Kerrisdale Capital: Intelsat S.A. (I) And SES S.A. (SESG) UpdateUmair Tariq
Disclosure: We are long shares of Intelsat S.A. and SES S.A. Please review our disclosures at the end of our email.
Since we released our report on Intelsat and SES a month ago, events have unfolded even more rapidly than we expected, and our confidence in the upside potential of these stocks, largely due to the underappreciated value of their US C-band spectrum, has only grown. Below we briefly summarize some of the key developments.
C-Band Momentum Continues
On July 12th, the FCC held an open meeting at which the commissioners voted unanimously to adopt the Notice of Proposed Rulemaking (NPRM) on “Expanding Flexible Use of the 3.7 to 4.2 GHz Band,” officially setting in motion the regulatory process that will allow Intelsat and SES to unlock the value of their C-band spectrum rights. The text of the NPRM came as no surprise, since a highly similar draft had already been publicly circulated ahead of the meeting. What did come as a surprise, though, was how passionately key commissioners argued in favor of the Intelsat/SES proposal and against other possible outcomes. In particular, Commissioner Michael O’Rielly, a Republican who for two years has spearheaded the search for a workable way to harness the C band for 5G, spoke out strongly (emphasis added):
The reality is that if everyone puts aside preconceived notions, a final proposal, in the very near future, can address everyone’s concerns and needs. Cooperation and avoiding unnecessary tangents will be paramount. … There are, however, some things I would have done differently than what is contained in the item. As previously stated, it is of utmost importance that this proceeding is concluded and spectrum is released into the marketplace quickly. There can be no unnecessary delays or distractions. Parts of this item, while interesting, are not practical and unlikely to be adopted. … Consider that the record clearly supports a market-based approach, but the item veers off seeking comment on various auction mechanisms, many of which were not suggested in the record and some of which are incredibly complex or downright troubling. … I am concerned that such ideas may detract time and attention from more viable options.
Although there are still many details to be worked out about the market-based approach, it is easy to see how the spectrum can be expeditiously put into the hands of the wireless industry and how the satellite industry and its broadcast, cable, and other customers can be made whole. The money received by the satellite industry from these free market negotiations will cover the costs of repacking their users, improving technology and filters, launching new satellites, laying fiber, or moving their customers to new technologies or satellite systems. It is not clear how some of these issues will be tackled under some of the other ideas proffered in this notice.
In short, Commissioner O’Rielly stated clearly that the Intelsat/SES market-based approach makes the most sense and can best achieve the goal of getting spectrum to the market for 5G quickly. Every other reallocation concept, in his view, is just a distraction. Republican Commissioner Brendan Carr likewise highlighted the importance of mid-band spectrum like the C band for winning “the race to 5G” and said specifically:
I want to draw particular attention to the notice’s section on a market-based mechanism for clearing the spectrum. … In my view, this could provide the quickest path to clearing the spectrum, and it could do so without the inevitable issues that arise when the Commission begins imposing mandates and repurposing the spectrum itself.
With the Republicans on the Commission strongly backing the Intelsat/SES proposal, votes from the other side of the aisle are unnecessary to secure approval of a final order. But Democratic Commissioner Jessica Rosenworcel not only voted for the NPRM but hit on many of the same themes as her Republican colleagues in her own statement, in particular the importance of staying ahead of other countries in terms of allocating mid-band spectrum for 5G use. While she did point out that “what incumbent providers [namely, Intelsat and SES] stand to reap from a secondary market sale of repurpose spectrum is significant,” she didn’t describe this potential profit as a reason to shoot down the proposal. Rather, she spoke broadly in favor of “a framework to ensure that this approach truly serves the public interest” – a relatively low bar for Intelsat and SES to meet, given the broad-based consensus about the economic benefits of accelerated, widespread 5G deployment.
Overall, we find it hard to see how the FCC process could be going any better for the satellite operators. In keeping with this positive momentum, Intelsat, while acknowledging the inherent uncertainties of any regulatory process, has now stated publicly, “We believe that it is possible the FCC may issue its final Report and Order in early to mid-2019.” One precedent for such a relatively fast-paced process is the so-called Spectrum Frontiers proceeding under prior FCC Chairman Tom Wheeler. That proceeding, which granted expanded flexible-use rights to incumbent millimeter-wave licensees previously authorized only for fixed (not mobile) use – all in the name of 5G deployment and speed to market – went from NPRM to final order in less than nine months, consistent with a first-quarter 2019 completion date for the C-band proceeding. If anything, the current FCC (and Commissioner O’Rielly in particular) have put even more emphasis on speed, suggesting that this proceeding might wrap up even faster.
