BT's Fundamental Value Is Overlooked

BT’s scale as the U.K.’s largest fixed-line, broadband, and wireless telecom operator provides the company with a narrow economic moat due to cost advantages. Additionally, BT benefits from efficient scale, as the costs for a new entrant to wireless telecom would be such that the operator would have an extremely difficult time earning its cost of capital.

We have long been proponents of converged services and have seen the U.K. as a laggard in this regard. Now, however, the country is beginning to focus on convergence, and BT is best suited to prosper from this trend as the only U.K. operator with both fixed-line and wireless networks. We have been disappointed with the slow rate at which the company has integrated EE, but we believe this was partly due to ongoing reviews by Ofcom, the U.K. telecom regulator.

In May, BT announced a much better integrated converged offering. We believe this will enhance the distinction between BT and other U.K. operators, as it will fully control the customer’s experience. BT controls the quality of the network, its maintenance and upgrade schedule, customer billing, and opportunities to cross-sell other services. The experience of multiple operators in various countries shows that the more services a customer subscribes to, the lower the churn and the more valuable the customer becomes. This enhances returns on capital and the company’s moat. But BT’s announcement has a twist: Rather than bundling multiple services for a discounted price, BT plans to maintain high prices but entice customers to sign up for the bundled products by offering additional services and increased data allowance. This creates a “more for more” service rather than a discounted product. Given the lack of competitors in the U.K. converged marketplace, we think this strategy has a decent chance of success. Ultimately, we believe the market is missing out on BT’s economic moat, which will be strengthened by convergence.

U.K. Communications Market Structure Favors BT Four wireless operators in the U.K own their own networks: BT (which owns EE), Telefonica TEF (which owns O2), Vodafone VOD, and 3 UK (owned by CK Hutchison). There are also many operators that don't own their own networks, but operate as mobile virtual network operators, or MVNOs, with the largest being Tesco, Virgin Media, and Talk Talk. BT has the largest market share at about 30%. However, that is only a recent part of the story, since the company's acquisition of EE. Historically, BT's strength has been in its fixed-line business, with the key increasingly being its broadband positioning.

Besides BT being number one in broadband with 37%, we think it is important to focus on what is lacking at the other three full-scale wireless operators: Vodafone, Telefonica, and 3 UK. Virgin Media and Talk Talk act as MVNOs on the wireless side and only have about 4% and 1% market share, respectively. These two have much larger market share in broadband at 20% and 12%, respectively, but Talk Talk again owns no network, although it is a reseller of BT’s capacity. Likewise, Sky SKYAY (at 24%) is a reseller of BT capacity. Only Virgin Media owns its own competing broadband network. When Virgin Media completes its Project Lightning in 2019, it will have expanded its cable TV network to cover approximately 70% of the U.K.’s population, up from about 55% before it started. This means that even after completion, roughly 30% of the population will still depend on BT’s network for broadband, either directly or through a reseller. It also means that currently, an additional 37% of U.K. broadband subscribers ultimately rely on BT’s network. While BT would rather have those subscribers on its own network so it can more easily cross-sell other services, its operating margin for resellers is similar to that for direct customers, as it avoids the costs of servicing them.

Where Sky and Virgin Media have an advantage over BT is in pay TV. Sky dominates the pay-TV market in the U.K. through its satellite system, and Virgin Media owns virtually all of the cable TV subscribers in the country. They both have more pay-TV customers than BT.

While BT’s weaker pay-TV positioning makes it more difficult to cross-sell a full quad-play service, it is still the best-positioned company, given Sky’s and Virgin Media’s much weaker positions elsewhere. Sky launched an MVNO wireless service last year, but it is tiny. With BT’s excellent positioning in fixed-line telephony, wireless telephony, and broadband, we believe it will increase its focus on offering converged services.

