mainly that the companies’ combined spectrum holdings will enable TMo to build an LTE network entirely in the AWS band, and a huge 40MHz one in some key cities. However, the new entity faces several years of difficult transition, since 2G and 3G services have to be moved out of much of the spectrum, plus it will face dilemmas over branding, roaming and pricing. During those years, it will have to sustain two incompatible 2G/3G networks as well as investing in LTE. It may not be quite as awful as Sprint’s merger with Nextel, but there are significant challenges on the network, device and marketing fronts. The merger will not enable TMo to challenge Verizon and AT&T in any significantly new way, though it will strengthen it against Sprint. Increasingly, as senior executives virtually conceded this week, these operators are stuck in the ‘value’ market, and will be forced to harness their new and expensive LTE capacity to make them more capable of waging an aggressive price war in the prepaid segment. If DT’s end game is to exit the US, it should try to move quickly.
In the past week we have seen two mobile mergers announced ‘ Softbank buying eMobile in Japan, and T-Mobile wedding MetroPCS in the US. There are some similarities. In both cases the larger firm wants to eliminate a smaller competitor, add to an already complicated mix of 4G spectrum, and gain some scale to challenge the country’s big two cellcos.
There the parallels end. In Japan, one of the world’s most advanced mobile carriers is snapping up a genuinely disruptive challenger, boosting its already strong chances of gaining ground on leaders Docomo and KDDI (see inset). In the US, the two partners are both squeezed into the overcrowded midmarket, while struggling with falling margins, spectrum constraints and operational difficulties.
There is on-paper logic to the TMo/MetroPCS deal, and by increasing their scale they can certainly get closer to third operator Sprint. The problem is, they are chasing Sprint by repeating its mistakes when it acquired Nextel, and while they struggle with those challenges, Verizon and AT&T will look down and laugh at all their rivals locked in an intensifying price war at the low end, while they calmly increase their own market leads in 4G.
The deal sees TMo parent Deutsche Telekom paying $1.5bn in cash to MetroPCS shareholders, as part of a stock and debt restructuring which leaves the German giant with 74% of the merged firm. This will trade on the New York Stock Exchange. There will be some way to go to overtake Sprint for the number three position. Combined, the new entity would have had 42.3m customers at the end of June, while Sprint had 56m. MetroPCS has previously been in talks with its fellow flat rate operator and national roaming partner Leap Wireless, and with Sprint, though Sprint’s board apparently rejected a merger proposal earlier this year.
Many believe this is a ‘shotgun wedding’, as a Bernstein Research note put it, with DT’s main objective being to begin the process of exiting the US. Bernstein points out that US stocks have traded at a 25% premium over European shares relative to earnings during the past decade, so the merger will help DT get more value from its asset. Shing Yin of Guggenheim Securities said he was ‘not sure it makes that much of a difference’ to the big two cellcos, but the deal did give DT ‘more options – to hold on to it, or to sell the shares in the open market, or spin out the shares to shareholders.’
Of course, unlike the Japanese deal, the primary driver behind the TMo agreement is ‘ as usual in the current US mobile climate ‘ to bolster a spectrum position for mobile broadband. The results of the merger in this respect are mixed, adding capacity but also complexity for the fourth cellco.
After the failure of TMo’s plan to be acquired by AT&T, and the period of paralysis that followed the collapse of that deal, the operator has played an unexpectedly strong hand in improving its previously lamentable spectrum position for 4G. It is repurposing its AWS spectrum, where it was the only carrier running 3G, for the more mainstream LTE, which will improve its access to key devices like the iPhone. It has also gained significant additional tranches in the AWS band thanks to its break-up deal with AT&T, and spectrum swaps with Verizon Wireless and Leap. Although it has to undergo a complex double refarming exercise over the coming months, shifting HSPA+ to the PCS band, turning off some 2G services, and launching LTE in AWS, the end result was looking to be a clean implementation in just one LTE band, with decent coverage and capacity in most major markets. That would do away with the need ‘ at least for the time being ‘ to acquire more spectrum or engage in a deal with Clearwire, a commonly suggested option.
