|Shares Out. (in M):||709||P/E||0||0|
|Market Cap (in $M):||16,660||P/FCF||0||0|
|Net Debt (in $M):||0||EBIT||0||0|
Altice USA ("the Company” or "ATUS") is a significantly better business than the market currently believes, trading at a valuation that sets the equity up for a double over the next 3 years. Idiosyncratic concerns about unsustainable margins at ATUS, as well as broader cable industry worries around cord-cutting and the threat of 5G fixed wireless, are either misunderstood or greatly exaggerated and have distracted the investment community from the highly attractive upside available in ATUS over the intermediate term.
Trading at 9.5-10.0x 2019E equity FCF and executing a sizable, levered buyback of $1.5bn per year, I believe ATUS has 80-120% upside or 18-25% IRRs through 2022, with the lower end of that range assuming no re-rating (10x FCF exit multiple) and the upper-end assuming an 8% FCF yield (12x FCF), which would be consistent with other pure-play cable peers and still sensible given a long runway for mid-single digit free cash flow growth.
In addition to the ATUS-specific and general industry concerns summarized above, the sell-side has also largely mis-modeled the Company’s capital spending trajectory, with consensus assuming flat or ever-increasing CapEx well beyond ATUS’ costly fiber deployment (currently running at an estimated $450m per year, or 1/3rd of total CapEx).
Altice USA is comprised of two primary cable assets: Optimum (previously branded as Cablevision) and Suddenlink.
Optimum is concentrated in highly dense, affluent areas including Long Island, The Bronx, Brooklyn, Northern New Jersey, southern Connecticut, and Westchester County (suburbs outside of NYC), and competes with Fios’ fiber product across c.50% of its passings. Outside of these passings, there is essentially no competition other than some legacy U-Verse and DSL which gives ATUS the enviable position of what is in practice a cable monopoly within one of the most densely populated, wealthy markets in the US.
Suddenlink operates primarily in Texas, Louisiana, Arkansas, Oklahoma, Missouri, West Virginia, and North Carolina, and contends with minimal fiber overbuild (10% as of 2015 – it is very likely somewhat higher now following AT&T’s required fiber build-out, which took place in some of Suddenlink’s markets). Legacy products (generally sub-100Mbps) from AT&T, Centurylink, Frontier, and a few others represent 80-90% of Suddenlink’s remaining competition.
Like essentially all publicly traded cable companies in the US, ATUS has been shedding video (43% of revenues) subscribers at a 3-4% rate, but growing internet (30% of revenue) subscribers at 2-4% rate and overall internet revenue at 10-12% given (i) significantly increased uptake of higher speeds and (ii) a higher concentration of customers taking internet-only offerings, which carry higher standalone prices than when bundled.
Video subscriber losses decelerated from 4% to 3% in 2018 vs. 2017, which appears to have been primarily a function of more rational pricing on the part of OTT providers. The Company typically institutes a price increase across its entire video footprint every year just to offset increases in programming costs, which results in c.0-2% annual revenue declines in video. I believe – and ATUS management has stated – that video remains a profitable business for ATUS, but only marginally so. On reasonable estimates, video is likely a 10-20% standalone EBITDA margin business and perhaps only slightly profitable on a free cash flow basis given associated CPE spend (e.g. set-top boxes). However, rather than being a negative, the fact that video generates only a small amount of ATUS’ overall EBITDA (c.15-20% of total) and even less of its FCF evinces the point that very little of ATUS’ value is related to video.
The profitability problem in video is actually not so much a function of subscriber declines as it is increases in annual programming costs, which have been increasing at a high single digit rate (per subscriber) for several years with no end in sight. Over time, ATUS and all of its cable peers will need to restructure the way they market and provide video products, as the business is clearly trending toward being loss-making as broadcast retransmission and sports programming costs continue to increase at mid-teens annual rates. Comcast has already partnered with Netflix to include its service as part of other Comcast offerings; I think it is likely that at some point over the next 5 years, ATUS and other cable providers will essentially become resellers of OTT products (everything from Netflix and Hulu to YouTube TV) and take a revenue split. If this were to develop, I think overall FCF margin dollars would actually be higher than they are today given the absence of programming costs, CPE, and customer service calls. Nevertheless, I have basically extrapolated recent trends in video, assuming revenue declines at a c.1% annual rate and programming costs continue to increase at 8% per subscriber.
