Business
The Business: A Cash Machine Under Assault, Wrapped in Leverage
Charter sells one thing that matters and three that are fading. The one that matters is broadband connectivity — the high-speed internet (now bundled with mobile) that runs into ~58 million homes and businesses across 41 states under the Spectrum brand [1]. It is one of the highest-return infrastructure businesses in America: spend billions once to bury a network past a home, then collect a high-margin recurring bill for decades. That is the part to fall in love with.
The trouble is that the moment of maximum competitive attack and the moment of maximum capital spending have arrived together — and the equity sits on top of $94.6 billion of debt at 4.15× EBITDA [2]. When EBITDA is flat and the balance sheet is geared 4–5×, the equity is a thin, volatile sliver. That is exactly what the market has done to it: the stock fell from roughly $209 in early January 2026 to $126 by mid-June 2026 — and it earns about $36 a share.
The investment question is not "is this a good business?" — at the asset level it plainly is. It is "what is a high-return but no-growth, secularly-pressured, 4–5× levered cash engine worth, and will broadband units stabilize before leverage does the damage?" Value it on EV/EBITDA and free-cash-flow-per-share through the capex peak — never on the optically tiny P/E.
Revenue FY2025 ($M)
Adjusted EBITDA ($M)
Free Cash Flow ($M)
Diluted EPS ($)
Return on Equity (%)
Net Debt / EBITDA (x)
Sources: revenue and Adjusted EBITDA — FY2025 10-K, Item 7 MD&A [3][4]; leverage — FY2025 10-K [2]; EPS and ROE derived from reported financials.
1. The economic engine: a sunk-cost network rented by the month
The reason cable broadband is a great business is structural, not managerial. The hard, expensive thing — a hybrid fiber-coax network passing tens of millions of homes — is already built and largely paid for. Once a home is passed, the cost of connecting and serving one more customer is small relative to the ~$120/month they pay, so incremental subscribers and incremental products drop almost straight to cash flow. Management's own strategy is explicit about the flywheel: sell more products to each relationship, which lowers churn and the cost to acquire and serve, which raises profitability [1].
The result is a business with a ~24% operating margin and a ~42% Adjusted-EBITDA margin on $54.8 billion of revenue [3][4]. But "margin" understates how good the asset is and "ROA" understates it too — the right frame is cash return on the incremental dollar of plant. The catch, which Section 6 develops, is that this is a no-growth cash machine: revenue was actually down 0.6% in 2025 [3].
Sources: revenue and Adjusted EBITDA, FY2025 10-K MD&A [3][4]; capex and free cash flow derived from reported cash-flow statements.
The ladder shows the whole tension in one picture: a $22.7 billion EBITDA engine is currently surrendering more than half of that to capital expenditure (Section 6), so the cash that actually reaches owners is $4.4 billion — and the rest of the EBITDA is spoken for by interest on $94.6 billion of debt.
2. The mix: where the money comes from — and where it leaks
Charter reports four residential products plus commercial and advertising. The single most important fact about the mix is that "connectivity" (internet + mobile) is growing while video and voice bleed out — and connectivity is the high-margin product, so the company is, slowly, getting better even as the top line stalls.
Source: FY2025 10-K, Item 7 MD&A, "Revenues by service offering" [3].
In 2025, internet rose 1.7% and mobile service jumped 22.0%, lifting combined connectivity revenue 4.1% to $27.5 billion; meanwhile video fell 9.4%, voice fell 6.0%, and advertising fell 17.6% (a political-cycle effect) [3]. The industry primer covers this mix shift in depth; the investor's takeaway for quality is this: the shrinking lines (video, voice) are the low-margin, high-cost-of-goods lines — video in particular is a near-pass-through of programming fees — so losing them hurts revenue far more than it hurts profit. The dangerous line to lose is internet, because that is the profit. Which brings us to the moat.
3. The moat: real, wide, and visibly narrowing
For two decades cable broadband enjoyed a near-monopoly over a large share of its footprint: the only wire into the house capable of high speeds. That moat is a sunk-cost barrier to entry — a rival must spend billions to build a competing network to take a customer worth ~$1,400 a year. It is real, and it still produces monopoly-like economics across much of the territory. But it is being attacked from two directions at once, and the attack is now visible in the unit numbers.
Attack 1 — fiber overbuild. AT&T and Verizon are building fiber-to-the-home that matches or beats cable speeds. Charter discloses that it already faces terrestrial broadband competition from AT&T in ~27% and Verizon in ~16% of its footprint, and that figure climbs every year [5].
Attack 2 — fixed wireless. T-Mobile and Verizon now sell home internet over their 5G networks at low prices, peeling off the price-sensitive and lighter-usage end of the market. Management concedes the obvious — "the introduction of fixed wireless access has impact on everyone's penetration" — while arguing it competes on usage and reliability [20].