Openreach Surviving Despite Being a Political Football BT's Openreach division owns the fixed-line telephone network in the U.K. and wholesales capacity to other operators. Thus, it determines how, when, and where the network is upgraded from the old copper network on which telephone systems were originally built. In 2005, Ofcom became the first regulator to require a functional separation between the physical telecom network and the operating business. BT's other divisions, primarily consumer and business, became Openreach's biggest clients. On a gross basis, Openreach accounts for about 19% of BT's revenue. However, about 58% of that is internal revenue paid from other divisions. Thus, on a net basis Openreach only accounts for 9% of BT's revenue. We expect this percentage to shrink in the short term due to mandatory price cuts from Ofcom but to largely recover over time as demand for fiber-based broadband increases. Ofcom's continual involvement in the prices that Openreach charges is partly driven by its having the highest EBITDA margin (almost 50%) of any of BT's divisions. This has allowed it to generate a return on capital above what Ofcom has designated appropriate. We expect Openreach's margins to decline as Ofcom continues to mandate lower pricing.

However, in our opinion, Openreach isn’t important because of the size of its revenue or margins; it is because the entire fixed-line and broadband network in the U.K., excluding Virgin Media and a few other small pieces, depends on its network. This makes it very political and brings Openreach into constant interaction with Ofcom. After Ofcom ordered the functional separation of Openreach, a few other countries followed, with Australia and New Zealand going a step further and creating full separation. In Australia, the government is building its own next-generation network, which includes acquiring some assets from Telstra, the incumbent telecom operator.

In 2017, Ofcom conducted a major review of communication markets in the U.K., which is required every decade. During this review, most of the other telecom operators, led by Vodafone, Talk Talk, and Sky, requested that Ofcom follow Australia and force a full separation of Openreach. These companies’ reasoning was that BT was rolling out fiber to the premises, or FTTP, too slowly, and its plan to use G.fast (which is more like a fiber-to-the-curb, or FTTC, network) instead would put the U.K. behind the rest of the continent, which is moving to FTTP. BT’s G.fast proposal is an intermediary stage. With FTTP, fiber is taken right to the home or apartment building.

FTTP currently generates broadband speeds up to 1 Gb/second. This equals or exceeds what is capable with advanced cable TV networks that have been upgraded to Docsis 3.0. However, Docsis 3.1, which is beginning to be rolled out, can theoretically go higher. FTTP is also theoretically capable of generating faster speeds with minimal additional capital expenditures. In our opinion, to ensure that phone companies can compete long term in broadband against cable TV operators, it is necessary to build FTTP networks.

FTTC shortens the copper loop by bringing fiber closer to the premises, but not all the way, and can theoretically produce speeds over 300 Mb/second. While BT’s proposed FTTC solution is probably sufficient for most subscribers today, it isn’t upgradable without laying more fiber. Hence, the other operators view FTTC as a half-step solution that won’t be sufficient by the time it is completed, and they would rather see BT just build the FTTP network from the start, which is easily upgradable if new technology is invented. We agree. While the initial cost for FTTP is higher than for FTTC, it is less than that for building FTTC first and then adding FTTP later. Additionally, it keeps the phone companies from always trailing the cable TV operators in broadband speeds.

BT’s counterargument is that demand for broadband speeds that fast on a large scale is debatable. Thus, the company has taken the slower and initially cheaper buildout route. With Ofcom’s review underway last year, we think BT was unwilling to demonstrate the power of its position and so was slow to push convergence. Ofcom ended up with a compromise: Openreach gets an independent board, but its budget is set by and it remains 100% owned by BT. We believe that with Openreach’s ownership settled for now, BT will begin to push converged services.