By adding MetroPCS to the mix, TMo has one big advantage ‘ it can keep its LTE network on one band, since its prey also operates in AWS. This promises a quick consolidation of the 4G systems next year, and an earlier LTE base than TMo had expected, if the merger goes through reasonably quickly. (Good news for Ericsson too, which is MetroPCS’s supplier, and shares the TMo LTE deal with Nokia Siemens.) Regulatory objections are not expected, given that this theoretically addresses one of the chief qualms about the AT&T/TMo plan, the removal of the fourth operator. Although the total number of cellcos does fall, there is an argument that TMo is now more likely to survive and be a strong challenger to the big two, than it would if left alone.
In the top 25 markets, T-Mobile will see its average spectrum holdings increase from 63MHz to 76MHz, and much of the gain will be in AWS. Currently, it has spectrum to support a 20MHz network in half of its metro areas, but only 10MHz elsewhere. Its acquisition does not fill all the gaps, but does enable TMo to build a huge 40MHz LTE network ‘ outdoing the big two ‘ in 11 major cities. In others though, such as Chicago, its position will be unchanged. As the map, created by coverage mapping firm Mosaik for GigaOM, shows, the best cities include Los Angeles, San Francisco, Dallas, Boston, Detroit, Atlanta and Miami, while New York and Philadelphia will also see a decent combined capacity.
If this were an LTE-only world, all this would make a strong case. A single-band LTE network, capable of roaming with Verizon and AT&T once they expand from 700MHz into AWS next year, and the removal of a competitor in the area where TMo has seen most growth recently, prepaid. However, it is not an LTE-only world. The majority of users, especially in the midmarket where TMo is most comfortable, will remain on 3G for two or more years. And that raises the specter of having to manage two incompatible older networks ‘ TMo’s HSPA(+) and MetroPCS’s CDMA ‘ with all the implications for economies of scale, roaming and device availability. Being able to buy all its LTE devices in one band has real advantages, but not if they all have to support PCS/HSPA and CDMA as well. Not to mention the complexities of adapting back office and management systems for three networks, and the dilemmas inherent in maintaining good customer service and quality of experience, on systems which the carrier just wants to shut down.
Sprint and Nextel went through this with CDMA and iDEN, and the mishandling of the situation helped create problems which still dog the carrier today. They will only really go away once it finishes its ambitious Network Vision project, which will support all its air interfaces on flexible base stations and a common core, allowing for all sites to be upgraded to LTE, in all its bands, over time, and to suit the business model. This is one extreme of what TMo could choose to do with MetroPCS, but it is an expensive and time consuming task and would waste some of the efforts it has already made in its HSPA+/LTE project. It would also be costly, and unlikely to gain support from parent Deutsche Telekom, which many believe is eyeing an exit from the US once it has boosted the value of its subsidiary.
At the other extreme, TMo could keep the two legacy networks entirely separate and only gain the benefits of expanded coverage and capacity when it comes to LTE. Then the challenge will be to move users as quickly as possible to the new networks to maximize the efficiencies and reduce the risk of a mass exodus of subscribers from ageing or under-supported 3G systems. This seems to be the preferred approach for new TMo USA CEO John Legere, and, he insisted, was why TMo would avoid the problems of Sprint.
‘The soundbite that this is a Sprint-Nextel do-over is absolutely completely wrong,’ he said on the analyst call to discuss the merger. ‘Nothing could be further from the truth.’ The difference is TMo will not try to keep MetroPCS’s CDMA network running for longer than absolutely necessary, but will put all its efforts into unifying the networks as soon as possible. In particular, it will market HSPA and LTE devices to MetroPCS users to try to accelerate the switch-off of CDMA. Of course, that can be tough since prepaid customers have to pay for their handsets upfront, but Legere said 60% of the cellco’s base swap devices on an annual basis, meaning it could convert most of the subscribers within two years.
He said the new carrier would then dominate the prepaid and budget mobile markets by leveraging the new LTE capacity to offer flexible pricing and heavyweight data packages. ‘When you add MetroPCS to an already aggressive challenger strategy, it acts as an accelerant,’ Legere said. He added that the new firm would be able to offer unlimited data plans to prepaid as well as contract subscribers.
