Please refer to our disclaimers, which can be found in the footnote of this page and here.
Update Only
We have owned Charter Communications (“Charter”) in various sizes since 2020. However, I first invested in the company in early 2016 on a personal basis, pre-dating our inception. The stock is down circa 40% since the 3Q 2023. As for Alibaba, our commentary here is intended as an update only, not a comprehensive investment memo.
For those unfamiliar, Charter is predominantly a cable broadband business. Broadband accounts for just under 70% of net revenue[1] and over 80% of EBITDA. We like the business because of its stable, recurring revenue and entrenched oligopoly/monopoly position. We estimate that just over 50% of its footprint is a monopoly, with no other competition capable of > 1 Gbps speeds[2]. Its core moat lies in its hard-to-replicate assets and broad ownership of customer relationships. The returns available for a new entrant overbuilding in cable’s core markets are typically too low to make it attractive. There is also an interesting opportunity for Charter in offering mobile services via its network agreement with Verizon combined with offloading traffic to its own infrastructure. This enables Charter to offer mobile services at a significant cost advantage relative to the mobile network operators.
While we like the business, there are some structural things not to like. First, its core moat relies on economic rationality, which in our experience is not always reliable (sometimes market participants do irrational things!). Second, the business is capital intensive. The network infrastructure requires upgrades to maintain its competitive position versus fibre. At various stages over the years, the investment community champions the stock on the basis that once the current network upgrade ends, cash flow and ROIC will improve. However, once an upgrade cycle ends, there is usually another announced. Third, the business has high leverage, though we are comfortable with the maturity profile, fixed-rate nature, and the serviceability of the debt. Management also has flexibility to reduce spending on rural initiatives or share repurchases if required. As a result, we do not think the business quality is as high as other businesses in our portfolio, and it therefore carries a higher cost of capital for us. We also size it more conservatively despite its more attractive valuation.
The recent underperformance predominantly relates to its decelerating residential broadband growth, which has been the result of fixed wireless offerings from T-Mobile, Verizon, and now AT&T (plus some impact from the ending of the government Affordable Connectivity Program). Fixed wireless is a service whereby the customer’s home connection is direct to the mobile network utilising mid-band 5G spectrum. T-Mobile has the most attractive mid-band spectrum for this. While its impact is recent, the technology is nothing new. In fact, I was a fixed wireless customer in London some years ago on 4G. As can be seen in the table below, fixed wireless has been taking most of the incremental net customer additions of late given its lower price point and positive early customer reviews. T-Mobile expects to reach 7-8m fixed wireless subscribers over the next two years, Verizon 5m, and AT&T has not released a target. This represents 9-11% of addressable households in the US and about 700k adds per quarter.
Broadband Net Subscriber Additions
We have spent considerable time with network engineers and business managers from the fixed wireless teams to understand the path forward. The consistent opinion was that the stated targets from T-Mobile and Verizon are fair, but that it would also be hard for them to go much beyond this range. Fixed wireless customers use considerably more data than mobile customers and so the revenue per gig of data, and therefore customer unit profitability, is an order of magnitude lower. In addition, the capacity of mobile spectrum is limited, which means most of the spectrum must be reserved for mobile customers. The strategy of the networks is to sell only spare capacity on a tower-by-tower basis. This also means customer acquisition is complicated since it is highly localised and customers may not be able to transfer their service when they move (moving equals churn). Further, fixed wireless customers rank below mobile customers and business customers when the network is congested and are the first to face speed-limiting. We have also heard that the fixed wireless teams must fight with the mobile teams for capacity allocations. As the spare capacity is used up, the positive experience of the fixed wireless customers is likely to degrade. Indeed, this was my personal experience after being a 4G fixed wireless early adopter (I churned back to wired broadband). Our understanding from these conversations is that it would require very significant capital expenditure and acquisition of new spectrum to go much beyond the stated subscriber targets, which is not feasible unless it would also lead to significant mobile subscriber acquisition (unlikely given already mature mobile market penetration). Our view is that fixed wireless will continue to gain share but will plateau over the next 24 months, allowing cable to return to a more normalised path. This has also been signalled by the mobile networks themselves.