In response to other operators’ complaints, Openreach announced it would pass 3 million homes with FTTP by 2020, with plans to reach 10 million by the mid-2020s, assuming favorable regulatory rulings. When it reported fiscal 2018 results, management said its discussions with Ofcom provide it with greater confidence that the fiber buildout past 10 million premises will occur. Other operators aren’t waiting around. Vodafone has announced a deal with CityFibre to pass up to 5 million homes by 2025. The first 1 million homes passed will start in the first half of 2018 and will be completed in 2021. Either party has the right to extend the deal to pass up to an additional 4 million premises by 2025. We think Vodafone’s ambition is to push Openreach to speed up its buildout of FTTP, not to own a competing network across the U.K. Building out its own network nationwide would be prohibitively expensive, and it would be difficult for Vodafone to earn a decent return on its capital. We think there is a good chance that after the Vodafone-CityFibre venture reaches its initial target of 1 million premises passed, Openreach will realize it needs to speed up its buildout. Now that Openreach appears to be catching on to the need for a FTTP buildout, however, it is constrained by how quickly it can roll out such a network. Thus, it will probably focus on the most competitive areas first. It will also supplement its FTTP buildout by increasing its G.fast rollout into other areas; this can be rolled out at a faster pace in order to maintain its competitive position.

We agree with the other operators that BT should move more quickly in building out a FTTP network, which would enable it to compete better with Virgin Media now and would also future-proof the network. Beyond the advantages fiber provides for broadband, it would also improve BT’s wireless positioning as 5G develops, which will need faster backhaul speeds that only fiber can provide.

Openreach’s main competitor in broadband is Virgin Media, the cable TV operator owned by Liberty Global LBTYA. Virgin offers broadband speeds ranging from 50 Mb/s to 350 Mb/s due to its hybrid fiber/coaxial-based network and Docsis 3.0. The top speed is already above the likely speeds reachable with G.fast, and matching or potentially beating Virgin will require FTTP. Virgin can also increase the speed of its services by moving to Docsis 3.1, which is starting to be rolled out around the world. The problem is that BT is always playing catch-up. We think it is its slow fiber rollout relative to its peers that most annoys BT’s wholesale customers and will eventually drive the company to increase the speed of its fiber rollout.

Ofcom has also created other regulatory headwinds. The regulator finished its review of the wholesale local access market in March, with the result that it is lowering the prices BT can charge other operators for some of its products. BT anticipates this will lower Openreach’s revenue and profit by GBP 80 million-120 million. We are a bit frustrated that Ofcom continues to regulate pricing at such a granular level despite the increases in competition, and despite the regulator having claimed for the past decade that it would relax micromanaging at the pricing level if competition were to increase. Despite these changes, we anticipate BT can offset these declines in other parts of the company.

BT's Extensive Spectrum Holdings Can Carry More Data In addition to being the only operator that owns its own network and having the largest fixed-line, wireless telephone, and broadband bases, BT leads in its wireless spectrum holdings. The company, which includes both the assets BT bought directly and those acquired with EE, has significantly more total spectrum than Vodafone, 3 UK, or Telefonica. BT's one spectrum weakness is that it has a limited amount of sub-1 GHz spectrum. Sub-1 GHz spectrum has favorable propagation characteristics. It can more easily penetrate walls, thus improving indoor wireless coverage. It also travels further, which means fewer towers are needed in rural areas. EE's network was built based on 1,800 MHz spectrum, so it has the necessary extra towers to generally offset its lack of sub-1 GHz spectrum, but this doesn't offset the value of penetrating walls, so BT would like to obtain spectrum in the 700 MHz auction expected next year.

Due to BT’s large spectrum holdings, 3 UK sued Ofcom to restrict BT’s ability to bid on the next lot of spectrum. BT countersued, but Ofcom won and both suits were thrown out. As such, Ofcom’s rulings stood. This prevented BT from bidding on the 40 MHz of spectrum in the 2,300 MHz band, as it was already at the cap of 255 MHz on immediately usable spectrum, but it was allowed to bid for a portion of the 150 MHz of spectrum auctioned in the 3,400 MHz band. There was a secondary cap on total spectrum of 340 MHz. BT ended up acquiring 40 MHz of spectrum in the 3,400 MHz band, which leaves it with 45 MHz of available capacity before it hits the total cap restriction. We assume the cap on usable spectrum will be increased before the auction of the 700 MHz band. We believe BT intentionally left space in its total spectrum cap for bidding in this auction, as we think it needs more sub-1 GHz spectrum capacity. Ofcom’s overarching ruling is that BT can’t own more than 37% of usable spectrum.

