TMUS: Wireless Ain’t What it Used to Be

sagawa
Print Friendly
Share on LinkedIn0Tweet about this on Twitter0Share on Facebook0

SEE LAST PAGE OF THIS REPORT Paul Sagawa / Artur Pylak

FOR IMPORTANT DISCLOSURES 203.901.1633 /.1634

psagawa@ / apylak@ssrllc.com

twitter.jpg @PaulSagawaSSR

July 9, 2015

TMUS: Wireless Ain’t What it Used to Be

We believe that TMUS is worth $55-60 based on projected future market share gains and margin improvements. It is winning a zero-sum game in the increasingly stagnant US wireless market, as consumer priorities shift from voice coverage to data service, flexibility and pricing. The market leading duopoly of VZ and AT&T cannot afford to follow TMUS’s bold “Uncarrier” initiatives and aggressive pricing for fear of killing their cash flows, while S remains mired in a never ending turnaround. We also see TMUS as well suited for potential future wireless growth markets, such as residential broadband, IoT, Automotive and others. Against this, top line expansion will yield leverage against TMUS’s network, spectrum and customer acquisition investments, including an anticipated $10B investment, at a FCC sanctioned discount, in the 600-700 MHz spectrum block to be auctioned in 2016. All of this could drive 50% upside to the current market valuation. A combination with DISH or S could yield operating synergies worth an additional 25% to the base valuation.

  • US mobile wireless is a mature business. Wireless carrier sales were down (1.7%) YoY in 1Q15, capping a 15-month slide from robust 7.4% growth in 4Q13. Market saturation is a major concern – total sub growth has decelerated to 4.3% in 1Q15, with total penetration up to 110% and smartphones representing 75.6% of the mix. Add to this ARPUs that appear to have peaked under pressure from TMUS’s aggressive pricing, a harbinger of future deflation across the industry. Despite this, analyst estimates for the 4 major US carriers project a return to sales growth, accelerating into next year. We are not so optimistic, and expect total US mobile wireless service revenues to trend down going forward.
  • The basis of competition is evolving. In the early days of smartphones, VZ and T were able to exploit their early 4G builds, and coverage advantage – plus T’s iPhone exclusivity – to drive sub gains and ARPU growth. 8 years into the smartphone era, TMUS’ 4G speeds and availability now exceed the big 2 in many major markets, and data performance, usage/upgrade policies, and price trump service coverage as a buying criteria for many consumers. The duopolists, with 75% collective market share to lose and 5.3% average cash flow yields to protect, are trapped in the classic incumbent’s dilemma, facing attack from below but hesitant to respond for fear of cannibalizing their own profits.
  • TMUS is winning the zero-sum game. TMUS is the only carrier posting sales gains, as it continues to churn subscribers from its 3 rivals with its customer friendly “uncarrier” moves – unbundling device sales, liberalizing usage terms, adding free bundled perks (zero rated music, international roaming), and cutting prices. Accompanied by an aggressive, hip advertising and promotional campaign, these moves have driven 260bp of share gain in the two years since the introduction of “Uncarrier 1.0” which set low, simple rate plans and eliminated bundled phone subsidies. As expected, VZ and T’s complacency and S’s bungled operations have allowed TMUS to build considerable momentum.
  • Top line growth feeding bottom line profits. The very high fixed cost, very low marginal cost nature of the wireless communications business rewards market share leadership and high prices. This dynamic sees TMUS operating with its COGS + depreciation at 2/3rds of sales, substantially more than T (59%) or VZ (52%). While TMUS anticipates significant further CAPEX to license additional low-band spectrum and expand its network to address its coverage and in-building deficits, its relatively underutilized capacity in more densely populated markets is considerable asset in its subscriber growth driven strategy. With time, we believe the resulting sales growth will narrow the COGS+D gap vs. the duopoly and yield significant margin and cash flow expansion.
  • Spectrum is a key variable. In the beginning, VZ and T’s wireless networks were endowed with lower frequency spectrum that enabled significantly wider coverage at significantly lower cost than either TMUS or S. This advantage widened with the duopolists’ acquisition of the biggest swaths of 700MHz spectrum auctioned in 2008. Another 80-100MHz of this unusually valuable spectrum will be auctioned in 2016. The FCC currently plans to restrict VZ and T from bidding for 30MHz of this block, a limit that TMUS has lobbied hard to expand and that the leaders propose be eliminated. We believe that the FCC will likely proceed with the current rules proposal and that TMUS will obtain at least 10MHz filling in important gaps in its coverage map and supporting its current service in its top markets. It is possible that partners (GOOG, DISH, etc.) could also acquire spectrum that would work to TMUS’ benefit.
  • There will be new opportunities. The internet of things, including connected automobiles, could be a new source of revenue for mobile operators, likely with a severe price sensitivity that will benefit lower margin operators like TMUS. We also believe that “5G” technology, based on a standard expected for 2018, will allow wireless to effectively compete with fixed residential wireless.
  • TMUS worth 50% upside to current valuation. Assuming 10.4% 5 yr annual sales growth and an expansion of operating margins from 6.4% to 17.7%, we believe TMUS shares are worth $55-60. This assumes that the company is able to license at least 10MHz in the 700MHz band auctions for roughly $10B, spurring an increase to 20% subscriber share by 2019 as coverage sensitive consumers begin to churn. We are not factoring significant revenues for IoT or residential broadband, but expect both to be important growth markets post 2020.
  • M&A or partnerships cold enhance value. TMUS is an often rumored object of acquisition interest. Of potential buyers, we see S as having the most synergy – assuming a post 2016 FCC would allow a deal – able to leverage spectrum assets, existing subscribers, and network operations to drive costs lower. DISH is also an obvious match, with 50MHz of unused spectrum nearly adjacent to TMUS’ primary holdings, a potentially synergistic video business, and similarly pugnacious approach to competition, although financing could be an obstacle. Either of these could add $10-20/share to our valuation. GOOG is already partnering with TMUS for its Fi MVNO and could be a source of capital for network expansion. We are less optimistic in a CMCSA acquisition, given its previous lackluster moves in wireless and a lack of obvious operating synergies.