This will certainly go head-to-head with Sprint, which has clung to unlimited tariffs as a key differentiator and aims to use LTE to improve its capability there, while Verizon and AT&T look to harness 4G to push their applications and their ARPUs upmarket
Many observers were unconvinced by the companies’ optimism that they would make smooth work of combining incompatible current networks (CDMA at MetroPCS, GSM/HSPA at TMo) at a time when both are going through complex network transitions, the former to LTE, the latter moving HSPA+ into refarmed 2G spectrum while preparing to launch LTE in AWS bands. ‘These are not two robust, growing entities that are combining,’ Kevin Smithen, an analyst at Macquarie Securities, told Bloomberg. ‘These are two companies that had significant operations issues. We are skeptical.’ And someone who should know pointed out the pitfalls. Paul Saleh, who was CFO at Nextel when it was bought by Sprint in 2005 and served as interim CEO in 2008, said: ‘I think it is easier said than done’ and that TMo’s task was ‘equally challenging’. Indeed, he said Sprint and Nextel were planning to migrate both their user bases to a common third platform too, like TMo, but that proved harder than expected.
And even if they manage the transition well, there are the challenges of trying to break out of the low margin prepaid sector. LTE cannot be guaranteed to shift the balance to high value or postpaid users, since these are most likely to choose between AT&T and Verizon ‘ and indeed Legere sounded as though he had already ceded that battle, talking almost entirely about how increased capacity supports a value model. If TMo can offer something different or better, it will gain some of these subscribers, but that will mean innovative thinking in services and pricing, since its spectrum and coverage position does not give it the upper hand over its two larger rivals in too many markets.
Even DT’s CEO Rene Obermann was stressing the ‘value’ angle in the combined strategy. ‘The T-Mobile and MetroPCS brands are a great strategic fit – both operationally and culturally,’ he said. ‘The new company will be the value leader in wireless with the scale, spectrum and financial and other resources to expand its geographic coverage, broaden choice among all types of customers and continue to innovate, especially around the next generation LTE network.’ But with MetroPCS’s customers all highly liable to churn because of the lack of contracts, TMo will have a delicate balancing act ‘ keeping those users loyal while improving its overall value proposition and margin. Its simplest option will be to operate its new unit as a separate brand, as Sprint does with Virgin Mobile.
Additional scale and spectrum will make TMo more viable, but the new management will not need to be thinking about keeping Verizon and AT&T awake at night. Their target is clearly Sprint and the value sector, and Bernstein sees the positive of the merger resting on nothing more disruptive than having more ammunition for the price war in the midmarket.
In the end, the plans sound hopelessly over-optimistic. TMo will learn from the Sprint disaster and will not endure the same problems, or at least not for so long. However, it has a complex task ahead. Both firms have to clear their AWS holdings of 2G and 3G services to gain the full LTE capacity, and even with the moves to convert MetroPCS subscribers, Legere admitted that could take until the end of 2015 to complete. That means three years when the company has not yet got to its ideal spectrum situation, and is also struggling with the inevitable debates and uncertainties over whether to operate one or two brands; how to hold onto low end customers for market share, while avoiding a precipitate shift of the base towards skinny margins; how to support roaming and integration while separate networks are still in play. A three-year transition, for a company in a poor competitive position and with an indebted parent begrudging further investment, is a tough call. Unlike Softbank in Japan, the merged firm will be making its changes in a period when the main two carriers are riding high, hoovering up the high end users with their LTE headstarts and their virgin spectrum. The US market is set to polarize between the high value duo and the rest, with Sprint and TMo investing vast sums in bringing LTE to their legacy bands, only to find themselves using the resulting capacity mainly to fight a data price war.
T-Mobile USA sells cell towers to Crown Castle
Deutsche Telekom has agreed the long expected sale of 7,200 cell towers owned by its T-Mobile USA unit, to infrastructure provider Crown Castle.
The latter firm will pay $2.4bn for the rights to operate the towers, a strategy DT adopted after the failure of its attempt to sell its US arm to AT&T. The sale will provide cash to support TMo’s LTE roll-out, which it has to accomplish on its own after the merger fell through, using a complex set of refarming initiatives in its 2G and 3G spectrum, plus additional AWS licences it gained as part of a break-up fee from AT&T.
Without the deep pockets of a top two US cellco, and with DT seeking to reduce debt and cope with the eurozone crisis in its home region, TMo needs to gain funding. The towers were always seen as the most straightforward route to this, and Crown Castle gains exclusive rights to lease and operate the infrastructure for 28 years, with the option to acquire them in full at the end of the lease. The deal should close during this quarter. T- Mobile can rent back antenna space on the towers.