Global Services Less Important Than in the Past BT's stock has had three major downturns this century. The first was when the tech and telecom bubble burst. BT came close to bankruptcy and had to sell its wireless business and cancel its dividend to stay afloat. The second was primarily caused by its global services division, which at the time was probably the second-largest global enterprise telecom business after AT&T. At that point, BT realized it had accrued revenue for several projects that turned out to be nowhere near as profitable as expected. It consequently took write-offs totaling over GBP 1 billion, and its EBITDA barely stayed in positive territory. This downturn was exacerbated by the global financial crisis and led to the dividend being cut by almost 80%. The third downturn is where we are today. Once again, the global services division is struggling. However, revenue has declined for several years as BT has focused more on profitability than revenue, higher revenue in the consumer and business divisions, and the acquisition of EE. As a result, global services now accounts for a much smaller portion of companywide revenue than in the past.

In fiscal 2009, the global services division accounted for 40% of revenue; now it accounts for about 21%. Its EBITDA margins have deteriorated, but they are still far higher than in 2009 and are coming off what we believe was an unsustainable level. Thus, part of the most recent pullback was deserved, as the market seemed to believe these margins were sustainable and some even predicted room for further expansion. We did not agree and rated the stock 2 stars for most of 2014 and 2015.

Also, similar to the previous stock pullback, the global services division has been hit by a write-off. This time, it was due to overstating results in Italy for several years, which caused a write-down of GBP 245 million. This deception required multiple people’s involvement, and the management team in Italy has been replaced. Additionally, the practices that allowed this to happen have been changed and brought into alignment with other countries. As an extra precaution, a detailed investigation into seven other countries’ operations was undertaken to verify that these problems were unique to Italy. Ultimately, we think the global services’ sustainable EBITDA margin is around 14%, and just as investors were overly optimistic from 2014 to 2016, they are overly pessimistic now.

Another way to view the reduced size of the global services division is to look at its backlog, which was cut by more than half from fiscal 2005 through fiscal 2018. These orders can be lumpy, but the trend is certainly down, which we expect to continue for a couple of more years. Part of this decline is intentional, as BT has focused more on margins and has intentionally dropped some low-margin business. Eventually, we anticipate the division’s margins to become less volatile.

Retail Margins Squeezed in Short Term The retail division has seen solid broadband subscriber and revenue growth, primarily driven by BT's push into sports broadcasting. The company has acquired the U.K. TV rights to UEFA Europa and UEFA Champions League football, two tranches of Premier League football, and several other sports. It first launched in August 2013. Not only did the sports channels attract more viewers to BT's pay-TV service, but also to its broadband service, as the sports service was initially provided free to paying broadband customers.

This strategy initially made us very leery. The company spent roughly GBP 2 billion on sports content rights, and we questioned whether it could generate a return on that investment if the service was provided for free. Part of the answer was that it roughly doubled BT’s number of pay-TV subscribers, which were paying for the service, and in 2017, it finally began to charge some broadband clients for access. Currently, all broadband subscribers are paying something for the sports channels. Finally, in 2017, BT and Sky agreed to sell each other’s Premier League packages. This allows BT’s subscribers to also watch Sky’s much larger package of Premier League games without jumping ship and provides some wholesale revenue from Sky, whose customers can buy the games BT shows.

Now that customers are paying for the sports channels, we think there is a chance of BT actually generating a decent return on its sports rights investment. However, in the short term, the high cost of programming is squeezing margins in the retail division. That said, in the auction earlier in 2018, BT agreed to pay just GBP 885 million for the next three years of Premier League Football versus its current contract of GBP 960 million, though the current contract allows BT to show more games per season. Sky also bid less for the next round of Premier League games, so we believe some pricing rationality is returning to content costs.