John and Charlie Sittin’ in a Tree …

TMUS is up nearly 44% YTD on a combination of strong wireless subscriber growth and a drumbeat of rumors that the company is fielding offers from interested suitors. We believe that TMUS could be worth $55-60 going it alone, a roughly 50% further upside, based on accelerating market share gains and resulting scale economies. Furthermore, synergies from a possible hook-up with DISH – or even S, if a post 2016 FCC would be open to it – could add another $10-15 per share in value.

TMUS is the lone bright spot in an otherwise troubled industry. 8 years after the iPhone, US smartphone penetration is high, industry subscriber growth has slowed and ARPUs are falling. In this context, overall wireless carrier revenues have decelerated sharply, from 6.7% YoY growth in 4Q13 to (-1.7%) in the most recent quarter, evidence that the market is, indeed, saturating and that TMUS’ aggressive pricing and service initiatives are taking its toll. Still analysts are blasé, expecting both VZ and T to reaccelerate over the rest of the year and rating both stocks a net “Buy”.

We see it as a zero-sum game that TMUS is winning. The primary use of wireless devices has abruptly changed from voice calling to Internet-driven mobile apps, and with that data speeds, local availability, and network congestion have replaced unbroken coverage maps as the most important consideration for users. TMUS has a robust, uncrowded network in most major markets, and a customer acquisition strategy that is working. The company’s “Uncarrier” initiatives – including paying termination fees from other carriers, cheap and unlimited data plans, unbundling devices from service plans and facilitating frequent upgrades, free international roaming, and significantly lower prices – reduce switching costs and eliminate pain points for users. Meanwhile, VZ and T, already with dominant market share and stagnant revenues, cannot follow for fear of intensifying earnings and cash flow pressures. This is a classic Clay Christensen “Disruptive Innovation” case of an upstart attacking from the bottom.

Additional 600-700MHz band licenses to be auctioned by the FCC in 2016 under rules expected to be favorable to TMUS, could help in several ways. First, lower frequencies travel much further and penetrate into buildings much better than TMUS’s existing bands, extending coverage and delivering better reception. This could catalyze further share gain from would-be subscribers holding back due to those deficits. Second, additional spectrum would expand the capacity of the TMUS network, enabling higher speeds with less congestion. Finally, the new spectrum could obviate the need to add cell sites as TMUS traffic grows, with significant scale economies contributing to improving margins.

We believe that there is considerable upside to TMUS based on ongoing share gains, scale driven margin expansion, and longer term new opportunities, such as “Internet of Things”, automotive and residential broadband. We also believe that there are also substantial synergies available to potential acquirers. Of these, we see DISH as the obvious match, assuming that Charlie Ergen can raise sufficient cash to pay Deutsche Telekom for their share. With highly complementary spectrum assets, synergies with the satellite TV franchise, and an intriguing early play in OTT television, the combination could raise the value of TMUS by 37%. While we are pessimistic about the likelihood of a CMCSA bid, time and a new FCC could bring back talk of a S/TMUS merger that would have very substantial synergies. We are also intrigued with the possibility of GOOG and its Project Fi as a source of capital for the 2016 spectrum auction and subsequent network build out.

The Thrill is Gone

The last few decades have been good to Verizon and AT&T Wireless under the many names by which they have been known. Given initial spectrum assignments in the valuable 850MHz band by fiat and lottery back in 1983, the two early incumbents had a huge coverage and cost advantage over later competitors who bought inferior 1900MHZ licenses at auction more than a decade after. These advantages are the product of basic physics – lower frequency radio waves remain robust over much longer distances than higher frequencies and penetrate through walls and other obstacles far better, as well (Exhibit 1). Verizon and AT&T, patiently patching together national coverage in these low bands through acquisition over the years while adding spectrum in higher bands to buttress network capacity, have turned spectrum advantage into a powerful duopoly with 75% combined market share, growing subscriber rolls, rising ARPUs, and hefty cash flow. Over the past decade, this has translated into 9.6% average revenue growth and dividend yields exceeding 6.4% (Exhibit 2).

Exh 1: Relative Coverage Advantages of Lower Frequency Spectrum versus Higher Frequencies

However, things have changed. Subscriber growth has decelerated sharply. Mobile penetration amongst US adults now tops 91% with 75% of those carrying data friendly smartphones (Exhibit 3). ARPUs have begun to decline under pressure from aggressive pricing tactics from market disruptor T-Mobile (Exhibit 4). As a result, mobile industry revenue growth has just dipped negative for the first time in many years. Consensus revenue expectations for the duopolists optimistically expect a rebound to growth for the rest of 2015 and on to 2016, but it is not clear what will change in the industry dynamic to accommodate that scenario. We see the US mobile carrier business as a mature, zero-sum game with slowly declining sales to the end of the decade.

Exh 2: AT&T and Verizon Annual Revenue, 2004-14

Exh 3: US Mobile and Smartphone Users, 2010-2014

Exh 4: Industry Wireless ARPU, 1Q09 – 1Q15

Can You Hear Me Now?

For the first 25 years of commercial wireless service in the US, coverage was king. Service quality was measured in dropped calls and communicated in maps colored in where ever a carrier had coverage. This was the games of haves and have-nots, with Verizon and AT&T having the best coverage and charging the highest prices and Sprint and T-Mobile offering steep discounts to attract the customers that were willing to accept dropped calls on their morning commute, poor reception in parts of the house, and unpredictable coverage when out of town.

Then things started changing. In the early ‘00’s AT&T Wireless merged with Cingular, and struggled to manage a huge network upgrade needed to combine the two networks. Service quality suffered badly and customers began to leave. In desperation, AT&T cut a sweetheart deal for Apple. At its launch in 2007, the iPhone was exclusive to AT&T, which quickly began to regain net subscriber share. However, the share gains came at a cost – Apple demanded substantial device subsidies, complete control over the user experience and an unlimited data. AT&T’s iPhone deal shifted the parameters for all of the players in the US carrier market, driving subsidies higher, spurring extraordinary data traffic growth, and weakening carrier relationships with their customers (Exhibit 5).