Longer-term, we remain more cautious of fibre competition than fixed wireless competition. Charter has a very credible network upgrade path to remain competitive with fibre on a symmetrical (download/upload) basis. However, fibre is still a superior asset. It has fewer outages, has a lower maintenance cost, and the back-end is easier to manage. New home passings by Charter are done with fibre (as is the case for their government subsidised rural build programs). Over the ultra-long-term (say 30 years), we think the cable companies will gradually replace their network with fibre.
Charter’s competitive overlap with fibre has risen from 25% of its footprint 5 years ago to circa 48% today, and we forecast this to rise to roughly 65% over the next 5 years. This has a much bigger impact in the medium-term than fixed wireless in our view. We conservatively assume in our base case that Charter declines to 30% penetration where fibre is present versus an increase to 65% penetration in non-fibre areas. Combined with Charter’s rural expansion initiative (not a topic in this letter), this implies broadband subscriber growth over the next several years will be relatively flat. We may prove to be too conservative here in Charter’s ability to maintain share in its competitive footprint. The end of zero interest rates has made the economics of new fibre projects even harder, which should help cool the recent fibre building.
On a more positive note, Charter has been growing its mobile subscribers more strongly than we anticipated and finished 2023 with 7.8m mobile lines (representing 47% growth). We think this is a great opportunity for Charter given the aforementioned cost advantage it enjoys. However, we are also cautious here. Management reports a strong 20-30% margin before customer acquisition costs, but there is a sleight of hand in these numbers. The company records a higher revenue with an offsetting 1st-year discount in the marketing line, which is excluded from their margin number (i.e. it is inflated). In addition, because the mobile business does not have high switching costs, the customers need to be reacquired every couple of years, whether through renewed line discounts or handset subsidies. As such, the marketing expense becomes a recurring cost item rather than a one-time customer acquisition cost and should be included in the contribution margin. We estimate a fully loaded EBITDA margin of -7% today, which should expand as the business scales, but remain significantly lower than that of the broadband business. Overall, we still see this as a strong positive for Charter and it is the strongest element of growth in the business. Although the margin is less attractive, it helps reduce churn in the broadband business and to grow overall FCF per subscriber, which is the real key to value creation. Because the mobile business is lower in quality than the old video business (not the current video business in decline), the end state of the mobile+broadband business is lower in quality than the old cable business of video+broadband.
With the picture painted above, how can the returns be achieved? Our conservative base case has flat revenue over next 5 years. Broadband has a low single-digits revenue CAGR [3] (slight volume decline with low single-digit pricing growth), while mobile is growing strongly, offset by continued video decline. We think the overall margin compresses as the lower margin mobile business expands. The key lies in falling capital expenditure. As mentioned above, we think the current network upgrade expenditure is renewed with yet another upgrade announcement and is therefore more recurring in nature. However, the capital expenditure load over the next few years is significantly elevated due to their rural expansion initiative. For brevity, we won’t go into the details, but early signs indicate that Charter is executing well here. This component of capital expenditure will only remain recurring if there is continued success and value creation (a good thing). This expenditure is growth orientated. If we normalise for this, run-rate FCF can expand meaningfully. Combined with share repurchase, our base case has FCF compounding in the mid-20s over a 5-year horizon. The business currently trades on a ~7.5% FCF yield (also on a depressed forward-FCF), which gives us a mid-20s IRR[4] without any compression in the yield (i.e. without multiple expansion).
Footnotes
[1] Revenue adjusted for pass through programming costs
[2] This was reported as 56% in 3Q 2022
[3] Compounded annual growth rate
[4] Refer to our disclaimers, we make no guarantees of future performance, and our projections should not be relied upon