Analysts were impressed with the price paid, which indicates the high demand for tower space as US carriers try to keep up with rising mobile broadband demand. Despite an increasing focus on expanding capacity via small cells, which are mounted on rooftops or lamp posts rather than towers, there is also significant investment in additional macrocells, while carriers are inclined to rent space from third parties rather than owning their own towers.
‘They got a very steep price at $2.4bn,’ Kevin Smithen, an analyst with Macquarie Capital, told Bloomberg. ‘And the 28-year deal signifies that T-Mobile is likely going to be in the US for the long term.’ Others also saw the structure of the agreement as a signal that DT had gone cold on efforts to sell off its US arm, at least until it can boost the cellco’s value significantly with programs to become a 4G player and improve its margins. ‘T-Mobile got a good amount of cash and good lease terms,’ added James Ratcliffe of Barclays Capital. ‘This frees up financial capacity for upgrades in the US, and they don’t need to be as dependent on financing from Germany.’
Parent DT said in a statement from CFO Timotheus Hottges: ‘We have found an intelligent way to strengthen T-Mobile USA among competitors and reduce the group’s net debt at the same time.’ The deal will reduce DT’s debt by €1.9bn ($2.4bn).
TMo USA announced a new CEO last month. Former Global Crossing CEO John Legere took the helm, pledging to restart subscriber growth and aim to overtake third placed Sprint. In February, DT said it would boost its US network spending budget by $1.4bn over two years to win back customers with 4G services.
‘T-Mobile USA is working aggressively to make our 4G network stronger, faster and more dependable for consumers, and this transaction will support our ongoing $4bn network modernization initiative,’ Legere said in his statement.
In 2008, Sprint sold about 3,300 towers to TowerCo to raise $670m in cash.
Meanwhile, Leap Wireless has completed a spectrum swap deal with TMo, to improve the coverage both firms offer in certain markets. The deal also involves Cook Inlet/VS GSM VII PCS, a joint venture between T-Mobile and Cook Inlet, and Leap’s majority-owned venture Savary Island Wireless. Under the transaction agreements, Leap acquired 10MHz of additional AWS spectrum in Phoenix, Arizona and in three Texas markets, while TMo and Cook Inlet received spectrum in parts of Alabama, Illinois, Missouri, Minnesota and Wisconsin. The transactions also included intra-market exchanges between Leap and T-Mobile in Philadelphia, Wilmington, Atlantic City and various markets in New Mexico and Texas, and between Leap and Cook Inlet in certain markets in Texas.
Latest FCC activities:
The FCC has has proposed opening up spectrum in the WCS band (2.3GHz) for 4G services, which will be good news for AT&T. The cellco has been buying up licences in this formerly unloved band, notably in a $650m deal with NextWave Wireless. The main problem with the spectrum in the US is that the band is bisected by the satellite radio frequencies held by Sirius. However, in June, Sirius and AT&T proposed a compromise over how to deploy LTE close to the radio services without causing interference. The operator still needs FCC approval for its plans, but a proposed order from the agency’s chairman is usually a good sign. The WCS spectrum is far less attractive than AWS in capacity or band plan, and AT&T has said it will use it for ‘alternative’ LTE services rather than its mainstream network. These could include fixed broadband or backhaul.
The FCC established rulemaking to initiate incentive auctions for additional TV broadcast spectrum, which could then be repurposed for mobile services, with some of the frequencies being opened up as licence exempt white spaces. The details of the channels and locations are left open as part of the voluntary auction process. As much as 120MHz of spectrum will become available though auctions are not likely to take place until 2014.