Despite those concessions, the early smartphone era was good to the wireless carrier business, driving both new subscriber growth and data plan adoption. By 2010, AT&T’s iPhone exclusivity had lapsed and Verizon, with its iconic coverage maps, regained the upper hand. Both Verizon and AT&T killed unlimited data plans in 2012, following a strategy of locking new smartphone users into two year contracts with aggressive subsidies, then enjoying ARPU expansion as usage growth drove subscribers to higher priced data tiers. Family plans looked to tie multiple users together into contracts with a slight discount, but that were hellish to unwind without incurring fees. The combination of contract lock in and that longstanding coverage advantage kept churn low and drove cash flow. This worked … for a while.

Exh 5: North America Mobile Data Traffic, 2010-2020

Can I Download a File Now?

Over the past five years, data usage by wireless subscribers has increased at a torrid 106% annual pace, while voice calling has declined by -2.8% (Exhibit 6). For many users, Verizon’s famous “Can You Hear Me Now?” has become an irrelevant question. Rather than blanket voice coverage into every corner of the country, these consumers care about their daily access to high speed data service at the times and in the places that they congregate.

This change is blunting the market leading duopoly’s longstanding advantage. Smaller, and often less congested, carriers like T-Mobile and Sprint may offer more consistent bandwidth where data-centric users need it most. In particular, T-Mobile, after three years of elevated CAPEX spent expanding its 4G network, offers surprisingly fast and available service in many major markets (Exhibit 7). This has been its table stakes for its ambitious Uncarrier gambit.

Exh 6: Average Monthly Voice Minutes per Mobile User

Exh 7: Wireless Carrier CAPEX, 1Q09-1Q15

The Uncarrier

In March 2013, upstart T-Mobile launched a broadside against the market status quo. Called “Uncarrier 1.0”, the initiative unbundled the cost of the smartphone from the cost of wireless service and eliminated the annual contract. Monthly service costs were the same whether or not a subscriber purchased a device, which could be financed and paid off monthly separately. This hit at a consumer pain point – cancelling a subscription at Verizon or AT&T triggered a significant penalty and the completion of a two year contract did not trigger monthly service price reductions despite having completed the bundled device amortization (Exhibit 8).

Uncarrier 2.0 hit 6 month later, introducing the $10/month “Jump” program, which offered comprehensive device insurance and allowed subscribers to return their devices after 6 months for an upgrade. This hit another major pain point – users resented being locked into a device by the leading carrier’s 2 year contract plans. The Jump program was just liberalized, allowing up to 3 upgrades per year and eliminating the $10/month fee (N.B. the previously bundled insurance must now be bought separately for $8/mo, comparable to similar programs at other carriers).

Exh 8: T-Mobile “Uncarrier” Initiative Summary

Uncarrier 3.0, in October 2013, brought free data roaming in more than 100 countries and reduced cost calling from the US to international markets, along with a free 200MB/month data allowance for connected tablets. Uncarrier 4.0 in January 2014 saw T-Mobile offering to pay off all early termination fees (ETFs) imposed by rival carriers on subscribers switching to its network. Uncarrier 5.0 in June 1014 provided potential customers with a one-week iPhone test drive before having to commit to a switch. Uncarrier 6.0, at the same time, granted free data capacity for music streaming services. Uncarrier 7.0, announced last September, added WiFi voice and text on all T-Mobile devices, enabling subscribers to send and receive calls and texts over WiFi when no T-Mobile signal is available, especially valuable in international markets. In December, Uncarrier 8.0 began to allow all T-Mobile users with capped data agreements to automatically roll unused data minutes forward for as much as a year. Uncarrier 9.0, launched this past March, greatly simplified and lowered service pricing for business accounts. And finally, just at the end of June, the company expanded its “JUMP!” offering to allow for up to 3 upgrades per year.

Kicking #$% and Taking Share

Collectively, these initiatives greatly reduce switching barriers for consumers (e.g. equipment bundling, ETFs), remove significant non-contract costs (e.g. early upgrade costs, roaming charges, data overage fees), and enhance the customer experience (e.g. WiFi calling, free music, equipment test drives, etc.). Combined with its investment to build out 4G capacity in major markets, consistently lower prices and an aggressive, digitally savvy marketing approach led by salty tongued CEO John Legere, Uncarrier has driven T-Mobile to steadily gain share of both subscribers and revenue in every quarter since it was introduced. In two years, the subscriber count is up more than 40%, as are revenues, adding 360bp to the company’s share (Exhibit 9-10).

Exh 9: Wireless Carrier Market Share – Subscribers, 1Q09-1Q15

Exh 10: Wireless Carrier Market Share – Revenue, 1Q09-1Q15

Market leaders Verizon and AT&T cannot afford to follow. The fat cash flows borne of years of duopoly pricing and controlled network investment fund exceptional dividend yields – Verizon is returning 4.6% and AT&T an even more impressive 5.2%. Both Verizon and AT&T operate with little spare capacity, with congestion evident in many major markets. Matching T-Mobile on price or eliminating fees would reduce ARPUs without adding much to the subscriber rolls. Matching T-Mobile on eliminating switching costs would simply make it easier for their own subscribers to switch. Matching T-Mobile on enhancing the customer experience would require investment. All of this would hurt near term cash flow and, potentially, endanger dividends. Verizon and AT&T are frozen in a classic game theory position – better to suffer the slow bleeding than to self-inflict a more serious wound (Exhibit 11).

Exh 11: Game Theory – Options and Likely Outcomes for Wireless Players

So Verizon and AT&T fight a war of words, launching programs that seem at first glance to offer some of the innovations offered by T-Mobile, most notably early upgrades and device financing, but leaving in place high service prices, data limits, hidden fees and contract lock ins, all while continuing to tout their superior coverage. But as subscriber priorities continue to shift toward data service price and performance, and as new spectrum and ongoing investment fill T-Mobile’s most pressing coverage gaps, the trickle of churn from the market leaders will likely grow to a more troubling stream. By then, it will be much more difficult and painful to counter.