One theoretical concern that previously hung over the Intelsat/SES proposal was that small, marginal C-band satellite operators in the US might try to hold the process ransom, refusing to cooperate unless promised an unfair share of the eventual C-band proceeds. While we never believed this holdout scenario was a realistic worry – Intelsat, SES, and the FCC simply have too many ways to twist the arms of potential holdouts – the risk has largely been eliminated now that, as of July 12th, Eutelsat, the only other C-band operator with a meaningful US presence, has officially joined forces with Intelsat and SES. The road ahead is clearer than ever. By this time next year, we believe, the satellite operators could already have inked a deal to monetize the first tranche of their spectrum.
The Value of the C Band Comes More Clearly into View
One of the key points we made in our original report was that investors and analysts have been wrong to focus exclusively on just the initial 100MHz clearing target set forth by the satellite operators. In reality, almost the entire band will eventually be in play, as long as its economic value for 5G terrestrial networks sufficiently exceeds its economic value to incumbent users (primarily media companies). In our discussions with other market participants, some have expressed concerns that SES has taken a more cautious stance on the C-band opportunity than Intelsat has, giving rise to fears that it might (irrationally, in our view) refuse to consider clearing more than the first 100 MHz. On this quarter’s earnings call, however, SES management put these fears to bed with more open-minded and optimistic commentary. In the words of SES CEO Steve Collar:
[T]he really encouraging thing about the NPRM is that it carries out our market-based proposal as the lead…We genuinely think this is an opportunity for a win-win. The only way that we can protect our customers, protect our video neighborhoods, and, importantly, from the FCC standpoint, free up important spectrum for 5G in the fastest possible way – it really is the only way to do that. … [T]he U.S. hasn’t got 10 years to [wait for] 5G, and that’s why this sort of market-based approach gives us, if we control this process and drive this process, we’re able to deliver that spectrum in a way that continues our business, continues our customer business that serves hundreds of millions of users in the U.S, gives us the opportunity to actually enhance those services and frees up that spectrum. … As for the amount of spectrum, look, I mean, we've been pretty clear that we have a plan that closes for 100 megahertz. We’re also – we also acknowledge, let’s say, the desire for more spectrum to be freed up, and we’re working on it, right? We're working on our plans to look at what we think is actually achievable with the absolute focus – laser focus on making sure that we can deliver for our customers and deliver not only current services but whatever our customers need in the future. … [W]e are equally engaged as good partners to make sure that whatever we can free up, we will do so and at the moment, that's 100 megahertz. We’re working hard to see what's possible.
While Collar was careful to stress his commitment to protecting incumbent users, he left the door wide open to additional spectrum clearing beyond the first 100 MHz, potentially in multiple discrete transactions with several counterparties (e.g. Verizon and AT&T). While it behooves neither SES nor Intelsat to get ahead of themselves in prematurely committing to specific targets, it’s becoming clearer that the two companies are looking at the medium-term opportunity beyond 100 MHz in the same logical way. (Indeed, SES has begun to put its money where its mouth is, offering to reimburse incumbent C-band earth-station operators for filing fees that they must pay when registering their antennas with the FCC to ensure legal protection from future terrestrial interference – a move presumably designed to foster goodwill and encourage constructive cooperation.)
If SES’s openness to freeing up more bandwidth provides greater clarity on C-band supply, Intelsat’s earnings-call commentary hinted at the strength of C-band demand. In response to a question about potential dialogue with “interested parties” (i.e. potential spectrum buyers, like the mobile carriers), Intelsat CEO Stephen Spengler said:
[W]ell, of course, we are engaged with the interested parties now. We’re engaged with all the stakeholders in this effort. And so we understand the importance of this spectrum to various players, and we’re engaged with them to try to understand their specific needs and technical needs in terms of operating the spectrum. So those dialogues are going on right now. … [W]e’re engaged with the various parties now to make sure that we are able to move quickly as possible once we see all the criteria come together for execution.
While it’s no great surprise to hear that the satellite operators are already engaged with potential spectrum buyers, it highlights the fast pace of the process and the strategic importance of the band.
Similarly, on July 25th, at the Senate Commerce Committee’s hearing on “The Race to 5G,” the head of the cable company Charter’s wireless technology efforts proactively noted in his prepared remarks that “[t]he lower C-band spectrum (3.7GHz – 4.2GHz) … holds promise as it provides both meaningful bandwidth and RF propagation that could enable ubiquitous 5G mobility.” He also repeatedly mentioned Charter’s interest in the band in response to questioning. Though we view Charter as a relatively unlikely buyer of at least the first 100 MHz of C-band spectrum (given the spectrum’s greater value to large-scale wireless players like Verizon and AT&T), it certainly doesn’t hurt to have a firm with a ~$150-billion balance sheet kicking the tires.