The regulator, burned by its over-hasty approval of LightSquared’s plan to run LTE in mobile satellite (MSS) spectrum ‘ killed by issues of interference with nearby GPS ‘ has initiated a review of all the MSS bands. In particular, it is looking at how they might be aligned with AWS. Dish Network’s chairman Charlie Ergen pushed the regulator this week to pass rules allowing it to use its AWS-4 (former mobile satellite) spectrum for a 4G network, otherwise it says it may be too late to be a viable contender and would then look to sell its holdings. Dish cannot use its 40MHz of spectrum for LTE without changes to regulations on the former satellite band. Ergen added that it ‘doesn’t really make sense any more’ to build its wireless network from scratch, and that it would instead look to partner (Clearwire being an obvious candidate). But he added: ‘We’d prefer not to sell the spectrum, we’d prefer to enter the business.’ Sprint is pushing the FCC to shift Dish Network’s AWS-4 band up 5MHz from 2000-2020MHz to 2005-2025MHz so adjacent H block PCS spectrum can be used for LTE. But Dish says allowing high power operations in the H block will ‘render at least 25% of the AWS-4 uplink band unusable.’
Another important issue is how to enforce open access rules in 700MHz and require open devices on unused portions of this band. Verizon’s swathe of 700MHz spectrum was auctioned under rules which supposedly required open devices and open access to applications. However, as principal analyst Robert Syputa of Maravedis-Rethink points out: ‘Since Verizon and AT&T control what devices appear on their networks by nature of their control over market share, in all practical regard they have locked out competitive devices and operators. That shows how screwed up the political situation is in the US today: the ‘free enterprisers’ in Congress often come down on the side of Verizon and AT&T and are ‘hands off’ to requirements that would force common devices and network access to develop. The smaller operators are left with islands of spectrum and no devices to build a viable market. That leaves the competition weak and unable to fund future auction bids.’
The FCC has voted to initiate a review of its rules regarding how much spectrum a carrier should be able to hold, aiming to achieve a more transparent process, which could benefit smaller carriers. The rulemakers will explore the agency’s spectrum screen, which it uses when reviewing frequency-based transactions. The screen is currently different for each proposed transaction.
LightSquared has filed a new proposal with the FCC to revive its plan for building a nationwide LTE network. Even from bankruptcy court, the firm is refusing to give up, even if its latest plan would take years to effect. In its latest idea, it has promised to vacate permanently the upper 10MHz of its MSS spectrum, and not use the lower 10MHz until the regulator comes up with new rules for that band, to satisfy all interested parties. In return for these concessions, LightSquared proposes using 5MHz of spectrum that is currently in the hands of federal government, and is contiguous to its own, to replace the upper reach frequencies which it will not use at all. Previously, the start-up had offered to leave the upper 10MHz idle, in order to reassure GPS players, but they were not satisfied, claiming there could still be interference with precision GPS devices lower down the spectrum.
The FCC has announced the recipients of $300m in funding to build out wireless 3G and 4G connections across 83,000 miles of road in 31 states where there are gaps in mobile service.
Wireless providers obtained the Mobility Fund money through participation in a reverse auction, in which 38 companies and their subsidiaries submitted almost 900 bids, with winning companies promising the lowest cost-per-mile in infrastructure build-out. US Cellular won $30m to cover more than 1,700 miles of road, while Allied Wireless got more than $45n to build out wireless on more than 4,400 miles of road. Most of the winners were small, rural providers.
Softbank to buy eMobile, becoming Japan’s second cellco
Softbank of Japan is to buy the country’s disruptive fourth cellco, eMobile, in a bid to leapfrog KDDI in the cellular market. It will acquire ‘180.2bn ($2.3bn) of stock in the smaller operator’s parent eAccess.
Softbank Mobile has been a serious challenger to KDDI’s second position since it bought out Vodafone KK in in 2006, and it is one of the world’s most innovative carriers in harnessing new technologies ‘ including TD-LTE, small cells and incipient HetNets ‘ to make the most of its spectrum and keep pace with the famously hi-tech NTT Docomo and KDDI. However, it lost a key commercial advantage, the iPhone exclusive, when KDDI gained an Apple device last year, and it has seen eMobile nibbling at its heels with its competitive data-focused plans.
Buying the fourth player, then, could kill two birds with one stone, removing a competitor ‘ especially one which has pulled down data ARPUs in Japan ‘ and gaining more scale to go after the big two. According to Reuters, the ‘52,000-per-share deal represents a premium of more than three times on eAccess’s closing price on Friday. ‘We are paying more than triple the current trading price of eAccess because the benefits come from the company’s existing customer base,’ Softbank Mobile’s founder Masayoshi Son told reporters. ‘We do not think our shareholders will view this as a bad tie-up at all.’