Meanwhile, Sprint is trying to grab some of T-Mobile’s thunder, offering its “All-In” program, bundling a new iPhone or Samsung G6 with an unlimited usage plan for $80/month plus taxes and fees. This deal, which slightly undercuts T-Mobile for a similar combination, prompted an expletive laced Twitter exchange between Legere and Sprint CEO Marcelo Claure. Sprint, with a wide swath of largely unused spectrum in a very high frequency band and the financial backing of its parent Softbank, is an intriguing potential competitor, but faces a high degree of difficulty with a poor track record. Combining that high frequency spectrum (acquired with Clearwire) with Sprint’s patchwork quilt of various other spectrum bands requires brand new technical functions only recently added to the 4G standard and not yet fully supported by all equipment and devices. Rolling out service has been balky, slow and expensive. Moreover, years of clumsiness, particularly during its near disastrous integration with Nextel, have damaged the Sprint brand amongst consumers, who have deserted the carrier in droves.

We believe that T-Mobile will continue to make hay while Verizon and AT&T play defense and Sprint struggles to regain its footing. We are modeling T-Mobile wireless service revenue to grow at a better than 9% CAGR through 2019, gaining some 600bp of market share as the other three carriers experience slightly shrinking sales.

Top Line to Bottom Line

While network investment does follow network demand, there are still scale economies in building for coverage. With total subscriber share still just less than 30% of Verizon’s and 40% of AT&T’s, T-Mobile’s network is considerably less congested in many major markets. This gives it room for improving network utilization with growth, and thus, lower total network costs and depreciation relative to revenues. Moreover, SG&A – primarily brand marketing and management costs – also can be expected to show scale economies.

In 2014, T-Mobile’s operating margins were just 4.7%, while AT&T Wireless was generating 25.6% and Verizon Wireless was generating 30.6%. We believe that T-Mobile will show significant leverage to scale as it grows, bringing profitability sharply higher despite its aggressive pricing strategy. Consensus expectations peg 2015 EPS at $0.72, up from $0.30 in 2014, on net earnings up 80bp YoY as a percentage of sales. 2016 consensus projects a further bump to $1.83, a 200bp YoY rise. While this may seem optimistic, it implies 2016 operating margins of less than 8%, still well less than a third of those currently generated by AT&T and Verizon’s wireless businesses. To achieve it, TMUS would have to grow its costs – COGS, SG&A and Depreciation – a third more slowly than it is expected to grow its sales, that is a roughly 5% annual growth in costs vs. a 7.5% sales CAGR. We believe that this is a very reasonable expectation.

Longer term, we believe T-Mobile can achieve operating margins in the mid to upper teens, even as its aggressive strategy brings overall industry profitability down.

Give Me Spectrum, or Give Me Death

T-Mobile USA’s roots stretch back to the 1994 launch of VoiceStream Wireless, a substantial winner in the PCS spectrum auctions of that year. Over the next decade, VoiceStream merged with fellow regional PCS operators Omnitel and Aerial, before being acquired by Deutsche Telekom and combined with a 4th regional PCS operator, Powertel, to create the basis of a national carrier, dubbed T-Mobile USA, after Deutsche Telekom’s European wireless brand. After the failed 2011 attempt to divest T-Mobile USA to AT&T, Deutsche Telekom agreed to merge it with budget carrier MetroPCS and list the merged company as a public company. Thus T-Mobile was born … with a handicap (Exhibit 12).

Exh 12: Historical Timeline of T-Mobile

Verizon and AT&T were also Franken-carriers, created from the mash up of multiple regional carriers into a national presence (Exhibit 13). The difference was the Verizon and AT&T networks were built with the two original 850 MHz spectrum licenses granted to the Bell system and to a variety of politically favored applicants back in 1982. This superior spectrum has dramatically longer range and better penetration through solid objects than the 1900 MHz licenses that make up the core of T-Mobile and Sprint’s networks, giving Verizon and AT&T the cost and performance advantages that, until very recently, completely defined the nature of wireless competition in this country (Exhibit 14).

Exh 13: Legacy Companies Comprising Current Mobile Players

Exh 14: Carrier Spectrum Holdings Summary

This spectrum advantage was further strengthened in the 700 MHz auctions of 2008, where Verizon and AT&T dominated bidding. Today, after last year’s FCC auction of 40 MHz of AWS-3 spectrum, Verizon has an average of 116 MHz of spectrum across its network, while AT&T has 157. Importantly, 46 to 55 MHz of Verizon’s and 51 to 56 MHz of AT&T’s spectrum is in the advantaged sub 1 GHz frequency range, while T-Mobile has 0-10 MHz in its covered markets.

Exh 15: US Mobile Wireless Frequencies and Auctions

Against this backdrop, the FCC is planning to auction 80-100 MHz of super prime licenses in the 600-700 MHz range early next year (Exhibit 15). T-Mobile has lobbied hard to have some portion of that spectrum reserved for bidders other than the advantaged duopoly. Currently, the proposed FCC rules have reserved 30 MHz, but T-Mobile and S continue to lobby for more, or for total low-band spectrum limitations that would effectively accomplish the same thing. After heavy spending in the AWS-3 auction and with the largest low band holdings of any carrier, Verizon has suggested that it may not need to be aggressive in next year’s auctions. If so, it could also leave more room for T-Mobile to snap up licenses at reasonable fees.

We believe that the auction is likely to generate value for T-Mobile and its shareholders, allowing the company to compete more directly for customers with high priority on coverage and to expand capacity more quickly and cheaply. We have modeled a $10B expenditure for 10MHz of licenses with national coverage, roughly $3 per MHz POP. This is slightly more than AT&T and Verizon paid for AWS-3 licenses at the end of last year, reflecting both the significantly higher value of the 600-700 MHz spectrum block, but also the benefit of having a portion of frequencies reserved for smaller carriers.

Houses, Cars and Machines

While we believe that mobile wireless service is becoming a mature market in the US, there are future opportunities that could catalyze a new era of growth. At one end, the growing speeds and system capacities for wireless carrier networks will begin to encroach on the traditional performance standards of fixed residential broadband. By decade end, LTE-C will have rolled out, with the possibility of 10 Gbps peak downstream speeds for carriers with ample spectrum to support it (Exhibit 16). Just as younger cohorts led a shift to mobile phone only voice calling a decade ago, we expect the same phenomenon to begin hitting wireline broadband carriers within the next five years. This is a sizeable future opportunity.