Business and Public Sector Division Hurt by Reduced Spending Ever since the financial crisis, the U.K. government has been reducing its spending on telecommunications and information technology. Projects are being reviewed for criticality, with some being dropped and others renewed at lower prices. New programs are few and far between. We expect the public sector will continue to struggle for a couple of more years.

On the corporate side, more-profitable legacy system projects are ending, and newer programs tend to have lower revenue and margins. However, we believe the worst of this transition is almost over. New projects, including security and cloud-based systems, are growing nicely but aren’t yet big enough to offset the legacy declines.

While the fiscal fourth quarter continued to see some end-of-year budget-flushing among public-sector and major business customers, the trend from this subsector is clearly negative. Eventually, though, the government and corporate sectors will need to update their networks. We expect this will take longer with government, especially with Brexit approaching, which will keep the pressure on to reduce costs. Still, we expect revenue declines to slow. With corporations, we anticipate a return to growth in fiscal 2021.

With continued slow growth from small and midsize enterprises and Ireland, we believe the division as a whole can return to revenue growth in fiscal 2022, despite continued declines in the public sector.

Large Pension Deficit Weighs on Stock BT's pension assets are more than double its market capitalization, but its pension liability is even larger. At the end of its last fiscal year, the pension fund had assets valued at GBP 49.9 billion but a pension obligation of GBP 55.8 billion, or a pension deficit of GBP 6.4 billion gross of tax. The deficit declined by GBP 2.8 billion in fiscal 2018, the biggest reduction in many years.

BT is required to meet with the government for a triennial review of its pension. The review for the deficit as of June 30, 2017, was completed in May 2018 and showed a deficit of GBP 11.3 billion. The methodology for measuring the deficit for the triennial review is different from the company’s quarterly pension review, which is what we use in our model. The differences in methodology and cash contributions BT has made since June reduce the deficit to GBP 6.4 billion. The main difference in the methodologies relates to the discount rate. For the quarterly review, BT has adjusted to using an average of AA rated corporate bonds that have no implied government guarantee like BT’s debt has, owing to its previous government ownership and some government agreements to cover some liabilities in the event of a BT bankruptcy from the time it was privatized.

The review includes an agreement on how BT will fund the GBP 11.3 billion deficit. BT has agreed to make cash payments of GBP 2.1 billion over the next three years, which includes GBP 850 million paid in March, and to issue the fund GBP 2 billion in bonds that will be redeemable between 2033 and 2042. Additionally, BT will pay about GBP 900 million annually from 2020 through 2030. However, the next triennial review will begin in July 2020 and will probably be agreed on after only one of those GBP 900 million payments has been completed. With interest rates likely to have risen by then, we expect payments beyond 2020 will actually be much lower than has been agreed.

Because of the huge size of the pension liability, small changes in its value have a magnified effect on BT’s valuation. Owing to lower interest rates, BT’s pension deficit continued to generally increase until the fourth quarter of fiscal 2018. As rates have dropped over the past decade, the present value of its future liabilities has increased due to the lower discount rate. While lower interest rates have also benefited the company through lower interest expense on its debt, this gain has been less than the pain on the pension deficit. We believe interest rates have probably bottomed and rates are likely to increase from here. While we anticipate the change will be gradual, we do think BT’s pension deficit will benefit rather than be hurt by interest rate moves. Just as interest rate declines hurt the pension deficit more than they benefited the company’s interest payments on its bonds, we expect interest rate increases to benefit the pension deficit more than they will harm interest expense on BT’s bonds.

Currently, BT’s debt is significantly below the European telecom average when compared with expected 2018 EBITDA. Even adding the pension deficit to net debt only puts the combined net debt/EBITDA ratio equal to the average.