The scoreboard tells the story better than any argument: residential internet customers peaked at 28.5 million in 2023 and have declined every year since, and the company lost another 120,000 internet customers (residential plus small business) in Q1 2026 alone [7].
Source: residential internet, video, and mobile-line counts — FY2025 10-K Item 1 Business [8]; prior years derived from FY2018–FY2024 10-K KPI disclosures, as reported.
Read those three lines carefully, because they are the thesis. Red (internet) is the profit engine, and it has rolled over. Grey (video) is the managed decline — losing it is fine. Cyan (mobile) is the new growth leg, and it is going vertical. The bull case is that cyan and ARPU outrun red's decline; the bear case is that red keeps falling and the moat was never as wide as the multiple once implied.
4. Mobile: the growth engine that is also a hedge — but a thin-margin one
Mobile is the most important strategic story and the most misunderstood economically. Spectrum Mobile is an MVNO — it resells service over Verizon's cellular network rather than owning spectrum and towers [9]. That is the key to its economics: low capital intensity, but also structurally thinner margin than internet, because Charter pays Verizon for the wholesale network. So mobile's value is less about the standalone profit per line and more about bundling — a customer who takes internet and mobile churns far less, which protects the high-margin broadband relationship.
And it is working as a volume story. Charter passed over 12 million mobile lines in Q1 2026, adding 370,000 in the quarter and 1.8 million over the trailing year — growth of over 17% — and management calls Spectrum Mobile "the fastest-growing mobile provider in our footprint" [10]. From a standing start in 2018, that is one of the fastest mobile ramps in US history — built with almost no network capex because it rides Verizon's radios.
The honest read: mobile is a defensive growth engine. It adds revenue and, more importantly, lowers broadband churn — but at a wholesale-MVNO margin, it does not replace lost internet profit dollar-for-dollar. It buys time and stickiness, not a new high-margin business.
5. Capital intensity: the cyclical peak that is eating free cash flow
Here is the second half of why the stock is where it is. Charter is spending at a cyclical capex peak for two reasons at once: it is upgrading the entire network to symmetrical multi-gigabit speeds (DOCSIS evolution — management expects ~50% of the network upgraded by end-2026) [6], and it is building into rural America, having spent $7.7 billion on subsidized rural construction since 2022 to activate ~1.3 million new passings [11].
The effect on owner cash has been brutal. Capex climbed from $7.6 billion in 2021 to $11.7 billion in 2025 (~21% of revenue) while free cash flow halved from a $8.6 billion peak to $4.4 billion.
Source: capital expenditure and free cash flow, FY2020–FY2025 consolidated cash-flow statements, as reported; FY2025 free-cash-flow walk in 10-K MD&A [12].
This matters enormously for valuation. If the network upgrade and rural build are genuinely temporary, then FCF is artificially depressed and should rebound sharply toward $8–10 billion as capex normalizes — making today's equity very cheap. If broadband units keep falling, the "growth capex" turns out to be "maintenance capex to stand still," and the rebound never comes. The bull and bear cases live almost entirely in this one chart. FY2025 already showed the first crack of relief — free cash flow rose $747 million as cash taxes and interest fell [12].
6. Capital allocation: the share-shrink machine, running on leverage
Charter has no dividend [13]. Its entire equity story for a decade has been the classic Liberty/Malone playbook: lever the cash machine to a target band and use every spare dollar to retire stock. It has worked spectacularly on a per-share basis. Net income has been roughly flat at ~$5 billion since 2020, yet diluted EPS more than doubled because the share count was cut nearly in half.
Source: shares outstanding and diluted EPS, FY2018–FY2025 income statements, as reported; FY2025 buyback detail in 10-K Note 11 [14].
In 2025 alone Charter bought back 16.1 million shares for about $5.0 billion [14] — and at a far lower price than the buybacks of 2021–2022, which is the optimistic spin. The buyback is funded not only by FCF but by debt capacity: the company runs a deliberate 4.0–4.5× net-debt-to-EBITDA target and borrows to keep leverage in the band as EBITDA grows [2].
This is the double-edged sword. Leverage is why ROE is 31% on a ~9% return on capital — financial engineering amplifies the asset's return. It is also why the equity is so fragile: with $94.6 billion of debt structured as $11.9 billion of credit facilities, $55.4 billion of investment-grade secured notes and $27.3 billion of high-yield unsecured notes [15], a few points of EBITDA erosion swing the equity value violently. Buying back stock while EBITDA is flat and units fall is a bet that the units stabilize — if they do, the share-shrink at today's depressed price is enormously accretive; if they don't, it is borrowing to buy a melting asset.
7. The two deals that remake the company: Cox and Liberty Broadband
Two transformational transactions are pending and they change the scale, the structure, and the leverage.