The company is targeting synergies of ‘360bn, split between customer base expansion (‘200bn), network sharing (‘110bn) and ‘other management efficiencies’ (’50bn). In addition, Softbank will gain more spectrum for its already mixed bag of FDD and TDD frequencies, the latter also the result of acquiring a smaller player, the 4G element of PHS operator Willcom. eMobile has a network of 10,000 3G and/or LTE base stations running in the 1.7GHz band while Softbank boasts 20,000 in 2.1GHz, as well as its TD-LTE deployment in 2.5GHz.
Softbank still sees the iPhone as the center of its consumer strategy, especially now it has secured the LTE-based iPhone 5, and the new deal will give it additional capacity to support the data-hungry device. ‘You asked if the iPhone 5 had a role as a trigger for this business merger. The answer is yes,’ Softbank CEO Masayoshi Son told reporters.
eMobile switched on LTE services in March, and pledged that by midyear, it would reach 99% of the population of Japan’s major cities with 4G, though only 40% of the total POPs, a figure targeted to rise to 70% by next April. Outside LTE zones, users fall back to the DC-HSPA+ service, which covers 93% of the population and is branded eMobile G4.
Softbank had 34.9m mobile subscribers as of the end of August, while eMobile had just 4.2m. The latter started life as an MVNO, founded in 2008 by KDDI founder Sachio Semmoto, but went live on its own HSPA network in 2010. Together, the two cellcos will have 28.8% share, ahead of KDDI’s 26.5% or 35.9m customers. Docomo has 60.6m subscribers, for 44.7% of the total base.
Sprint mulls counterbid for MetroPCS?
Sprint could step in to spoil T-Mobile’s party by making a higher offer for regional US carrier MetroPCS, sources told Bloomberg on Thursday. This would be yet another misstep in Sprint’s M&A strategy ‘ its board reportedly rejected a proposal to bid for the flat rate carrier earlier this year, having also failed to foresee AT&T’s move on T-Mobile a year earlier. However, the third cellco is supposedly evaluating the pros and cons of creating a bidding war, given that its own growing prepaid business will be the primary target for an enlarged TMo.
Deutsche Telekom, TMo USA’s owner, announced the merger plan with MetroPCS earlier this week, looking to boost its spectrum position and its scale in order to become the leading brand in the ‘value’ end of the LTE market. This would differentiate it from the clear 4G leaders, Verizon and AT&T, rather than seeking to go after their headstart, but would put Sprint firmly in its sights.
Sprint’s CEO Dan Hesse has indicated he expects to take part in a consolidation process in the US, but with MetroPCS gone, the firm’s options are limited to smaller carriers like Leap or its own joint venture, Clearwire.
One factor that could be an obstacle to a Sprint bid would be the break-up fee if TMo backed away from the deal, it would have to pay its prey $250m, while if the smaller operator reneged, it would owe TMo $150m.
Deutsche Telekom is prepared for a potential counterbid from Sprint and would consider better terms if necessary, according to Bloomberg’s sources. Based on the current price of MetroPCS shares, which will account for 26% of the combined TMo entity, the Texas firm’s equity could be valued around $3bn, leading to an equity value of $11.4bn for the combined entity excluding liabilities, said Bloomberg. Accounting for joint net debt of $21.4bn, the enterprise value would total about $32.8bn.
Sprint abandoned plans to buy MetroPCS around February, after its board rejected the transaction, which may have cost as much as $8bn. Sprint shares have more than doubled in value since then while those of the regional player are flat. MetroPCS currently has a market value of $4.35bn while Sprint’s is about $15bn. Sprint was also said to be in talks with DT about a merger with T-Mobile USA, before AT&T rushed in with its own (failed) bid.
If Sprint and MetroPCS did merge, they would not have TMo’s problems with sustaining incompatible 3G networks, since both have CDMA, but their spectrum for LTE would be less harmonized. Both operate CDMA in 1.9GHz but many of MetroPCS’s 2G systems, and most of its LTE, runs in AWS (where TMo is also deploying 4G), while Sprint is the only major player with no holdings in this band and is initially rolling out LTE in its 800MHz iDEN band. As GigaOM points out, the only real match for MetroPCS’s spectrum is its roaming partner Leap, with which it has held repeated merger talks over the years.