[Note: Charter’s management has also stated this it expects its video business to no longer be profitable within the next several years, at which point it would be “indifferent to video on a standalone basis.” Providing video services at a loss seems like an unsustainable dynamic for cable distributors; deeper relationships with OTT providers seems to be where things are headed].
While video subscriber losses continue to grab a large share of the headlines, the majority of ATUS’ value is very clearly concentrated in its broadband business, which has continued to grow residential relationships at low-single digit rates as customers of legacy telecom products continue to churn and upgrade to high speed data (“HSD”). Layered on top of 2-4% subscriber growth is 6-8% ARPU growth, which is driven by rapidly increasing uptake of higher speeds, features (i.e. whole home WiFi), and standalone HSD offerings which typically carry higher allocated ARPUs than they do when bundled with other products. Accordingly, ATUS’ overall HSD revenues have been growing at 10%+ for over 3 years and, given a large block of subscribers still on legacy telecom services and rapid increases in both data consumption (+25% annually) and overall speed tiers, this approximate rate of growth is likely to continue. For conservatism, I have assumed a step decline in HSD revenue growth, resulting in a revenue CAGR of 6% through 2023.
The majority of ATUS’ other revenue (27% of total) is in surprisingly attractive businesses.
14% of ATUS’ revenue is in its Business Services segment, which consistently grows top-line at 5-6%. The majority of the value within this segment is the Lightpath business. Lightpath’s key business is IP, Ethernet, and other connectivity services which it sells to enterprise and other business customers across c.6,000 fiber route miles; Lightpath also performs some fiber backhaul for cell carriers, which is a beneficiary of the transition to 5G. Management has suggested they might sell Lightpath given high valuations in the space (typically 15x EBITDA for similar assets), which would generate c.$3 billion of value given a rumored $200m of EBITDA for the business. I am indifferent as to whether they sell the business. I understand the logic behind selling into an active M&A market where valuations are relatively high and using the proceeds to repurchase undervalued equity or other cable assets, but I also believe the Lightpath business is a unique asset (significant high capacity fiber infrastructure in the most attractive market in the US) that continues to grow nicely, and it may be worth retaining the optionality to aggressively scale the business.
The Company also has a nascent advertising business (5% of revenue) which grew 23% during FY 2018 and is basically a play on targeted advertising. Much of this growth was driven by political year ad spend, but the business appears to be getting some traction and along with Business Services is a helpful offset to modest top-line declines in video.
Landline telephone services represent the remaining slice of ATUS’ revenue and are declining at a mid-single digit revenue rate. However, the Company is currently in the process of rolling out its MVNO with Sprint, which will provide some additional revenue growth at an attractive margin profile as the business scales. While some prospective investors may be scared off by the significant upfront losses incurred at Comcast and Charter as they deployed their MVNO offerings, there are some key fundamental differences between those MVNOs (both with Verizon) and the one ATUS has with Sprint.
1. Economics – ATUS receives favorable discounts on its mobile network lease costs in exchange for permitting Sprint to use its own infrastructure to mount small cells, and to also use Altice’s cable network for cellular backhaul (by linking these small cells to Altice’s network). Sprint has publicly stated that given the depth of the relationship and shared infrastructure as part of this MVNO, both companies are receiving highly compelling economics.
2. Network Flexibility – Related to favorable network lease costs, ATUS has stated that “we basically own and control everything apart from spectrum and base stations,” and it is able to direct mobile data to travel over its own cable network within its footprint, which dramatically reduces use of Sprint’s network and contains lease costs.
3. Bring Your Own Device – ATUS has elected to go with a BYOD model for its MVNO, which eliminates the need for working capital burn associated with device sales. This may have the impact of slowing the rate of customer uptake but seems like a sensible framework to start with.
4. Near-Term Profitability – As a result of these three factors, ATUS management has stated the MVNO launch will be cash flow profitable fairly quickly and that it only spent $50 million in total CapEx to prepare for deployment. In contrast, Comcast and Charter have both incurred significant cash flow burn in launching their respective light MVNOs with Verizon, which is partly a function of higher network lease expense and partly a reflection of those companies’ decisions to sell handsets to improve customer penetration. Based on a haircut of Comcast’s trajectory of mobile adds, I assume only c.3% of ATUS’ eligible subscriber base (c.400,000) subscribes to the mobile service by 2023, which generates c.$200m of annual revenue (sub-$40 APRUs) at 30% EBITDA margins.
