TMUS - Charter: Fiber And FWA Won't Kill Cable

2023-10-30 01:10:30 ET

Summary

• Like many Covid-profiteers Charter has been a profitable short for the past two years.
• In addition, at first sight, operational trends are looking bad, while competition seems to be gaining share.
• In this article, based on a thorough analysis of operational trends, I show why longs should not worry.

Why did Charter sell off after earnings?

The market feared lackluster results from both telcos and cable companies. Hence, when subscriber numbers and free cash flows turned out to be surprisingly high for the telcos, massive rallies ensued. In contrast, broadband subscriber net adds were bad for both Comcast ( CMCSA ) and Charter ( CHTR ), resulting in huge sell-offs: Comcast lost about 15% and Charter about 18% from their respective recent peaks.

Moreover, subscriber net adds have been bad for two years now, while they had been fabulous in 2020. After adding about 1,300K subs annually for five years, in 2020 net adds at Charter jumped to 2,215K, which took the stock to record highs. (As if those numbers were the start of a sustainable trend.)

Since then, the stock is down over 50% and subscriber additions have been very disappointing: for both 2022 and 2023 the final tally will be around 300-400K.

At the same time, especially T-Mobile ( TMUS ), but also Verizon ( VZ ) are making millions of new fixed wireless subscribers, adding a combined total of around 3,000K annually. Moreover, fiber builds are expanding and AT&T ( T ) and Verizon are growing their own broadband subscriber bases.

So it looks like cable’s competition is winning on several fronts.

Last but not least, cable has entered a massive investment cycle to upgrade and expand its footprint. This weighs on FCF, reduces buybacks and makes leverage (close to 4.5x EBITDA at Charter) look even more frightening, while financing costs are skyrocketing. At the same time, the new investments still don’t yield returns.

Why shorts and longs can both be right

I always say that shorts and longs can both be right at the same time – it’s the different time horizon that makes the difference. Shorting Charter at $800, when a foolish market had bid up the stock on an unsustainable one-off boost to subscriber numbers, was relatively smart. It was quite likely that the stock would come down a bit in the near term.

Shorting it at $370, when subscriber additions have been low for several quarters, while interest rates are expected to peak soon and footprint expansion and upgrade investments have been going on for roughly the same period, is probably not so smart anymore.

To make money on a Charter short position today, you have to be right on many more things compared to 2021: For example, you have to be right on the prospects of fixed wireless as a durable alternative to cable and fiber, you have to be right on the future path of financing costs and on the success or failure of the footprint expansion and upgrade initiative.

The footprint expansion: success or desperate move?

As Charter stated on its Q3/23 call , the uptake within the expanded footprint areas is exceeding expectations:

We expect to add approximately 300,000 new subsidized rural passings in 2023 and to accelerate that pace in 2024. Our penetration gains in subsidized rural passings continue to grow at a better-than-expected pace. At the 12-month mark, our rural builds are achieving nearly 50% penetration, faster than our initial expectations.

What Charter wants us to understand is that within 12 months, subscriber adds just from the new builds will likely exceed 300K. This is what the company is likely to reach for 2023 on the entire footprint.

Now the shorts say, wait a minute: If so far Charter has added around 200K new builds and net adds from that area so far are close to 50%, since all net adds for the entire footprint in 2023 have been just around 200K, this means that net adds on the legacy footprint are really cratering.

Further down on that path, a short seller might argue that spending all those billions, increasing interest expense and leverage to add just a few thousand customers is a bad use of money.

So could the footprint expansion be just a desperate move to at least somewhat increase subscriber adds (albeit at a huge cost) and to hide the fact that the legacy business is rapidly using value?

To answer this question, we first need to look at the competitive landscape.

Is fixed wireless a durable alternative to cable broadband?

In my opinion the debate about this issue is focusing on the wrong aspects. This is not about the technology (“Can we achieve a stable high-speed internet connection wirelessly?”), but simply about economics.

Many of us can rent a private plane, fly to Paris, drink a bottle of champagne and get back safely within 24 hours. The simple fact is that for most of us, it is simply too expensive, and for almost all of us it would be stupid to do even if we could afford it.

Given the cost of 5G network deployment, it doesn’t make any sense to sell unlimited data plans to everybody. Telcos need phone lines to recover their investments. That said, they can sell some excess capacity in some areas of their footprint.