Exh 16: Evolution of Broadband Standards, 2000 – 2020

At the other end, are the many millions of potentially connected machines requiring cheap, secure, low latency networks to make the so-called “Internet of Things” economically viable. Within a building, these devices may use private, local networks, like ZigBee or BluetoothLE to make the connections, but when these machines are not tethered to a local network, low power, narrow band modes efficiently supported on carrier networks will be key. Latency will be particularly important for some applications, such as automotive, telemedicine or factory control systems.

5G wireless standards aim for maximum speeds of 10GBps or more, the ability to operate across licensed and unlicensed spectrum bands at frequencies much higher than those in use today, the flexibility to cost effectively connect to many millions of low-power, narrow band IoT machines, and the ultra-low latency needed to extend the reach of virtual reality and to support mission critical vertical applications. Initial research proposals for 5G began in 2012. Samsung is pledging to demonstrate 5G with live networks at the 2018 Winter Olympics in South Korea, a rough time line for working prototypes shared by Ericsson and Nokia. On that schedule, meaningful commercial deployments could begin as soon as 2020. While these futures could be delayed, and T-Mobile could execute poorly in reaching for the opportunities, we do see meaningful option value for all US wireless carriers half a decade ahead.

Valuation

We tend to be suspicious of DCF models, as these models are so sensitive to the assumptions put into them that a savvy analyst can manipulate the inputs to shift valuation sharply in either direction. Still, it is a useful exercise to understand the assumptions that may be embedded in the market price and as a sanity check on the possible upside should a particular scenario play out.

Exh 17: T-Mobile Revenue Assumptions, 2014-19

Our 5 year scenario assumes that T-Mobile can grow its total branded subscribers at a 12.5% CAGR, growing to just over 74 million in 2019 (Exhibit 17). We project ARPUs to decline at about 3% per year to less than $31 at the end of the period. This implies overall average revenue growth of 9.7%, taking T-Mobile’s share of US carrier sales from 13.7% in 2015 to 18.5% in 2019. To put this in perspective, T-Mobile gained 150bp of revenue share in 2014 alone.

On the cost side, we project T-Mobile’s operating margins, including depreciation, to grow from 6.4% in 2015 to about 17.5% in 2019, a steep improvement, but still well short of the 25.6% generated by AT&T Wireless or the 30.6% earned by Verizon Wireless in 2014 (Exhibit 18). A big piece of this is stemming the losses sustained in device sales, which are projected to lose roughly $2.5B in 2015. We are projecting that this can be cut in half over five years, as T-Mobile’s equipment financing approach should allow it to scale back subsidies over time, particularly given inherent financial incentives to defer device upgrades. We do assume that T-Mobile spends $10B at auction buying 600-700 MHz spectrum licenses in 2016 and then maintains somewhat elevated CAPEX as it builds out its network, a move the contributes to ongoing reductions in network operations costs relative to sales but that has substantial impact on near term cash flows.

Exh 18: T-Mobile Cost Assumptions, 2014-19

In this scenario, our DCF model projects a $58/share fair value price target for T-Mobile, a more than 50% premium from the current price, assuming a 2.5% terminal growth rate and a 1.5 Beta. Shifting the assumptions on the terminal growth and the Beta suggests a range of values between $45 and $80. In contrast, the current share price closely reflects the assumptions embedded in consensus estimates, which project an abrupt deceleration in operating cash flow growth post 2016, but do not suggest a bump in CAPEX for the 2016 auctions. We note that just 7 analysts include cash flow projections out to 2019, so the aggregated numbers may not be fully representative of the perspectives of all 24 analysts that cover the company.

Exh 19: T-Mobile DCF Valuation

It Can Be Yours IF the Price is Right

The scuttlebutt is that Dish Network CEO Charlie Ergen wants to merge his company with T-Mobile, assuming he can finance it. That is, if 66% owner Deutsche Telekom doesn’t convince its favored buyer, Comcast to make a bid. Even Softbank, which backed off on its ambitions to merge T-Mobile with its own Sprint subsidiary, could return to make a play if presented with a more favorable FCC and Justice Department under a Republican administration.

Of these, we believe a Dish combination is the most likely. Ergen is sitting on roughly 80 MHz of unused spectrum in the AWS-3 (25 MHz), AWS-4 (40 MHz), H Block (10 MHz) and 700 MHz (5 MHz) bands, with full national coverage. The AWS-4 and H Block spectrum is roughly adjacent to blocks already in use in T-Mobile’s LTE network. Combining with Dish, particularly if the resulting company is still granted favorable status in the upcoming 600-700 MHz auctions, would give T-Mobile means to close its remaining regional coverage gaps, to resolve busy cell congestion without expensive cell splitting, to offer faster network speeds and smooth future technology transitions. This could accelerate share gains and improve the trajectory of gross margins. It would also reduce CAPEX, including the amount of spending on 600-700 MHz spectrum. Moreover, marketing and overhead functions could be shared with Dish’s satellite and online streaming TV businesses. Importantly, Dish is also a good cultural fit, with T-Mobile CEO John Legere likely to rise to CEO of the merged business and Charlie Ergen focusing on his role as Chairman. We believe these synergies could be worth $10-$20/share (Exhibit 20). At issue is whether Dish would be able to raise sufficient cash to satisfy Deutsche Telekom, which is rumored to be unwilling to accept stock for its share.

Exh 20: TMUS – DISH Synergies

A Comcast combination is less attractive, particularly given that company’s lackluster previous history in bundling wireless services with its cable and broadband. Very modest savings in back office and management functions could amount to something less than $5/share. Reports suggest that Deutsche Telekom prefers a Comcast deal and would be more willing to accept equity for that combination.

We also note that a Republican Presidential victory in 2016 could open a door to a revival of merger talks with Sprint. This combination would have even greater synergies than a Dish deal, but it is likely not possible to close before 2018, even if the political landscape shifts enough to allow it (Exhibit 21). There has also been interest from foreign carriers, such as the French Iliad, which made a $33 a share bid in 2014, or American Movil, rumored to have considered a bid earlier this year. We do not see real synergy in a global footprint, particularly since Deutsche Telekom has not been able to find it.