Summary of Assumptions
Briefly summarizing, I have assumed a -1% CAGR for video, 6% CAGR for HSD, -7% for Telephony, 5% for both Business Services and Advertising, and assume the Mobile business generates $200m of revenue in 2023. This results in an overall revenue CAGR of 2.8% through 2023.
On the cost side, I’ve assumed programming costs / sub continue to increase at an 8% rate, and that Cable OpEx ex-programming is flat over the period. I think this is a highly conservative assumption given (i) some remaining near-term synergies (billing systems integration and some real estate savings initiatives, which are currently underway), and (ii) likely, significant cost savings associated with the Company’s deployment of Fiber-to-the-Home (“FTTH”) across its Optimum footprint (further detailed later in this write-up). I assume Mobile OpEx based on a 30% EBITDA margin on modeled revenues.
ATUS has been utilizing a significant NOL, and will be a cash tax payer starting in 2020 – I assume a 25% tax rate on a continuously declining D&A shield (I model only $1.8bn of D&A in 2023 vs. $2.4bn in 2018).
Management has guided to a range of $1.3-$1.4bn of Capex over the next few years as it completes the FTTH deployment, expected to be substantially completed by the end of 2022. I believe the FTTH deployment is running at approximately $450m per year, based on management commentary that they can do better than European peers who have disclosed an average all-in cost of $580 / home passed associated with FTTH deployment. Assuming management believes they can do this for 25% less than European peers, that works out to a total cost of $2.3bn to fiberize the entire Optimum footprint of 5.2m passings, or $450m / year. It is also worth noting that CapEx was $1.0bn in FY 2017 when there was essentially no FTTH spend.
I assume average CapEx of $1.4bn / year through 2022, declining to $1.0bn in 2023 or the same as the pre-FTTH figure seen in FY 2017. I think this will likely prove conservative as the overall business shifts away from CPE-intensive video (accounted for in CapEx) and the FTTH deployment reduces overall network costs. Management has stated that CapEx will decline materially following completion of the FTTH deployment.
ATUS is currently leveraged at c.5.0x but generates $1.5bn of equity FCF even after elevated CapEx spend. Management repurchased $500m of stock in H2 2018 and is guiding for $1.5bn of repurchases in 2019, while targeting a sub-5.0x leverage ratio given EBITDA growth. I assume $1.5bn of annual repurchases over the period, representing 90-100% of annual FCF; I believe the only reason why ATUS would not deploy all equity FCF into repurchases would be if they go out and acquire one of the few remaining sizable cable assets.
Tying it all together, I think ATUS will in 2023 generate $11bn of revenue, $4.8bn of EBITDA (43% margin vs. 44% in FY 2018), and just under $2.1bn of equity FCF. Assuming a YE2022 share count of 490m, this results in Fwd. FCF / share of $4.24 in 2022. At a 10x multiple, this is a $42 stock, representing c.80% upside or 18% equity IRRs over the period. I believe there is a reasonably high likelihood that the valuation re-rates to a c.12x multiple (in-line with CHTR) as various ATUS-specific and industry concerns prove unfounded, as detailed below.
Market Perception of ATUS
I believe part of the ATUS investment opportunity is created by an analyst community that is highly skeptical of ATUS management, which is really driven by two things: (1) management has dramatically improved EBITDA margins across the business (c.30% in 2015 vs. 44% in 2018) by taking out c.$1 billion of operating costs in 3 years, which many view as overly aggressive and too high relative to peers; and (2) Altice NV (ATUS’ former parent company) has been on the financial ropes for a couple of years as a consequence of leveraging up the balance sheet to acquire telecom assets. The implication is that this is largely the same, aggressive management team and board that jeopardized Altice NV, so why trust them here?