Cox Transactions (announced May 16, 2025). Charter is combining with Cox Communications, the third-largest US cable operator (a privately held residential cable plus commercial-fiber and managed-IT business). Charter pays $4.0 billion of cash and assumes approximately $12.6 billion of Cox net debt and finance leases [16], with Cox Enterprises taking a large equity/partnership stake. Post-deal, Spectrum would reach over 70 million households [6]. Strategically the logic is sound: Cox's footprint has low mobile and video penetration, exactly the gap Charter has filled before — management explicitly compares the integration opportunity to its successful Time Warner Cable acquisition [7]. Charter plans to lower its long-term leverage target to 3.5–3.75× after closing [2].
Liberty Broadband Combination (announced Nov 12, 2024). Charter is absorbing Liberty Broadband, whose principal asset is ~41.5 million Charter shares [17]. This is essentially a structural clean-up — retiring the Malone holding-company layer — expected to close alongside the Cox deal.
Governance is a permanent feature, not a bug to wish away. After the deals, both Cox Enterprises and the Advance/Newhouse Partnership (A/N) hold large equity and voting stakes with rights to influence major corporate actions and preemptive rights on new equity [18]. Minority public shareholders ride alongside controlling insiders — a structure that can align long-term interests but limits outside influence.
8. Cyclicality and the real risks
This is not a cyclical business in the commodity sense — broadband demand is secular and recession-resilient; people keep the internet on before almost anything else. The cyclicality that matters here is different and more dangerous: a secular/competitive transition layered on a financial cycle (the capex peak and the debt load). The genuine risks, in order:
1. Terminal broadband decline. If fiber and fixed wireless keep taking units, the high-margin profit engine shrinks structurally — and no amount of mobile or cost-cutting offsets a permanently falling internet base. This is the risk; everything else is secondary.
2. Leverage in a flat-EBITDA world. At 4.15× with $94.6 billion of debt, refinancing the high-yield tranche at higher rates, or an EBITDA stumble, hits equity hard. Management's own framing is that 2026 EBITDA growth "has faced challenges" from broadband declines [20].
3. Deal execution. Integrating Cox while defending the core and completing the capex cycle is a lot at once; the 10-K's own risk factors warn the market price "may decline as a result of the Cox Transactions."
The mitigants are equally real: management is repricing the base (roughly 45% of residential customers are already on the new converged pricing and packaging [19]), capex is set to roll off as the upgrade and rural build complete, and the network upgrade should make cable competitive on speed with fiber for a fraction of fiber's build cost.
9. How to value it: ignore the P/E, underwrite EV/EBITDA and FCF/share
The most common mistake on Charter is to see a ~3.5× P/E and call it absurdly cheap. The P/E is meaningless here because the equity is a small, levered residual on a large enterprise. The right lens is enterprise value.
Source: derived from reported financials at a $126.23 share price (June 18, 2026) and ~137.7M shares; net debt and Adjusted EBITDA per FY2025 10-K [2][4].
At ~$126, Charter trades around 4.9× EV/EBITDA — a multiple that prices in genuine terminal-decline fear, well below where stable infrastructure assets change hands. The equity offers a ~25% FCF yield, but on FCF that is artificially depressed by the capex peak; normalize capex and the yield is even more striking — which is precisely why the bulls are loud. The bear retort is equally clean: at 4–5× leverage, "cheap on EBITDA" is only cheap if EBITDA holds, and EBITDA holds only if broadband units stop falling.
Where the genuine peers sit. Charter's true comparison set is small — the other facilities-based broadband/cable operators and the telcos attacking them. Comcast (Xfinity) is the near-identical cable twin; AT&T, Verizon and T-Mobile are the fiber/fixed-wireless attackers; Cable One and the former Altice USA (Optimum) are the smaller, more-levered MSOs whose distress shows what too much leverage plus competition does to a cable equity.
Sources: peer equity market caps from run-staged competitor snapshots (as reported, date unavailable); business-model confirmation from each peer's own FY2025 10-K and Charter's competition disclosure [5]; CHTR market cap and EV/EBITDA derived at the June 2026 price.
The peer table carries a warning as much as a comparison: Cable One and Optimum show the failure mode. Both are facilities-based cable operators with the same economics as Charter but more leverage and weaker footprints, and their equities have been crushed as competition met debt. Charter is bigger, better-capitalized, and investment-grade on most of its debt — but it is on the same spectrum, and the market is now pricing it closer to that end than to a serene infrastructure compounder.
The verdict. At the asset level this is a high-quality, wide-moat, cash-generative infrastructure business — but it is a no-growth one under real secular and competitive assault, geared 4–5×, spending at a capex peak. That combination is why a business earning $36 a share trades at $126. It is genuinely cheap on EV/EBITDA and normalized FCF/share if broadband units stabilize and the Cox deal deleverages as planned; it is a value trap if internet keeps bleeding. Underwrite the unit trajectory and the capex roll-off, not the headline multiple — that is the whole investment.