Meanwhile, the technology and infrastructure for mid-band 5G continue to move forward. On July 23rd, Qualcomm announced the unveiling of its 5G modules for smartphones and other mobile devices, including a sub-6 GHz module designed for “optimum massive MIMO applications” with “support for 3.3-4.2 GHz (n77), 3.3-3.8 GHz (n78) and 4.4-5.0 GHz (n79) sub-6 bands.” The numbers prefixed with “n” refer to official bands defined by the cellular standards body 3GPP. The C band is thus already part of the 5G standards and already part of Qualcomm’s first-generation technology.
On the base-station side, American Tower, the largest cell-tower owner in the US, took pains on its July 31st earnings call to highlight the growing interest among carriers in mid-band spectrum, including the C band, as key to widespread 5G deployment. American Tower’s CEO noted that “previously undeployed mid-band spectrum assets will be critical components of 5G rollouts in the future that are macro focused, which will drive incremental growth opportunities in our towers. This is one of the things that we concluded in our most recent technical assessments…that mid-band spectrums can be very, very important in the U.S. and the deployment of 5G throughout the country.”
In our previous work, we have stressed that, with the benefit of an important 5G technology known as massive MIMO, C-band spectrum can enjoy propagation similar to that of bread-and-butter cellular bands in the vicinity of 2 GHz, enabling it to work with existing cell towers instead of requiring dense, costly small-cell buildouts. American Tower made the same point:
And then there’s another band, a little bit higher, 3.7 to 4.2 gigahertz, it’s called the C band, and the government is actually actively looking at how to get that ready for deployment over the next few years. … [P]ropagation characteristics as far as coverage aren’t quite as wide as you get in the low band, of course, but we are starting to get modeling that shows that these kind of frequency ranges, 2.5 to, say, 4.2, are ideal for wide-area network mobility and capacity building, right? So again, macros and rooftops, tower macro sites are going to be again very well-positioned for this in the suburban areas and the less dense urban environments where we are also have presence. This kind of spectrum, you can actually aggregate 8- to 16-layer MIMO kind of technology using the – with the spectrum that’s available for this in those mid-bands. And so this is the most important emerging facet of 5G deployment expectations for us, at least for me, is that this mid-band spectrum can play kind of a foundational role for this going forward.
With the C band playing a “foundational role” in 5G deployment, boasting a blend of bandwidth and propagation that is “ideal for wide-area network mobility and capacity” and compatible with existing tower-centric cellular infrastructure, we believe the bidding will be fierce.
Core Satellite Operations Are Stable
Though our report and investment thesis focus on the C band as the key driver of upside for Intelsat and SES, recent quarterly earnings results reinforce our belief that the companies’ core business trends and outlooks remain stable.
For SES, despite market consternation over potential changes to guidance (as evidenced by the ~15% pullback in shares leading into earnings), the company delivered 2Q revenue and EBITDA modestly ahead of median consensus forecasts and reaffirmed FY 2018 guidance (with revenue revised upward to the top end of the range). On an underlying basis (excluding satellite health issues and IFRS accounting changes), Video revenue contracted -0.2% in the quarter (vs. -0.9% in 1Q18). Sequential improvement was also evident in Networks, where growth rates from Mobility, Government, and Fixed Data sub-segments all accelerated vs. 1Q18. 2020 guidance, a key source of debate among investors given the ambitious targets implied for a Video segment facing secular headwinds, was revised downward as anticipated – roughly -3% at the midpoint for revenue and -5% for EBITDA – driven by more “prudent” expectations of Video growth. But this proved a non-event, as the magnitude of the 2020 guidance revision was in line with consensus expectations.
Intelsat’s earnings were a bit more mixed but with the same punchline: results broadly in line and FY guidance reiterated. Total revenue fell 3.9%, consistent with the rate of decline witnessed in 1Q18. If not for some collection issues stemming from media customers in Iran and Thailand, we believe the slight miss versus consensus estimates would have been a slight beat. Adjusted EBITDA was 1% ahead of consensus estimates, benefiting from sequentially lower direct costs. As even an underweight-rated sell-side analyst conceded, on balance the quarter was “solid.”
Though lacking in fireworks, 2Q18 earnings for SES and Intelsat were important events in that the core businesses for both exhibited steady to improving underlying trends. With these results now in the rear-view mirror, investors are free to refocus their attention on the path to C-band monetization and the enormous upside it offers – not in some far-off future but, we believe, within months.
Article by Sahm Adrangi, Kerrisdale Capital