With respect to the margin story, many analysts cite ATUS’ high EBITDA margins in the 42-45% range, compare them to Charter’s 37% margins, and conclude that ATUS must be “over-earning.” This gap is very simply explained by the fact that Charter has been undergoing an enormously expensive integration of 3 cable assets (legacy Charter, Time Warner Cable, and Bright House) which collectively comprise 50 million passings and are spread across most of the country. Additionally, Charter generates ARPUs which are c.25% lower than ATUS, which primarily reflects geographic differences (legacy Cablevision always commanded premium ARPUs relative to peers) but also slightly more viable competition in Charter’s footprint than exists in ATUS’ markets. In addition to competing with Fios in a few key markets (e.g. Manhattan), Charter also competes with AT&T fiber and Comcast’s Xfinity product (up to 1GB speeds) in the majority of its markets; this partly explains why Charter has elected to pursue a strategy that is more focused on increasing penetration than it is on increasing prices – many of its larger markets will not support price increases. There is nothing wrong with Charter’s competitive positioning, but these dynamics provide important context needed to understand the margin differential between the two businesses.
More to the point, there are two metrics we can use to evaluate whether ATUS is under-spending on its product and network: OpEx ex-programming costs / Customer Relationships (“OpEx / Relationship”), and OpEx ex-programming + CapEx / Customer Relationships (“Total Cost / Relationship”). This analysis illustrates that ATUS, contrary to consensus opinion, continues to invest productively in the business at a rate commensurate with Charter:
Opex / Relationship; Total Cost / Relationship (FY 2018):
CHTR: $49 / month; $76 / month
ATUS : $47 / month; $69 / month
Importantly, Charter is guiding to a reduction in Cable CapEx of about $1.5 billion in 2019 as it has now completed the digitization of its network, at the same time as ATUS is ramping up network CapEx (more on this below). This will result in ATUS actually having a higher Total Cost / Relationship than CHTR in 2019.
Finally, despite 56% customer penetration, ATUS has continued to grow internet subscribers at 2-3% per year under the new management team despite the AT&T fiber roll-out within the Suddenlink footprint and continued competitive response from Fios in the Optimum market. I believe the combination of the relative cost per relationship analysis and continued growth in the Company’s subscriber base sufficiently demonstrates that ATUS is not under-investing in its business. By extension, I am quite comfortable that ATUS will not suffer from any self-inflicted market share losses or margin pressure as a result of anything related to chronic under-investment.
In contrast to consensus, I believe ATUS margins are highly defensible, with a high likelihood to continue expanding modestly from here. Furthermore, the current set-up provides investors with a free option on significant incremental margin improvement as ATUS completes its FTTH initiative over the next few years, as I will shortly set out in more detail. I have assumed no margin improvement in my base case.
The second concern about management specifically is obviously related to the perception that ATUS is over-earning, which is addressed above. There are a few other ways to counter the view that management’s struggles in Europe are a likely portent of its future failure in the US. Of particular note is the fundamental differences between the European Altice NV telecom assets, where the primary issue for Altice NV is a highly saturated French mobile and broadband market where four competitors are engaged in a war of attrition with no end in sight. However, I think it is most helpful to think about what management has actually done strategically since acquiring Suddenlink and Optimum other than reducing operating expenses.
Upon acquisition in 2015, Suddenlink was already capable of delivering speeds of up to 1Gbps in 60% of its footprint, with a top speed of up to 150Mbps in 90% of its markets; management has continued to expand the 1Gbps offering across the Suddenlink footprint and has taken minimum speeds up as well. Optimum, however, was at the time going to market with minimum speeds of 10Mbps and had maximum speeds of 300Mbps. Today, Optimum typically markets with minimum speeds of 200Mbps, a current maximum speed of 400Mbps, and management is in the process of rolling out DOCSIS 3.1 across its competitive Optimum footprint over its existing hybrid Coax-Fiber network over the next year, which will permit 1Gbps download speeds. Additionally, both the Optimum and Suddenlink video products were, prior to acquisition, extremely outmoded and unable to integrate OTT applications such as Netflix. Almost as soon as management took over, they began developing a modern, integrated device which they dubbed Altice One, which now covers about 10% of Altice’s video subscriber base. While management has suggested it is still too early to tell if the product will improve churn, video subscriber losses did decline from 3.6% in FY 2017 to 2.9% in FY 2018.
Increased investments in taking up broadband speeds and improving the quality of CPE equipment is in no way unique to ATUS, but these actions are not consistent with those of a management team that is simply running the business for cash. This provides a good seque into the Altice Fiber roll-out, summarized below.
Fiber Roll-Out – a Free Option on Market Share Gains and Continued Margin Expansion
In markets where Optimum competes with Fios, its pitch is essentially that it is slightly cheaper on a price / Mbps basis and offers more comprehensive video packages than Fios. While management has taken up minimum speeds across the Optimum footprint, the reality is that Fios’ fiber product is still vastly superior in that it offers symmetrical speeds (same upload and download speeds) and maximum speeds of 1Gbps. While these features have historically not been necessities for many customers, they are being increasingly sought after in a world of extremely bandwidth-intensive activities such as OTT streaming, online gaming / streaming, and remote access to workplace networks. Additionally, being able to market a 1 Gbps fiber product is no doubt a meaningful foray into consumer mindshare - even if target customers don’t ultimately purchase a 1 Gbps offering, being able to offer it in the first place is nonetheless a helpful competitive advantage from a marketing perspective.
ATUS management is well aware not only of its existing disadvantage relative to Fios, but also of the fact that this disadvantage will widen significantly over time if it does not fiberize its footprint today. Accordingly, management is currently in the process of deploying FTTH across the entire Optimum footprint (as well as some of Suddenlink), first starting with the markets in which it competes with Fios. This will enable ATUS to go to 10Gbps symmetrical speeds, which would be industry-leading and a highly powerful marketing tool. As of early 2019, Altice is already selling FTTH in certain markets in Long Island and it expects to have the Fios footprint fiberized within the next couple of years, with the rest of the project completed by 2022.
I don’t expect that the project will be a homerun from a revenue perspective, but management has suggested that given reduced maintenance CapEx and declines in incidence rates (and attendant truck rolls / customer service calls) associated with FTTH networks, it expects a 2.5 year payback “on just being able to eliminate the OpEx and CapEx related to [the fiber build].” As detailed earlier, I believe ATUS is currently budgeting about $2.3 billion total or $450m / year for its FTTH project, which it expects to be completed over a 5-year period. I interpret management’s commentary to mean they are expecting annual OpEx and CapEx savings of about $1 billion as a result of the FTTH project, irrespective of the c.$450m annual CapEx reduction associated with simply completing the deployment.
Even if half of the $1 billion of prospective savings materialized, it would represent enormous unexpected upside for ATUS ($500m of pre-tax profit results in fully taxed incremental FCF of $375m, or $7.50 of value / share at a 10x multiple). I don’t think anyone has incorporated even modest success associated with the FTTH project into their forecasts, and neither have I. Nonetheless, it represents an extremely attractive free option at the current share price.
I believe I have largely addressed the ATUS-specific issues and broader industry concerns that I mentioned at the outset, with the exception of 5G fixed wireless acess (“FWA”).
The first thing to understand about 5G as it relates to ATUS is that of the major cable providers, ATUS is the most insulated from any 5G threat, if it exists, for one simple reason: it already competes with the lone telecom that believes there is a real business case for 5G FWA. Importantly, Verizon has stated that its ambitions in 5G FWA (which it has begun to unashamedly walk back over the past several months) lie outside of its existing Fios product, which exists across 50% of the Optimum footprint. This stands to reason. Why would Verizon cannibalize its Fios product – a fiber to the home offering with 1Gbps speeds – with a 5G FWA product that is unproven, carries significantly lower average speeds and would require additional investment?
More generally, the second thing to note about 5G FWA is that the technology, as it relates to the sorts of speeds that would be required to displace an HSD incumbent like ATUS offering 200 Mbps or higher, is only economical in extremely dense, metropolitan areas where Verizon can reach more customers with its mmWave technology (which has poor propagation and range but plenty of bandwidth to carry high speeds). Optimum already competes with Fios in these sorts of dense markets, and the remainder of Altice’s footprint across both Optimum and Suddenlink is largely in suburban or rural areas in which FWA would not be economically viable at competitive speeds.
There are other reasons why I do not believe Verizon (or anyone else) will ever scale 5G FWA at competitive speeds, but I think the above two points satisfy the 80/20 rule as the issue relates to ATUS.
Giving no credit to any operational efficiencies associated with the Company’s FTTH deployment and assuming a 10x FCF multiple on a fairly conservative projection of future FCF, ATUS offers at least 80% upside over a c.3 year hold period, generating 18% IRRs. Material upside exists in the form of a likely re-rating and / or management being even partially right with respect to its payback analysis on the FTTH project.
Progression / Completion of FTTH Deployment