We Believe the Dividend Is Safe When BT last got into financial trouble, it slashed its dividend. The previous time, it eliminated the dividend completely. The market seems to think this could happen again, but we disagree. In August 2009, BT cut its final dividend for fiscal 2008 from GBX 10.4 to GBX 1.1 and the December 2009 interim dividend for fiscal 2009 to GBX 2.3 from GBX 5.4 the previous year. Since then, the company has steadily increased the dividend.

When management announced fiscal 2018 results, it said it would hold the dividend steady for fiscal 2018 and the following two years due to plans to increase capital expenditures in order to speed up the rollout of fiber and a 5G wireless network. We believe the dividend is secure and could start to increase again in fiscal 2021. In 2009, without the dividend cut, BT’s free cash flow would have turned negative. That is not the case today.

Reported EBIT in fiscal 2017 declined due to extraordinary costs related to the acquisition of EE and fines in Italy and the U.K. In Italy, BT was found to have manipulated its revenue for years to the tune of GBP 268 million. Beyond adjusting its previous year’s results, the company was fined GBP 245 million. In the U.K., Ofcom found BT had not properly compensated other companies when it failed to complete ethernet connections on time. The regulator fined BT GBP 42 million, the largest fine Ofcom has ever given. However, with the removal of these one-off items, BT’s EBIT and free cash flow have held up fairly well, with just small reductions in fiscal 2018 and expected in fiscal 2019. Additionally, despite our projections for continued dividend increases, we anticipate the company will hit a new high of free cash flow (post-dividend) of over GBP 2.2 billion by 2023. Even when adding in the required additional pension payments, free cash flow remains positive. The reason for zero cash pension payment in fiscal 2019 is due to the issuance of the GBP 2 billion bond, which doesn’t affect this year’s cash flow. We believe this free cash flow generation provides plenty of room to continue increasing the dividend, make extra payments into the pension fund, and invest in the business.

Clearly, BT’s free cash flow has benefited from the steadily declining tax rate. We model the company as if currently passed tax rate declines will occur. One risk is Jeremy Corbyn, the leader of the U.K.’s Labour Party. He is leading in the polls, and if he is elected as the next prime minister, there is a good chance that taxes not only won’t be reduced but instead might be increased. If the U.K.’s tax rate were to increase to 25% starting in fiscal 2020, our fair value estimate would drop about 9%. Free cash flow after the dividend with a 25% tax rate would decline by almost GBP 400 million in fiscal 2022. While this is significant, it wouldn’t require a cut in the dividend. Of course, an increase to a higher tax rate would be even more damaging. Even though a 25% tax rate would push free cash flow slightly negative in fiscal 2020, we don’t think management would cut the dividend. However, it might choose to hold it steady for a longer period.

BT's Fundamentals Deserve a Higher Valuation We believe the market is punishing BT's stock too harshly for some legitimate concerns. The market seems to be focused on the global services division's EBITDA margins, which are currently depressed. While we don't expect the division to ever sustain an EBITDA margin in the high teens or 20s, which it has hit in some quarters, we do believe it can average about 14% versus the roughly 9% it has averaged over the past two years.

In addition to our dividend discount model, which generates our fair value estimates of $24 and GBX 360, we look at enterprise value/EBITDA multiples. BT trades at 4.4 times enterprise value to our estimate of 2018 EBITDA, which is currently the lowest multiple of the European telecoms we cover. Adding in the pension deficit pushes BT’s EV/EBITDA up to 5.3. Additionally, the pension deficit is partially exaggerated due to the artificially low interest rate environment, which is beginning to reverse. In November 2017, the Bank of England raised the bank rate to 0.5% from 0.25%. At its last meeting in May, the central bank maintained the rate at 0.5% but reiterated that it expects continued but slow additional rate increases.

Also, the stock yields 7.2%, with a dividend that we believe is sustainable. This means shareholders are being paid to wait for the market to recognize BT’s value. This is currently the second-highest yield in European telecom and higher than Verizon’s VZ or AT&T’s T in the United States. With the stock’s continued pullback, BT moved into 5-star territory on May 18, making it one of only two 5-star-rated European telecom companies we cover right now.