Exh 21: TMUS – Sprint Synergies

We are also intrigued with the potential for partnerships. Network sharing with Sprint in the 600-700 MHz spectrum could reduce CAPEX and network operating costs, spreading them between the two carriers. It is also possible that a 3rd party could step into the auction and/or fund the build out of a shared wholesale network. Google, already working in partnership with T-Mobile and Sprint for its Project Fi MVNO network, is an obvious candidate. Partnerships like these could deliver $5-$10/share in synergies, and could be additive to a combination with Dish.

Risks

Of course, our scenario of continued subscriber growth, increasingly at the expense of market leaders AT&T and Verizon, and improving scale economies may not play out. One risk is that the appeal of T-Mobile’s data service and price forward “Uncarrier” campaign wanes, either because the demographics to which it appeals are narrower than we believe or because the market leaders capitulate and follow T-Mobile’s lead. We believe that the former reason is more plausible than the latter, but remain confident that the demand for “Uncarrier” will spread to new segments as T-Mobile fills its coverage gaps and as consumers grow more confident in its network quality.

It is also possible that T-Mobile will not be able to achieve the economies of scale that we have assumed in its network operations and SG&A. This is obviously linked to relative growth achieved on the top line, but also to management cost discipline, the impact of falling ARPUs on margins, and the cost of Uncarrier initiatives, such as buying out early termination fees from competitors. It is also possible that T-Mobile will not be able to temper the losses that it is currently generating on equipment sales, despite its strong move to financing rather than subsidizing devices. A big piece of this will be establishing an effective program for reselling the recent model devices it takes in as part of its early upgrade “Jump” initiatives. This is not assured.

Spectrum is also a risk. FCC bidding rules currently intended to favor T-Mobile and Sprint could be challenged and changed. In this scenario, the price of spectrum could be much higher and T-Mobile could be shut out by the deeper pocketed Verizon and AT&T. It is also possible that a new party could spoil the bidding – perhaps Dish Network or a cable operator – driving up spectrum prices or again, beating out T-Mobile. Our confidence around our spectrum assumptions will grow as the auction approaches.

Exh 10: Wireless Carrier Market Share – Revenue, 1Q09-1Q15

Market leaders Verizon and AT&T cannot afford to follow. The fat cash flows borne of years of duopoly pricing and controlled network investment fund exceptional dividend yields – Verizon is returning 4.6% and AT&T an even more impressive 5.2%. Both Verizon and AT&T operate with little spare capacity, with congestion evident in many major markets. Matching T-Mobile on price or eliminating fees would reduce ARPUs without adding much to the subscriber rolls. Matching T-Mobile on eliminating switching costs would simply make it easier for their own subscribers to switch. Matching T-Mobile on enhancing the customer experience would require investment. All of this would hurt near term cash flow and, potentially, endanger dividends. Verizon and AT&T are frozen in a classic game theory position – better to suffer the slow bleeding than to self-inflict a more serious wound (Exhibit 11).

Exh 11: Game Theory – Options and Likely Outcomes for Wireless Players

So Verizon and AT&T fight a war of words, launching programs that seem at first glance to offer some of the innovations offered by T-Mobile, most notably early upgrades and device financing, but leaving in place high service prices, data limits, hidden fees and contract lock ins, all while continuing to tout their superior coverage. But as subscriber priorities continue to shift toward data service price and performance, and as new spectrum and ongoing investment fill T-Mobile’s most pressing coverage gaps, the trickle of churn from the market leaders will likely grow to a more troubling stream. By then, it will be much more difficult and painful to counter.

Meanwhile, Sprint is trying to grab some of T-Mobile’s thunder, offering its “All-In” program, bundling a new iPhone or Samsung G6 with an unlimited usage plan for $80/month plus taxes and fees. This deal, which slightly undercuts T-Mobile for a similar combination, prompted an expletive laced Twitter exchange between Legere and Sprint CEO Marcelo Claure. Sprint, with a wide swath of largely unused spectrum in a very high frequency band and the financial backing of its parent Softbank, is an intriguing potential competitor, but faces a high degree of difficulty with a poor track record. Combining that high frequency spectrum (acquired with Clearwire) with Sprint’s patchwork quilt of various other spectrum bands requires brand new technical functions only recently added to the 4G standard and not yet fully supported by all equipment and devices. Rolling out service has been balky, slow and expensive. Moreover, years of clumsiness, particularly during its near disastrous integration with Nextel, have damaged the Sprint brand amongst consumers, who have deserted the carrier in droves.

We believe that T-Mobile will continue to make hay while Verizon and AT&T play defense and Sprint struggles to regain its footing. We are modeling T-Mobile wireless service revenue to grow at a better than 9% CAGR through 2019, gaining some 600bp of market share as the other three carriers experience slightly shrinking sales.

Top Line to Bottom Line

While network investment does follow network demand, there are still scale economies in building for coverage. With total subscriber share still just less than 30% of Verizon’s and 40% of AT&T’s, T-Mobile’s network is considerably less congested in many major markets. This gives it room for improving network utilization with growth, and thus, lower total network costs and depreciation relative to revenues. Moreover, SG&A – primarily brand marketing and management costs – also can be expected to show scale economies.

In 2014, T-Mobile’s operating margins were just 4.7%, while AT&T Wireless was generating 25.6% and Verizon Wireless was generating 30.6%. We believe that T-Mobile will show significant leverage to scale as it grows, bringing profitability sharply higher despite its aggressive pricing strategy. Consensus expectations peg 2015 EPS at $0.72, up from $0.30 in 2014, on net earnings up 80bp YoY as a percentage of sales. 2016 consensus projects a further bump to $1.83, a 200bp YoY rise. While this may seem optimistic, it implies 2016 operating margins of less than 8%, still well less than a third of those currently generated by AT&T and Verizon’s wireless businesses. To achieve it, TMUS would have to grow its costs – COGS, SG&A and Depreciation – a third more slowly than it is expected to grow its sales, that is a roughly 5% annual growth in costs vs. a 7.5% sales CAGR. We believe that this is a very reasonable expectation.

Longer term, we believe T-Mobile can achieve operating margins in the mid to upper teens, even as its aggressive strategy brings overall industry profitability down.

Give Me Spectrum, or Give Me Death

T-Mobile USA’s roots stretch back to the 1994 launch of VoiceStream Wireless, a substantial winner in the PCS spectrum auctions of that year. Over the next decade, VoiceStream merged with fellow regional PCS operators Omnitel and Aerial, before being acquired by Deutsche Telekom and combined with a 4th regional PCS operator, Powertel, to create the basis of a national carrier, dubbed T-Mobile USA, after Deutsche Telekom’s European wireless brand. After the failed 2011 attempt to divest T-Mobile USA to AT&T, Deutsche Telekom agreed to merge it with budget carrier MetroPCS and list the merged company as a public company. Thus T-Mobile was born … with a handicap (Exhibit 12).

Exh 12: Historical Timeline of T-Mobile

Verizon and AT&T were also Franken-carriers, created from the mash up of multiple regional carriers into a national presence (Exhibit 13). The difference was the Verizon and AT&T networks were built with the two original 850 MHz spectrum licenses granted to the Bell system and to a variety of politically favored applicants back in 1982. This superior spectrum has dramatically longer range and better penetration through solid objects than the 1900 MHz licenses that make up the core of T-Mobile and Sprint’s networks, giving Verizon and AT&T the cost and performance advantages that, until very recently, completely defined the nature of wireless competition in this country (Exhibit 14).

Exh 13: Legacy Companies Comprising Current Mobile Players

Exh 14: Carrier Spectrum Holdings Summary

This spectrum advantage was further strengthened in the 700 MHz auctions of 2008, where Verizon and AT&T dominated bidding. Today, after last year’s FCC auction of 40 MHz of AWS-3 spectrum, Verizon has an average of 116 MHz of spectrum across its network, while AT&T has 157. Importantly, 46 to 55 MHz of Verizon’s and 51 to 56 MHz of AT&T’s spectrum is in the advantaged sub 1 GHz frequency range, while T-Mobile has 0-10 MHz in its covered markets.

Exh 15: US Mobile Wireless Frequencies and Auctions

Against this backdrop, the FCC is planning to auction 80-100 MHz of super prime licenses in the 600-700 MHz range early next year (Exhibit 15). T-Mobile has lobbied hard to have some portion of that spectrum reserved for bidders other than the advantaged duopoly. Currently, the proposed FCC rules have reserved 30 MHz, but T-Mobile and S continue to lobby for more, or for total low-band spectrum limitations that would effectively accomplish the same thing. After heavy spending in the AWS-3 auction and with the largest low band holdings of any carrier, Verizon has suggested that it may not need to be aggressive in next year’s auctions. If so, it could also leave more room for T-Mobile to snap up licenses at reasonable fees.

We believe that the auction is likely to generate value for T-Mobile and its shareholders, allowing the company to compete more directly for customers with high priority on coverage and to expand capacity more quickly and cheaply. We have modeled a $10B expenditure for 10MHz of licenses with national coverage, roughly $3 per MHz POP. This is slightly more than AT&T and Verizon paid for AWS-3 licenses at the end of last year, reflecting both the significantly higher value of the 600-700 MHz spectrum block, but also the benefit of having a portion of frequencies reserved for smaller carriers.

Houses, Cars and Machines

While we believe that mobile wireless service is becoming a mature market in the US, there are future opportunities that could catalyze a new era of growth. At one end, the growing speeds and system capacities for wireless carrier networks will begin to encroach on the traditional performance standards of fixed residential broadband. By decade end, LTE-C will have rolled out, with the possibility of 10 Gbps peak downstream speeds for carriers with ample spectrum to support it (Exhibit 16). Just as younger cohorts led a shift to mobile phone only voice calling a decade ago, we expect the same phenomenon to begin hitting wireline broadband carriers within the next five years. This is a sizeable future opportunity.

Exh 16: Evolution of Broadband Standards, 2000 – 2020

At the other end, are the many millions of potentially connected machines requiring cheap, secure, low latency networks to make the so-called “Internet of Things” economically viable. Within a building, these devices may use private, local networks, like ZigBee or BluetoothLE to make the connections, but when these machines are not tethered to a local network, low power, narrow band modes efficiently supported on carrier networks will be key. Latency will be particularly important for some applications, such as automotive, telemedicine or factory control systems.

5G wireless standards aim for maximum speeds of 10GBps or more, the ability to operate across licensed and unlicensed spectrum bands at frequencies much higher than those in use today, the flexibility to cost effectively connect to many millions of low-power, narrow band IoT machines, and the ultra-low latency needed to extend the reach of virtual reality and to support mission critical vertical applications. Initial research proposals for 5G began in 2012. Samsung is pledging to demonstrate 5G with live networks at the 2018 Winter Olympics in South Korea, a rough time line for working prototypes shared by Ericsson and Nokia. On that schedule, meaningful commercial deployments could begin as soon as 2020. While these futures could be delayed, and T-Mobile could execute poorly in reaching for the opportunities, we do see meaningful option value for all US wireless carriers half a decade ahead.

Valuation

We tend to be suspicious of DCF models, as these models are so sensitive to the assumptions put into them that a savvy analyst can manipulate the inputs to shift valuation sharply in either direction. Still, it is a useful exercise to understand the assumptions that may be embedded in the market price and as a sanity check on the possible upside should a particular scenario play out.

Exh 17: T-Mobile Revenue Assumptions, 2014-19

Our 5 year scenario assumes that T-Mobile can grow its total branded subscribers at a 12.5% CAGR, growing to just over 74 million in 2019 (Exhibit 17). We project ARPUs to decline at about 3% per year to less than $31 at the end of the period. This implies overall average revenue growth of 9.7%, taking T-Mobile’s share of US carrier sales from 13.7% in 2015 to 18.5% in 2019. To put this in perspective, T-Mobile gained 150bp of revenue share in 2014 alone.