On the T-Mobile call , the company precisely described how their FWA “footprint” works:

We generally talk about marketing [FWA] to about 50 million homes right now. But it's a dynamic number and it changes based on penetration of given neighborhoods. And so what happens, I'll remind you is that on every sector of every tower, we have an assessment of capacity, not just now, but out into the future, assuming ongoing wireless smartphone share taking and ongoing rapid increase in wireless consumption per smartphone. And once we plot all of that out, there are sectors of towers where no normal amount of share taking or wireless smartphone consumption will use up our capacity anytime soon. And in those places, and only those places are we approving applicants for our home broadband service.

And what that means is, we're essentially monetizing and selling excess capacity through this initial 5G broadband strategy. And so those are the "homes passed". Now, if three people in your neighborhood sign up, or four or five people, depends on the sector, the whole neighborhood comes off our list until such time as we've got that excess capacity again.

This means that FWA is certain to enjoy a rapid, initial net add boost, but it is also a certainty that net adds will slow down massively once the target capacity is reached. Moreover, as T-Mobile pointed out, marketing of this excess capacity is somewhat uneconomical, since you can market the service only to a very limited number of customers. Usually, when ten of your neighbors have subscribed to a service, another ten are waiting in line. In the case of FWA, when word-of-mouth has spread, it might already be too late to subscribe.

So cable is right that this FWA issue is likely mostly transitory.

In addition, FWA might not even be competing with cable directly, since most of its subscribers have been located in rural areas with no alternative broadband provider. So not every FWA sub is a sub lost to cable. Some will come back in the future and the rapid uptrend is certain to grind to a halt.

In the meantime, cable itself is expanding into rural areas and apparently gaining subs rapidly.

Is Charter losing the battle with fiber and fixed wireless?

This is the question many investors have on their minds, yet we first need to ask ourselves whether Charter is competing with fiber and fixed wireless at all .

Many investors in the telco space forget that this is a local business. There are many (relatively unknown) excellent telco businesses with just a few thousand subs. It all depends on the exact footprint and the competition therein.

T-Mobile’s FWA subs may be really happy with their wireless broadband, but the fact is that quite often they don’t have many alternatives to choose from. Charter’s subscribers may be unhappy with their hated cable provider, but the sad fact is that it may still be their best available choice.

The best telco businesses are those which are not available everywhere. This is because they can choose where to invest depending on potential returns.

Historically, Charter has squeezed a growing return out of every passing in its footprint. In 2016 the company had built 48.8m internet passings and realized revenues of $40B, i.e. $819 per passing. Fast forward to 2020, passings where 53.4m and revenues had grown to $48.1B, thus $900 per passing. In Q3/2023 (i.e. two “bad” years later), passings are 56.6m and annualized revenues $54.4B, or $961 per passing.

At the same time, operating expenses per passing have grown from $525 in 2016 to $575 in 2023, i.e. operating income per passing has enjoyed considerable operating leverage, going from $294 in 2016 to $386.

This is also because internet footprint penetration has increased from 46% in 2016 to 54% in 2023.

Now, the new footprint has a cost of $3,800 per passing (as stated by Charter on the recent call). Penetration is likely to end up right around the average for the entire footprint, so we should expect the company to make at least average operating income per passing on the expansion area for a 10% yield on investment before interest and taxes.

Most importantly, these are very durable earnings that will almost certainly grow from here. While the sell-side has calculated an effective value per new passing around $9,000 (which we might argue with), the ROI seems certainly sufficient and doesn’t make Charter look like a failing business. This money is well spent.

? And what about the legacy footprint? Isn’t this legacy business losing share?

First of all, even if we look at the evolution of EBIT per passing in the past 11 quarters only, it has still grown from $368 in Q1/21 to $386 in Q3/23. This is while FWA made millions of new subscribers.

– Sure, you might say, but how sustainable is this? You can’t hike prices ad infinitum.

– Right, but Charter actually almost didn’t hike its prices at all: ARPU grew 3% for residential customers and decreased 1% for SMBs, which is far below inflation, GDP growth and also below EBIT/passing growth.

– And what about market share? – This is largely a fake argument in my opinion. This is not like Coke or Pepsi. Not every customer can choose between all competitors.