On the cost side, we project T-Mobile’s operating margins, including depreciation, to grow from 6.4% in 2015 to about 17.5% in 2019, a steep improvement, but still well short of the 25.6% generated by AT&T Wireless or the 30.6% earned by Verizon Wireless in 2014 (Exhibit 18). A big piece of this is stemming the losses sustained in device sales, which are projected to lose roughly $2.5B in 2015. We are projecting that this can be cut in half over five years, as T-Mobile’s equipment financing approach should allow it to scale back subsidies over time, particularly given inherent financial incentives to defer device upgrades. We do assume that T-Mobile spends $10B at auction buying 600-700 MHz spectrum licenses in 2016 and then maintains somewhat elevated CAPEX as it builds out its network, a move the contributes to ongoing reductions in network operations costs relative to sales but that has substantial impact on near term cash flows.

Exh 18: T-Mobile Cost Assumptions, 2014-19

In this scenario, our DCF model projects a $58/share fair value price target for T-Mobile, a more than 50% premium from the current price, assuming a 2.5% terminal growth rate and a 1.5 Beta. Shifting the assumptions on the terminal growth and the Beta suggests a range of values between $45 and $80. In contrast, the current share price closely reflects the assumptions embedded in consensus estimates, which project an abrupt deceleration in operating cash flow growth post 2016, but do not suggest a bump in CAPEX for the 2016 auctions. We note that just 7 analysts include cash flow projections out to 2019, so the aggregated numbers may not be fully representative of the perspectives of all 24 analysts that cover the company.

Exh 19: T-Mobile DCF Valuation

It Can Be Yours IF the Price is Right

The scuttlebutt is that Dish Network CEO Charlie Ergen wants to merge his company with T-Mobile, assuming he can finance it. That is, if 66% owner Deutsche Telekom doesn’t convince its favored buyer, Comcast to make a bid. Even Softbank, which backed off on its ambitions to merge T-Mobile with its own Sprint subsidiary, could return to make a play if presented with a more favorable FCC and Justice Department under a Republican administration.

Of these, we believe a Dish combination is the most likely. Ergen is sitting on roughly 80 MHz of unused spectrum in the AWS-3 (25 MHz), AWS-4 (40 MHz), H Block (10 MHz) and 700 MHz (5 MHz) bands, with full national coverage. The AWS-4 and H Block spectrum is roughly adjacent to blocks already in use in T-Mobile’s LTE network. Combining with Dish, particularly if the resulting company is still granted favorable status in the upcoming 600-700 MHz auctions, would give T-Mobile means to close its remaining regional coverage gaps, to resolve busy cell congestion without expensive cell splitting, to offer faster network speeds and smooth future technology transitions. This could accelerate share gains and improve the trajectory of gross margins. It would also reduce CAPEX, including the amount of spending on 600-700 MHz spectrum. Moreover, marketing and overhead functions could be shared with Dish’s satellite and online streaming TV businesses. Importantly, Dish is also a good cultural fit, with T-Mobile CEO John Legere likely to rise to CEO of the merged business and Charlie Ergen focusing on his role as Chairman. We believe these synergies could be worth $10-$20/share (Exhibit 20). At issue is whether Dish would be able to raise sufficient cash to satisfy Deutsche Telekom, which is rumored to be unwilling to accept stock for its share.

Exh 20: TMUS – DISH Synergies

A Comcast combination is less attractive, particularly given that company’s lackluster previous history in bundling wireless services with its cable and broadband. Very modest savings in back office and management functions could amount to something less than $5/share. Reports suggest that Deutsche Telekom prefers a Comcast deal and would be more willing to accept equity for that combination.

We also note that a Republican Presidential victory in 2016 could open a door to a revival of merger talks with Sprint. This combination would have even greater synergies than a Dish deal, but it is likely not possible to close before 2018, even if the political landscape shifts enough to allow it (Exhibit 21). There has also been interest from foreign carriers, such as the French Iliad, which made a $33 a share bid in 2014, or American Movil, rumored to have considered a bid earlier this year. We do not see real synergy in a global footprint, particularly since Deutsche Telekom has not been able to find it.

Exh 21: TMUS – Sprint Synergies

We are also intrigued with the potential for partnerships. Network sharing with Sprint in the 600-700 MHz spectrum could reduce CAPEX and network operating costs, spreading them between the two carriers. It is also possible that a 3rd party could step into the auction and/or fund the build out of a shared wholesale network. Google, already working in partnership with T-Mobile and Sprint for its Project Fi MVNO network, is an obvious candidate. Partnerships like these could deliver $5-$10/share in synergies, and could be additive to a combination with Dish.

Risks

Of course, our scenario of continued subscriber growth, increasingly at the expense of market leaders AT&T and Verizon, and improving scale economies may not play out. One risk is that the appeal of T-Mobile’s data service and price forward “Uncarrier” campaign wanes, either because the demographics to which it appeals are narrower than we believe or because the market leaders capitulate and follow T-Mobile’s lead. We believe that the former reason is more plausible than the latter, but remain confident that the demand for “Uncarrier” will spread to new segments as T-Mobile fills its coverage gaps and as consumers grow more confident in its network quality.

It is also possible that T-Mobile will not be able to achieve the economies of scale that we have assumed in its network operations and SG&A. This is obviously linked to relative growth achieved on the top line, but also to management cost discipline, the impact of falling ARPUs on margins, and the cost of Uncarrier initiatives, such as buying out early termination fees from competitors. It is also possible that T-Mobile will not be able to temper the losses that it is currently generating on equipment sales, despite its strong move to financing rather than subsidizing devices. A big piece of this will be establishing an effective program for reselling the recent model devices it takes in as part of its early upgrade “Jump” initiatives. This is not assured.

Spectrum is also a risk. FCC bidding rules currently intended to favor T-Mobile and Sprint could be challenged and changed. In this scenario, the price of spectrum could be much higher and T-Mobile could be shut out by the deeper pocketed Verizon and AT&T. It is also possible that a new party could spoil the bidding – perhaps Dish Network or a cable operator – driving up spectrum prices or again, beating out T-Mobile. Our confidence around our spectrum assumptions will grow as the auction approaches.

Print Friendly