Between 2014 and 2019 Comcast and Charter combined added 2,500K new subscribers annually at a very regular pace. There were no noteworthy fixed wireless or fiber subscriber numbers for competitors, so the entire broadband market switching pool was 2,500K.

As soon as fixed wireless entered the scene, the switching pool grew by 1,000K and, at least in the past two years, FWA took over 80% of this switching pool, i.e. of the 3,500K around 2,800K went to FWA. This certainly looks like a massive loss of share.

However, it is extremely likely that the 1,000K additional “switchers” actually were no switchers at all: They would not have subscribed to cable. They represent an expansion of the accessible market thanks to a new technology at affordable prices.

Yet cable apparently still lost about 1,800K subscribers annually to FWA. Many of these would actually have subscribed to cable absent the FWA offer, but the FWA offer is not permanent . In any case these are one-time losses (and some might come back later).

I also guess that most of those that were tempted to purchase a cable subscription, did so in 2020, thus somewhat reducing the available customer pool for new subscribers in the years immediately afterwards. So not all these 1,800K subs represent actual market share losses.

We will have more to say about this below.

Is fiber overbuild a concern for cable broadband providers?

So far, fiber subscriber net adds at AT&T and Verizon have done little more than compensated their legacy broadband losses. Basically, as DSL becomes too outdated, the telcos are switching those customers that become too unsatisfied with that unreliable technology to a more modern alternative. The net subscriber gains in broadband for AT&T and Verizon combined over the past 5 years are below 400K in total.

Again, we need to look at the economics. Given the huge costs of fiber builds, it is highly unlikely that a fiber internet access can be cheaper for consumers than cable, which can be updated to very competitive speeds for as little as $100 per passing.

Hence, overbuilds certainly add to competition, but cable remains competitive.

The elephant in the room

What the cable bears don’t talk about much is the huge success cable is making with its wireless service. Both Comcast and Charter are piggybacking on Verizon's network investments and can offer wireless plans for extremely competitive prices.

In the past three years (the “bad” years for cable), Comcast and Charter combined have taken about 26% of all new mobile lines created. In 2023 this share should come in around 35%, so it is still growing.

Speaking about market share, when looking at all net adds (internet + phone) combined for the entire market, it looks like cable took about 30% of them over the past three years.

While noteworthy earnings from this success have still to materialize, cable is certainly not losing. As I have pointed out in a recent article on AT&T , there is a “land rush” going on in the telco space, as customers increasingly prefer a single provider for all their connectivity needs. You cannot win the game if you offer only broadband and your client needs to buy a different plan from another provider for texts and calls. You need to offer all services as a bundle.

T-Mobile is actually aware of this and is already preparing for the time when it will have run out of excess capacity to sell for FWA by exploring fiber broadband partnerships. AT&T and Verizon are both expanding their fiber footprints.

However, the only connectivity providers that already today offer a future-proof complete bundle are the cable companies. And they can do it in a much more capital efficient way than the traditional carriers, including the “un-carrier”.

Basically, cable already has all networks in place, while the carriers have still to complete both their 5G networks and the fiber footprint.

Actually, the fact that the carriers are embarking on a multi-billion spending spree to build dozens of millions of fiber passings might be considered a “desperate move”, since without this investment they would be certain to lose share over time.

Has Charter too much leverage?

It certainly looks high. Charter has $97.6B of net debt, which equals 4.5x adj. EBITDA. This debt carries an average cost of 5.3% and 84% of all debt is fixed. The remaining weighted average life of Charter’s debt is almost 13 years. This means that the company won’t suffer much because of interest rate changes for the next few years. Only 10% of its debt matures within two years.

Assuming upon its renewal interest increases 50% on this 10% of debt, Charter will have to pay about $260m more interest in 2026 compared to today, which looks perfectly manageable.

As long as earnings grow – and I guess I have shown why we should expect that to happen – Charter will be able to service its debt.

Is Charter cheap?

On an EV/EBITDA basis, Charter trades for 7.3x, which is far below its usual trading range around 9-10x. At 9x EBITDA, the stock would stand around $600.

Looking at FCF, we have to consider that there is a massive upgrade and expansion cycle ongoing. Excluding these growth expenditures, Charter should have a run-rate of maintenance capex of around $6.5B, which puts its steady-state FCF at about $8.5B. So the equity trades for 7.4x steady-state FCF.

I think this is cheap enough to purchase a starter position.

For further details see: