Charter Communications, Inc. (NASDAQ: CHTR) — The Cash Machine the Market Left for Dead

Date: June 7, 2026 Price reference: $132.12 (June 5, 2026) | 52-week range $127.88–$422.29 | Market cap ~$20.7B | Enterprise value ~$117B Sector: Communication Services — Cable / Broadband & Pay-TV Fiscal year: December 31 | CIK: 0001091667 | Auditor: KPMG

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice and not a recommendation to buy or sell any security. The analysis that follows it takes no position and carries no price target; the only stated view, and the only price levels, appear in this clearly-labeled block.

Verdict: BUY / accumulate-on-weakness — a high-conviction-that-it’s-cheap, medium-conviction-that-it-compounds deep-value situation. Fair-value zone ~$240–$340 (roughly 4–5.5x a normalized ~$60/sh post-Cox 2028 free cash flow, or a re-rate toward peers’ EV/EBITDA on a merely-stable EBITDA base). Attractive accumulation below ~$150; bear-case downside ~$80–100 if broadband EBITDA enters genuine secular decline. Not a position to size as a “safe compounder” — size it as an asymmetric, leveraged bet on broadband stabilization.

Tag: “The cash machine the market left for dead.”

The market is pricing Charter as a melting ice cube — 5.3x EV/EBITDA, ~3.5x earnings, the cheapest decile in its own history, Berkshire heading for the exit. The bear story is real: broadband subscribers are falling (−120K in Q1’26, four straight quarters of losses), fixed-wireless and fiber are permanent new competitors, revenue has rolled over, and $94B of debt sits on a sliver of ~$21B equity. But the price embeds something stronger than “no growth” — it embeds perpetual EBITDA decline of ~3%/year forever, and the data doesn’t yet support that: EBITDA still grew +0.6% in 2025, residential connectivity revenue is still rising, video losses are shrinking sharply, mobile is compounding at 17%, and the company remains the low-cost high-speed provider across most of its footprint. What the market is mispricing is the free-cash-flow inflection: a near-certain capex cliff (from ~$11.7B in 2025 to <$8B by 2028) mechanically roughly doubles free cash flow regardless of whether EBITDA grows a dollar — taking FCF/share toward ~$60 post-Cox against a $132 price. You are being handed a >25% forward FCF yield on a business that has to merely not collapse to be worth multiples of today’s price.

This is a contrarian/value call, not a quality-compounder call — I want to be honest about the framing. The moat (local sunk-cost network scale) is real but eroding, the industry is structurally bad, and the levered-equity model runs in reverse if broadband EBITDA truly melts (Altice USA, down ~96%, is the cautionary tombstone). Management also bought back $963M of stock at $225 that’s now $132 — the model has no circuit-breaker for “structurally, not cyclically, cheap.” But the asymmetry is too large to ignore: a near-mechanical FCF doubling, a decade-trough multiple, disciplined Cox consolidation at ~6.4x (paid in paper struck above market), insider open-market buying by the CEO and three directors into the drawdown, and a deleveraging path that transfers enterprise value from creditors to equity. Conviction: medium-high. Flips bullish if broadband net adds stabilize toward flat for two consecutive quarters (proof the melt is ACP/cyclical, not secular). Flips bearish if broadband ARPU pricing power cracks under FWA pressure while EBITDA turns negative — at which point 4.6x interest coverage on a shrinking asset becomes the story.

1. Executive Summary

Charter Communications is the second-largest US cable operator, serving ~31–32 million customer relationships under the Spectrum brand across a ~57-million-passing footprint, generating $54.8B of revenue (FY2025) and ~$22.7B of adjusted EBITDA at a ~41.5% margin. It is, at its core, a sunk-cost broadband network monetized through internet, a fast-growing mobile MVNO, a structurally declining video/voice business, and a commercial arm — financed by ~$94.6B of debt held at 4.15x net leverage with no dividend, where essentially all discretionary cash has historically been routed into share repurchase.

The investment situation is defined by a violent dislocation: the stock has fallen ~69% from a $422 high to $132, compressing to 5.3x EV/EBITDA and the cheapest decile of its own decade-long valuation history on P/E, P/B and P/S. The cause is a genuine structural shift — for the first time, cable faces two credible broadband substitutes (fixed-wireless access from the mobile carriers and fiber overbuild from the telcos), and Charter has now posted four consecutive quarters of internet subscriber losses (−120K in Q1’26). Revenue peaked in FY2024 and declined in FY2025; the local-scale moat that made cable a quasi-monopoly is eroding home-by-home as competitors sink the capital that once created Charter’s cost-to-serve asymmetry.

Against this sits an unusually quantifiable bull lever. Charter’s free cash flow is set to inflect not because EBITDA must grow, but because capex is rolling off a multi-year network-evolution and rural-buildout peak — from ~$11.7B (FY25) toward a management-guided <$8B run-rate by 2028. Holding EBITDA merely flat, that ~$3.7B reduction flows almost entirely to free cash flow, taking FCF from ~$5.0B (FY25) toward ~$11–13B and lifting FCF/share toward ~$60 on the post-Cox as-exchanged share count of ~179M. Layered on top: the pending Cox Communications acquisition (~$34.5B EV, ~6.4x EBITDA, ≥$800M synergies) and Liberty Broadband buy-in reshape the entity into a ~38-million-subscriber footprint, paid largely in equity-linked paper struck above the current price, while clearing a long-standing holding-company overhang.

The central debate is therefore not whether free cash flow rises (it almost certainly will, mechanically) but whether the rising cash is captured by equity holders or bled away by EBITDA decline and absorbed by the $94B+ debt stack — the Altice path. The market’s 5.3x multiple is a vote that broadband EBITDA enters permanent decline. The operating data so far shows flatness, not collapse, and a credible mobile-convergence offset that is measurably reducing churn. This report takes no position and sets no price target; it lays out the evidence on both sides and the single variable — terminal broadband EBITDA direction — on which the outcome turns.

Key facts: Revenue $54.77B (FY25, −0.6%); Adjusted EBITDA $22.71B (+0.6%, 41.5% margin); FCF $5.0B (FY25); net income to Charter $4.99B; ROIC ~8.6% / ROE ~31% (a leverage artifact); net debt 4.15x EBITDA at 5.2% weighted cost; Class A shares 126.6M (12/31/25), ~142M economic standalone, ~179M pro-forma Cox; no dividend.

2. Business Overview

Charter Communications operates a regional, wired broadband-communications network under the Spectrum brand, serving residential and commercial customers across 41 states. At its foundation is hybrid fiber-coaxial (HFC) plant passing approximately 57 million homes and businesses — physical infrastructure that has been built and largely depreciated over decades, and which is the source of both the company’s economics and its strategic vulnerability. Charter is the product of a 2016 roll-up in which “Legacy Charter” (5 million customers in 2013) acquired Time Warner Cable and Bright House Networks to become the second-largest US cable operator behind Comcast.

How it makes money. Charter sells connectivity and content access over its network. The revenue base divides into four streams, whose diverging trajectories are the entire story of the business:

The crucial structural point is that the headline −0.6% masks two opposing currents: a growing connectivity core (+4.1%) — residential internet revenue rose 1.7% even as Charter lost 393K internet customers, because rate and product-mix more than offset volume — netted against a shrinking legacy video/voice business. Roughly $322M of the video decline is an optical “seamless entertainment” reallocation (bundling Disney+, Max, Peacock and other streaming apps into video packages, which nets revenue against roughly matching cost), and the advertising drop is biennial political-ad cyclicality rather than structural impairment. The real economic engine — selling internet access and mobile lines to households — is still growing.

Operating footprint and KPIs (12/31/2025): ~29.7M total internet customers (residential 27.6M), 12.6M video customers, 6.0M voice customers, and 11.8M mobile lines (+1.9M / +19% in the year) — the standout growth metric. Charter serves these through ~91,900 employees, with a deliberate strategic emphasis on a 100% US-based workforce for sales and service (a differentiator management leans on heavily versus offshored competitors). The business is overwhelmingly recurring: subscription revenue billed monthly, with very low (and improving) non-pay churn, which lends the cash flows utility-like predictability — the quality that makes the FCF thesis credible even as units decline.

Spectrum Mobile deserves separate mention as the strategic pivot. It is a mobile virtual network operator (MVNO) riding Verizon’s network wholesale, supplemented by Charter’s own CBRS spectrum and Wi-Fi offload. It allows Charter to bundle mobile with broadband at aggressive prices (the “$1,000 savings guarantee” launched February 2026), which both adds high-margin, near-zero-incremental-capital revenue and — more importantly — lowers broadband churn by deepening the customer relationship. Two-product penetration reached 35.2% (from 33.4% YoY). Convergence is the one area where cable is on offense, and cable MVNOs collectively have been taking the plurality of US postpaid phone net adds from the facilities-based carriers.

Verdict: A high-quality, recurring-revenue connectivity utility with a genuine growth engine (connectivity +4.1%, mobile +22%) hidden inside a declining legacy shell (video/voice). The business is far healthier at the connectivity level than the flat headline revenue suggests — but it is also a mature, capital-intensive, heavily-indebted network whose unit economics are now being contested for the first time in its history. Understandable, durable cash generation; contested growth.

3. Growth History and Forward Opportunities

History — from roll-up engine to stall. Charter’s growth from 2013 to roughly 2022 was a combination of the TWC/Bright House integration (scale, pricing normalization, product upgrades on acquired bases) and the broadband secular tailwind — the steady migration of households to higher-speed internet, accelerated by COVID. Revenue compounded from ~$40B (2017) to a peak of $55.1B (FY2024), and the broadband subscriber base grew almost every quarter for a decade. That era is over. Revenue declined to $54.8B in FY2025, the broadband base has now contracted for four straight quarters (−393K internet customers in FY25; −120K in Q1’26), and adjusted EBITDA growth has decelerated to under 1%.

The growth decomposition today is stark and bifurcated:

• Internet: Negative on units (−120K Q1’26), positive on revenue (+1.7% FY25) — pricing/mix is still outrunning volume loss, but the buffer is thinning. The deterioration is driven by a top-of-funnel problem (weak gross adds, especially in the low-income segment post-ACP) rather than churn, which remains at historical lows. Management’s diagnosis — that yield-at-point-of-sale and churn are fine but consideration/traffic is weak — is important: it implies the problem is competitive consideration and marketing, partly fixable, rather than a customer exodus.
• Mobile: The unambiguous growth engine — +1.9M lines / +19% in FY25, 12M+ lines and +17% YoY by Q1’26, +22% service revenue. Spectrum Mobile is now among the fastest-growing wireless lines in the country, leveraging the MVNO + CBRS economics. Runway remains large: mobile penetration of Charter’s own broadband base is still well below carrier-store norms, and the Cox footprint (very low mobile penetration) is a fresh cross-sell pool.
• Video/Voice: Managed decline. The strategy has shifted from defending these as standalone products to using a re-bundled, app-inclusive video package as a churn-reducer for broadband — and it is working (video losses narrowed to 60K in Q1’26 from 181K a year earlier).
• Rural/subsidized expansion: A genuine new-passings growth vector — +41K rural net customer adds in Q1’26, subsidized rural passings +483K over the trailing year — though its ROI is increasingly capped by LEO satellite in the lowest-density areas.

Forward opportunities. Five identifiable levers: (1) Cox integration — applying Spectrum pricing/packaging, mobile and video to Cox’s low-penetration base (>70M passings pro forma), a play Charter has executed repeatedly (TWC, Bright House, Bresnan); (2) mobile penetration — continued attach into the existing and Cox broadband bases; (3) DOCSIS 4.0 / multi-gig — defending share against fiber by closing the product gap at a fraction of fiber’s cost-per-home; (4) B2B/enterprise — Cox brings best-in-class hospitality and B2B capabilities to extend across Charter’s footprint; (5) convergence-driven churn reduction translating into higher customer lifetime value even without unit growth. Management has also floated longer-dated network-asset monetization (edge compute, fiber-powered DAS, B2B GPU-as-a-service) — optionality, not a near-term driver.

Verdict — low-quality top-line growth, but improving cash quality. On the metric that historically defined Charter — broadband subscriber and revenue growth — the growth is currently negative-to-flat and low-quality, driven by price over volume in a contested market. The genuine high-quality growth is mobile, which is real, capital-light and accelerating, but not yet large enough to offset the legacy declines at the consolidated level. The honest read: this is no longer a growth company; it is a stabilizing, cash-harvesting utility with one fast-growing product line and one large integration opportunity (Cox). The investment case does not require a return to revenue growth — it requires the top line to not fall apart while capex normalizes. That is a lower bar, but in a structurally deteriorating industry it is not a trivial one.

4. Industry Dynamics

Market structure: a saturated end market in a two-front oversupply war

US residential broadband is, at the household level, a mature, near-saturated market. Fixed-broadband household penetration has plateaued in the low-90% range, so the industry’s organic growth rate is now roughly the rate of household formation — ~1%/yr. In a no-growth market, one operator’s gains are another’s losses; the relevant question is no longer “how fast is the pie growing” but “who is taking whose slice.” For two decades cable answered that question decisively. That era is over.

Cable still dominates the installed base: top cable operators held ~62.5% of US broadband subscribers at end-3Q2025 (Leichtman Research Group). But share of the flow — gross additions, the leading indicator of the future installed base — has collapsed. In Q4 2025, cable captured just 41% of broadband gross adds (its lowest share ever), while fixed wireless access (FWA) took 32% and fiber ~26% (Leichtman, via Light Reading). The net-add reversal is starker: the cable category went from +5.2M broadband net adds in 2020 to ~1.3M net losses in 2024. Essentially all of the category’s marginal growth since 2022 has been captured by FWA, which surpassed 13M customers by early 2025. Charter’s own trajectory mirrors the category — internet customers fell 120K in Q1’26 versus ~59K a year earlier; the bleed is accelerating, not stabilizing.

Profit pools remain large and, for now, intact: incumbents run ~40% EBITDA margins (Charter) to ~53% (Cable ONE) on broadband-led revenue because the network is sunk and incremental connections are nearly free. But profit pools and profit trajectory differ. With unit volumes flat-to-down and pricing power capped by two new substitutes, the pools are slowly draining at the margin even as headline margins hold.

Threat 1 — Fixed wireless access (the swing variable)

FWA is the single most important variable for the entire cable thesis. Bear framing: FWA is a structurally cheaper, “good-enough” substitute delivered over spectrum the carriers already own, with near-zero incremental plant cost per home — the worst kind of competitor for a sunk-cost incumbent. Bull framing: FWA is a capacity-constrained, best-effort product with a hard ceiling, skimming the low-usage, price-sensitive tail rather than the high-ARPU core. The evidence increasingly favors the bull framing at the margin:

• The ceiling is real and approaching. New Street Research estimates the big-3 carriers can currently support up to ~32M FWA subscribers, and they have filled ~50% (~16M). Bernstein puts Verizon at ~83% and T-Mobile at ~68% of total network capacity utilized. FWA is parasitic on the same towers that carry mobile data — every fixed line consumes capacity that could carry higher-value mobile traffic, which is why carriers gate FWA availability sector-by-sector.
• Adds are decelerating and quality is degrading. New Street expects FWA net adds to taper from 2026. Ookla data show FWA speeds for all three carriers declined in 2025, and the CTIA warns networks may fail to meet peak demand “as early as 2026.”

Interpretation: FWA has done real, permanent damage — it captured the category’s growth and took cable’s price-sensitive marginal customer. But it is not a like-for-like replacement for a wired multi-hundred-Mbps connection in a heavy-usage household, and its runway is finite. If the ceiling binds by 2027, cable’s net-add bleed slows materially. That is the crux of the recovery thesis.

Threat 2 — Fiber overbuild (the structural one)

Fiber is the more durable threat: once built it is a strictly superior product (symmetrical multi-gig, low latency) at comparable price, and it does not face FWA’s capacity ceiling. MoffettNathanson estimates the share of cable plant overbuilt by fiber rising from ~40% toward ~55%+ over the next decade; Comcast guided to ~60% fiber overlap by end-2025. AT&T had 29M fiber passings at end-2024 (targeting ~50M+ by 2029–30, >60M incl. Lumen); Verizon closed its Frontier acquisition January 2026 (>30M combined fiber homes); Brightspeed passed >3M. This is the threat with no obvious ceiling. The mitigant is DOCSIS 4.0: Charter is upgrading coax to symmetrical multi-gig at a fraction of fiber’s cost-per-home, targeting ~half its footprint multi-gig-capable by year-end 2026. Whether DOCSIS 4.0 closes the perceived speed/symmetry gap before fiber finishes the overbuild is a core thesis variable.

Threat 3 — LEO satellite (the rural flank, “frenemy”)

Starlink scaled from ~2.3M subs (end-2023) to >10M (Feb 2026); Amazon’s “Leo” targets US service in 2026. For Charter, LEO is primarily a rural/unserved threat with limited overlap with its dense suburban core — a “frenemy” that caps the ROI on subsidized rural expansion but does not attack the high-density base, and which management suggests could become a reseller/bundle partner.

Convergence: the one place cable is on offense

The strategically important counter-current is mobile. As MVNOs riding wholesale capacity, cable operators have become the de facto fourth national wireless force: cable MVNOs (Spectrum + Xfinity + Optimum Mobile) took ~45% of total US industry postpaid phone net adds through Q3 2025 — more than any facilities-based carrier. This generates incremental high-margin revenue on near-zero incremental capital and lowers broadband churn via the converged bundle — the only durable defensive weapon cable has. Caveat: cable mobile net adds are themselves decelerating, and convergence reduces churn rather than restoring growth.

Regulation

• ACP termination (ended June 1, 2024): the Affordable Connectivity Program subsidized ~23M households at peak; an estimated ~5M dropped service post-lapse — a real, identifiable drag on low-income broadband net adds across all of cable in 2024–25.
• BEAD ($42.45B): tech-neutral revisions now favor lower-cost FWA/LEO over fiber and target unserved locations, largely bypassing Charter’s served footprint — a modest negative (subsidizes competitors more than Charter).
• FCC / M&A: the Carr FCC is light-touch and consolidation-friendly. It approved Charter–Cox ($34.5B) on Feb 27, 2026 (with onshoring/rural-investment conditions); the California PUC is the last remaining approval. Net neutrality is dead federally (6th Circuit struck Title II authority, early 2025).

Capital-cycle read (Marathon)

A textbook supply-side oversupply setup. Cable earned high, stable returns for two decades; high returns attracted capital; that capital is now flooding in — but from the telcos and carriers, not from cable — into a no-growth, saturated end market. By the framework’s logic, returns on that capital should mean-revert toward mediocrity; fiber-overbuilder ROI fatigue is a question of when, not if. The non-obvious inflection: cable is exiting its own capex cycle while the telcos are mid-boom. Charter’s capex is rolling from ~$11.7B toward <$8B by 2028; Cox consolidation removes a competitor’s reinvestment. The bull case is that survivors with the lowest cost-per-bit emerge from the overbuild war as cash-flow fortresses once entrants’ capital is exhausted and FWA’s ceiling binds.

Verdict: Structurally bad — a deteriorating, ex-growth, consolidating industry. A saturated end market, a once-cozy local-monopoly cable structure now fighting a two-front oversupply war (fiber + FWA) plus a rural LEO flank, capped pricing power, and negative unit-add economics. The mitigants — FWA’s capacity ceiling, cable’s own capex discipline, consolidation-friendly regulation, the convergence land-grab — are real and could mark a capital-cycle bottom for the survivors, but they argue that the best-positioned incumbents are cheap cash generators in a declining industry, not that the industry is attractive. The structural verdict is negative.

5. Competitive Position

Naming the moat (Greenwald): local sunk-cost scale + weak captivity

Charter’s advantage is best classified as a local economies-of-scale advantage grounded in a sunk-cost network, weakly reinforced by customer captivity. The mechanism is specific: Charter has already built — and largely depreciated — coaxial plant passing ~57M homes. For a home it already passes, the marginal cost to connect and serve a customer is near zero. A new entrant (fiber or FWA) wanting that customer must sink fresh capital — ~$1,000–$1,500/home for fiber, or scarce spectrum capacity for FWA — to compete for revenue Charter can already capture at almost no incremental cost. That cost asymmetry on already-passed homes is the entire moat. It is a classic Greenwald scale advantage (high sunk cost, near-zero variable cost) and it is intensely local — exactly where Greenwald says scale advantages are strongest.

Crucially, the captivity leg is weak. Broadband is a near-commodity; switching costs are low (no data lock-in, no retraining, professional installers make a swap trivial). Customers do not stay because they are captive; they stay because, historically, there was no better/cheaper alternative at the home. The moat was never captivity — it was the absence of a credibly-cheaper-to-serve competitor. That premise is precisely what fiber overbuild and FWA dismantle. Per Greenwald’s own warning — “market growth and new capital are the enemy of economies-of-scale advantages” — the flood of entrant capital into the footprint is the textbook way a local scale moat erodes: once a fiber builder has sunk the plant, the cost asymmetry on those homes vanishes and the entrant has the better product. The moat dissolves home-by-home as the overbuild advances.

The defensive weapons: DOCSIS 4.0 cost advantage and convergence

Charter is not defenseless. Two genuine, financially-grounded advantages remain:

1. Multi-gig at a fraction of overbuild cost. Charter can upgrade existing coax to DOCSIS 4.0 symmetrical multi-gig for a small fraction of a greenfield fiber build’s cost-per-home (~half the footprint multi-gig-capable by year-end). This narrows the product gap versus fiber while preserving the cost gap — if a 2Gbps DOCSIS connection is indistinguishable from 2Gbps fiber, the entrant’s sunk capital earns a far lower return (the “fiber-overbuilder ROI fatigue” the capital-cycle thesis depends on).
2. Convergence as a churn-reducer. Spectrum Mobile (12M+ lines, +17%/yr) lets Charter bundle mobile with broadband at attractive prices, lifting two-product penetration to 35.2%. Convergence does not restore broadband growth, but it measurably reduces churn and raises household LTV at near-zero incremental capital — the single most important defensive lever, and the one place cable is on offense.

Head-to-head: does Charter actually hold share?

Management’s central claim is that Charter holds or gains broadband share even in mature fiber-overlap markets, and that FWA takes the low-usage/price-sensitive tail, not the high-ARPU core. Pressure-testing: vs. fiber, the worst share loss occurs in the first 18–24 months after a fiber launch, then stabilizes as the remaining base is genuinely sticky and DOCSIS multi-gig closes the gap — plausible and partly evidenced (cable still holds ~62.5% of the installed base; net losses are low-single-digit-% of base per year, not a rout), but it is management’s framing of its own data and is a hypothesis. The honest read: Charter loses share where fiber arrives, but more slowly and to a higher floor than the bear case assumes — erosion, not collapse. Vs. FWA, the capacity-ceiling and speed-degradation evidence supports the claim that FWA cannot durably take the heavy-usage core — the more defensible half of management’s argument.

Pressure test: durable or eroding? Eroding.

Be direct: the moat is real but eroding, and the erosion is structural, not cyclical. The advantage rested on a cost-to-serve asymmetry plus the absence of a credible competitor — not on captivity (switching costs are low). Two well-capitalized, strategically-motivated entrants are systematically sinking the very capital that creates the asymmetry, and fiber once built is a superior product with no capacity ceiling. The erosion is already visible: accelerating net subscriber losses, share of gross adds at an all-time low (41%), revenue rolling over. What the moat still buys is time and cash — a sunk network with ~40% EBITDA margins, a capex cliff converting to a large FCF step-up, a low-cost DOCSIS defense, and a convergence bundle that slows churn. It is durable enough to harvest enormous free cash flow through the transition and to consolidate (Cox); it is not durable enough to resume subscriber growth or restore pricing power against a permanently more-competitive structure.

Verdict: A real but structurally eroding moat — challenged, not broken. A cash-flow fortress defending a slowly-shrinking castle, not a widening-moat compounder. The investment question is therefore not “is the moat durable forever” (it isn’t) but “does the fortress generate enough cash, fast enough, before the moat narrows” — a valuation/timing question, not a quality one.

Peer comparison

The peer set frames Charter as the deep-value, well-managed-but-challenged name. Comcast trades at a near-identical EV/EBITDA — the multiple compression is sector-wide, not Charter-specific, which both validates the cheapness and warns it reflects a real structural derate. Altice USA is the cautionary tale: ~7.8x net leverage, revenue down 5%+, a franchise-rights impairment — the literal picture of what excess debt does to a cable operator in this cycle, and the reason Charter’s deleveraging path matters. (Sources: Comcast Q4’25 8-K; Altice USA Q3’25 8-K; Cable ONE Q4’25 8-K; stockanalysis.com, Mar 2026.)

6. Financial Quality

Revenue composition — a connectivity utility wrapped around a melting video business. FY2025 revenue was $54,774M, down 0.6% — the top line peaked in FY2024 ($55,085M) and has now rolled over. The composition matters more than the aggregate, because two large, opposing currents net to roughly flat. The growth engine is Connectivity (Internet + Mobile, +4.1%) — residential Internet grew revenue 1.7% despite losing 393K customers, because rate/product-mix added $785M while volume cost only $380M (FY25 10-K, p.43). That is the single most important quality fact in the model: Charter still has residential broadband pricing power even as subscribers decline. Mobile (+22%, $3.76B) is the clearest secular tailwind. Against that, Video (−9.4%, −$1,426M) and Voice (−6.0%) are in structural decline — though ~$322M of the video drop is the “seamless entertainment allocation” (bundling programmer streaming apps, with roughly-matching cost reallocation, so partly optical), and Advertising’s −17.6% is political-ad cyclicality. Stripping those, the underlying core decline is milder than the headline.

Adjusted EBITDA and margin structure — plateaued, not expanding. Adjusted EBITDA was $22,708M (+0.6%), a 41.5% margin. Charter defines Adjusted EBITDA as net income plus NCI, net interest, taxes, D&A, stock comp ($673M), and special/M&A items — a generous, capitalization-heavy definition that excludes $673M of real stock comp:

Margin crept up ~150bps over five years, driven mostly by declining programming costs ($8.8B/27% of opex in FY25 vs $9.7B/29% in FY24) as video shrinks — i.e., margin expansion is a byproduct of the video business dying, not operating leverage on a growing base. EBITDA growth has decelerated to <1%. This is a low-growth, high-margin cash machine, not a compounder.

Free cash flow — the entire thesis lives here. FCF (mgmt definition: OCF − capex + change in accrued capex) was $5,004M FY25 vs $4,257M FY24, $3,490M FY23, $6,102M FY22. The FY25 step-up came partly from non-recurring items (lower cash taxes +$669M, mobile-device working capital +$398M, lower cash interest +$347M). The bull case is entirely about capex falling, not EBITDA rising. Capex composition:

Capex-cliff math: the elevated buckets are line extensions ($3,940M, of which $2,202M subsidized rural) and the network-evolution upgrade/rebuild ($1,937M). Management guides total capex from ~$11.7B (FY25) → ~$11.4B (FY26) → <$8B run-rate by 2028. A ~$3.7B reduction, taxed-through, is the ~$28/share of incremental FCF management points to. Two cautions: (a) FY25 capex ($11.66B) still far exceeds D&A ($8.71B) — FCF will mechanically improve as capex converges toward depreciation; (b) the cliff is a guide, not a contractual fact — rural obligations and FWA-driven defensive spend could keep it sticky.

Balance sheet & debt — investment-grade structure, sub-IG leverage. Total assets $154,213M, overwhelmingly PP&E net $46,444M, indefinite-lived franchises $67,471M (never amortized, impairment-tested), and goodwill $29,710M. Cash is a trivial $477M — the company runs on its revolver and OCF. Debt principal is $94,617M: ~$11.9B credit facility, ~$55.4B IG senior secured notes, ~$27B CCO Holdings unsecured + legacy TWC debentures. ~87% fixed-rate (weighted ~5.07%); weighted cost ~5.2%. The maturity ladder is long and well-spread (2026 $1.1B, 2027 $3.6B, 2028 $5.4B, 2029 $7.3B, 2030 $13.9B incl. revolver, $63.5B thereafter) — trivial near-term refinancing against $16B of OCF. The yellow flag: marginal new debt is repricing up — January 2026 issuance was 7.000% and 7.375%, ~200bps above the existing stack, so as long-dated bonds mature, interest expense has a slow upward bias if rates hold. Net leverage 4.15x. Interest coverage: $22.7B EBITDA / $4.98B cash interest = 4.6x — adequate but not comfortable for a 4x-levered melting-video business; coverage deteriorates if EBITDA stalls while refinancing climbs toward 7%.

Share count — Class A understates the economic claim. The most commonly mis-modeled item:

• Class A common: 126.6M shares at 12/31/25 — down from 141.9M (12/31/24) and 152.7M (12/31/22) on buybacks. (The “~141M” sometimes quoted is the FY24 figure.)
• A/N (Advance/Newhouse) holds 15.5M Charter Holdings common units, exchangeable 1:1 into Class A — economically equivalent shares sitting below the consolidation line as NCI.
• Standalone as-converted economic shares ≈ 126.6M + 15.5M ≈ ~142M.
• The ~179M pro-forma figure is post-Cox: Cox issues ~33.6M Charter Holdings common units + $6.0B of 6.875% convertible preferred to Cox Enterprises; as-converted, that lifts the count toward ~179M (offset by the ~6.8M net reduction from the Liberty buy-in). Per-share math must use ~142M (standalone) or ~179M (PF Cox), never the 126.6M Class A count alone — doing so overstates per-share value by ~12–40%. Note also the Tax Receivable Agreement: Charter pays A/N 50% of tax benefits from unit-exchange basis step-ups — a recurring cash leakage to the minority.

ROIC/ROE — adequate, not excellent; ROE is an illusion. NOPAT ≈ operating income $12,908M × (1 − 22.7%) ≈ $9,978M; invested capital ≈ debt $96.2B + equity $20.5B − cash $0.5B ≈ $116.2B → ROIC ≈ 8.6%, modestly above an estimated ~7–8% WACC. ROE of ~31% ($4,987M / $16,054M equity) looks spectacular but is an artifact of a thin, buyback-depleted equity base (accumulated deficit $(5,393)M) levered 4x — it measures financial engineering, not business quality.

Verdict — do the economics justify the leverage? Borderline, and the answer depends entirely on the capex cliff. The business throws off real cash (~$5B FCF, heading toward ~$8–9B+ if capex falls as guided), interest is covered 4.6x, and broadband still has pricing power. But this is a low-growth (~flat revenue, <1% EBITDA growth), ~8.6%-ROIC asset carrying 4.15x leverage against a structurally declining video base and intensifying competition on the core broadband product. The leverage is justified only if (a) the capex cliff materializes and (b) broadband revenue stays roughly flat. The economics support the leverage today; they offer little margin of safety if the top line accelerates downward.

Multi-year data block (reconciled to filings)

Operating KPIs (12/31, 000s): Total Internet 30,083 (FY24) → 29,680 (FY25); Video 12,892 → 12,605; Voice 6,884 → 6,046; Mobile lines 9,858 → 11,766 (+19%). FY25 net changes: Internet −393K, Video −287K (ending), Voice −838K, Mobile +1,908K. Customers >60 days past due improved to 82,300 (from 102,500) — a modest credit-quality positive. (Sources: FY25 10-K, FY23 10-K, EDGAR XBRL companyfacts CIK 1091667; FY21 figures from the FY2021 earnings release.)

7. Quality of Earnings & Filings Review

Run-rate distortions to normalize before valuation: (1) Seamless-entertainment allocation (~optical) — bundling programmer apps cut FY25 video revenue by $322M with offsetting cost; reported video declines overstate the economic deterioration. (2) Political-ad cyclicality — advertising fell 17.6% (−$312M) almost entirely on the 2024-election comp; biennial, not structural. (3) Programming-cost tailwind — costs dropped to $8.8B/27% of opex; a margin tailwind as long as video keeps shrinking. (4) Capex >> depreciation gap closing — FY25 capex $11,659M vs D&A $8,711M; the customer-relationship intangible has amortized to $440M and rolls toward zero (a forward D&A tailwind). (5) Deal noise in other opex — rose to $416M including a $142M asset-disposal loss, $129M M&A costs, $18M special charges (Liberty/Cox noise to normalize out). (6) Cash-tax timing — FY25 FCF flattered by a $669M cash-tax decline and $398M mobile-device working capital; don’t extrapolate the full FY24→FY25 FCF jump.

Accounting conservatism — standard for cable, but EBITDA-flattering. Charter capitalizes installation labor and line-extension construction into PP&E (industry-standard, but it means Adjusted EBITDA overstates true free economics versus a pure-opex framing). Franchises ($67.5B) are carried indefinite-lived and only impairment-tested. Depreciation lives are reasonable (distribution 6–22 yrs, CPE 3–8 yrs). No aggressive revenue recognition or off-balance-sheet red flags surfaced. The two flatters are EBITDA’s exclusion of $673M stock comp and heavy labor capitalization.

8-K material-event timeline (~36 months): routine quarterly earnings through 2023–24; Liberty Broadband merger agreement (Nov 2024); Cox transaction agreement + voting agreements with Cox Enterprises & A/N (May 2025); deal financing 8-Ks and Jan-2026 7.00%/7.375% note issuance; FY25 earnings + 10-K (Jan 30 2026); Q1’26 earnings (Apr 23 2026). Theme: an operationally quiet company that turned deal-heavy over 18 months — simultaneously absorbing its parent (Liberty Broadband) and acquiring a major peer (Cox). Integration risk and pro-forma leverage are now central.

Insider transaction read — a genuine conviction-buying cluster. Sampling the most recent ~13 Form 4s: open-market PURCHASES (code P), April–May 2026, after the ~69% drawdown — CEO Christopher Winfrey bought 6,936 sh @ $172.23 (Apr 28); director Wade Davis 5,728 @ $173.72; director Balan Nair 1,000 @ $175.46; director Mauricio Ramos 9,929 @ $140.93 (May 15, near the current price). A sitting-CEO-plus-three-director discretionary buying cluster at depressed prices is a real, mildly bullish signal (code-P buys are rare in cable). Offsetting: ex-CEO Thomas Rutledge sold ~87.8K sh @ $144–147 (a legacy holder, not current management); Liberty Broadband’s 1.26M-share disposition @ $204.33 is the contractual Charter repurchase (deal mechanics, not a market signal). Net read: the insider tape skews mildly bullish.

One governance flag: the SEC’s November 2023 settled action charged Charter with an internal-controls violation tied to the 10b5-1 trading-plan provisions governing its own buyback program — a process lapse, not fraud, but a reminder that even the buyback machine has had control weaknesses.

8. Capital Allocation

The model: lever up, buy back, repeat. Charter runs the most concentrated equity-shrinkage program in US large-cap. The philosophy is explicit and unchanged for a decade: hold net leverage in a 4.0x–4.5x band, fund all capex internally, pay no dividend, and route essentially all discretionary capital plus the debt capacity created by EBITDA growth into repurchasing Class A stock. Since the September 2016 TWC/Bright House close, Charter has repurchased roughly $73.7 billion of equity — taking share count from ~290M (2016) to ~127–142M today, a ~46% reduction in the economic base. For nearly a decade this was textbook value creation: EBITDA grew, the multiple held, and a shrinking denominator levered modest enterprise growth into outsized per-share gains.

The levered-equity math — why this cuts both ways. Equity is the thin residual on a vast enterprise: ~$21B market cap atop ~$117B EV — roughly 6:1 leverage on enterprise value. When EBITDA and the multiple hold, buying back the equity sliver at a low multiple is intensely accretive. When enterprise value erodes, the same leverage runs in reverse — the debt is fixed, so the entire decline lands on equity, and buybacks executed at higher prices become value-destructive in hindsight. The ~69% drawdown from $422 to $132 is precisely this reverse-gear at work. The recent timing is the honest indictment: Charter repurchased $5.13B in FY2025 and $963M in Q1’26 at ~$225/share — now $132. The model has no circuit-breaker for the possibility that the asset is structurally, not cyclically, cheap. Buying back stock at $225 that is now $132 is a real destruction of shareholder capital that no track record erases.

M&A: a defensible record, including Cox. The transformational TWC/Bright House deals (2016) built the platform. The pending Cox acquisition (DEFM14A, 2025-07-02) continues the discipline: Cox is valued at $34.5B EV (incl. ~$12.6B assumed net debt), an implied ~6.4x EBITDA (~$5.4B Cox EBITDA) — below recent cable precedents (7.6x–10.9x). Crucially, Charter is paying largely in partnership units priced at $353.64 and convertible preferred struck at $477.41 (35% premium) — issuing equity-linked paper far above today’s $132, not cheap stock. Synergies were raised $500M → ≥$800M run-rate opex. The concurrent Liberty Broadband buy-in (0.236 exchange ratio) is share-count accretive and removes the long-standing Liberty/GCI holdco overhang. This is sensible consolidation at sane prices — the opposite of empire-building.

Incentive alignment (2026 DEF 14A). Comp is heavily equity/option-weighted, aligning management directionally with shareholders — but the metrics matter. CEO Chris Winfrey’s annual bonus (target 300% of a $2.5M base) keys on Adjusted EBITDA (55%), Revenue (15%), Strategic Objectives (30%); 2025 paid at 94.3%. There is no FCF-per-share or ROIC hard metric — FCF enters only as a soft component. Long-term incentives are ~90% at-market stock options / 10% RSUs (2025 strikes $267–$350); the 2023 Performance Equity Program layered stock-price hurdles (5%–20% CAGR). The design rewards absolute stock appreciation and EBITDA defense — strongly aligned in spirit (genuine owner-operators in the Newhouse/A/N family and Liberty/Malone legacy hold large stakes), but it rewards EBITDA level, not per-share value creation or returns on capital. The 5-year pay-vs-performance window shows CHTR down ~68% cumulatively versus peer gains.

Verdict. Intelligent capital allocation historically and on M&A price discipline — but the model is inseparable from a single bet (broadband EV holds), it buys mechanically without a structural circuit-breaker, the recent $225 vintage is deeply underwater, and the incentive design rewards EBITDA defense over per-share value. Call it disciplined financial engineering on an asset that is no longer clearly growing — defensible if EBITDA stabilizes, value-destructive if broadband erosion compounds.

9. Changes & Headwinds — Last ~2 Years

1. ACP wind-down (mid-2024) — the demand air-pocket. The federal Affordable Connectivity Program expired, removing the ~$30/month subsidy for several million Charter low-income relationships (~5M ACP customers). The resulting non-pay churn, layered on competitive losses, turned net adds negative. It is a transient headwind (the cohort washes through) but it masked and accelerated an underlying competitive problem — hard to separate “ACP roll-off” from “real competitive loss,” which is why bull and bear read the same prints differently.

2. Broadband subscriber losses — the dominant headwind. Four consecutive quarters of internet losses, including ~120K in Q1’26 (worse than ~100K consensus), driven by FWA (low end) and fiber overbuild (premium end). The single most important negative of the period and the direct cause of the share collapse. Thesis-weakening; the open question is stabilization (bull) vs. compounding (bear).

3. Pricing & packaging relaunch (late 2024) + convergence. The strategic answer: a relaunched architecture bundling broadband with Spectrum Mobile and ad-supported streaming apps at no incremental cost, to lower churn and defend ARPU rather than chase gross adds. Early evidence is mixed-positive — video losses narrowed to 60K in Q1’26 from 181K a year earlier. A credible, rational pivot from a volume model to a value/retention model — but it trades subscriber growth for ARPU/retention, so headline sub trends may stay ugly even as economics improve.

4. Spectrum Mobile scaling — the genuine growth engine. 12M+ lines, +17%/yr; the convergence flywheel’s hub and the only segment adding customers at scale. Thesis-strengthening; the open question is segment profitability as it scales.

5. Cox acquisition + Liberty Broadband merger — the structural reset. The $34.5B Cox deal (FCC-approved Feb 2026; California CPUC the last gate) and the Liberty buy-in reshape the entity: >70M passings / ~38M-subscriber combined footprint, Cox Enterprises ~23% owner via partnership units, ≥$800M synergies, low-mobile/low-video Cox base as a cross-sell pool. Thesis-strengthening — scale, sane price, paper issued above market, overhang cleared — though it adds ~$12.6B Cox net debt and near-term integration/leverage concerns.

6. Leadership & network evolution. Nick Jeffrey (ex-Vodafone CEO) joined as COO (Sept 2026) — a convergence-credentialed operator hire for the exact strategic gap. Winfrey (CEO since Dec-2022) remains the architect. The network-evolution program (DOCSIS 4.0 / multi-gig) is the capex driving current intensity and the source of the coming <$8B capex cliff.

Verdict. The period strengthened the structural/strategic case (consolidation at good prices, convergence pivot with early proof, mobile scaling, capex cliff in sight, overhang removed) while weakening the core operating case (entrenched broadband losses). The stock has priced the operating deterioration violently while discounting the structural improvement — the heart of the variant perception.

10. Risk Analysis

The risk profile is unusual: the business risks are well-understood and slow-moving, but the capital-structure leverage converts modest business deterioration into large equity outcomes. The dominant risk is not a single catastrophe but the compounding of broadband EBITDA decline against a fixed $94B+ debt stack.

Risk of catastrophic / total loss. Low but non-trivial and worth naming explicitly. The path to severe permanent impairment is the combination of (a) broadband EBITDA declining several percent per year, and (b) refinancing the debt stack upward as the low-coupon bonds mature — at which point interest consumes a rising share of a shrinking EBITDA and equity is squeezed toward zero (the Altice trajectory, down ~96%). What protects against total loss in the near-to-medium term is the long, well-laddered maturity profile ($63.5B due after 2030) against $16B of annual operating cash flow and 4.6x interest coverage — the company is not at refinancing risk this decade absent a severe EBITDA collapse. The realistic bear outcome is a value trap / large drawdown, not a zero, on any reasonable horizon.

Verdict: A high-impact, medium-likelihood risk cluster centered on a single variable (terminal broadband EBITDA), amplified by leverage. This is not a low-risk security; it is a deeply-discounted one whose discount is commensurate with genuine structural risk. The margin of safety lives in the FCF/valuation, not in the business durability.

11. Valuation Discussion — Embedded Expectations

No price target, no buy/sell — this section backs out what the $132 price implies and stress-tests whether it is defensible.

Where the stock trades. At $132.12: market cap ~$20.7B, EV ~$117B, EV/EBITDA ~5.3x, trailing P/E ~3.6x / forward ~3.0x, P/S ~0.38x, no dividend. AZI’s own-history valuation index puts CHTR at a decade-low decile on P/E, P/B and P/S — the cheapest the stock has ever been on its own multiples.

Comp set — cable peers, all cheap, for a reason.

Source: yfinance, 2026-06-07; reconcile to filings. The entire US cable complex trades at 4–5x EBITDA — multiples normally reserved for terminally declining businesses — because the market is pricing broadband as a structurally challenged, ex-growth utility under FWA/fiber assault. Altice USA is the cautionary extreme (equity down ~96% from its 2017 peak even while still generating ~$3.4B EBITDA). CHTR sits a notch above peers on EV/EBITDA but at the bottom on P/E (leverage amplifies equity cheapness).

Embedded expectations — what the $132 price underwrites. At EV ~$117B on ~$22.7B EBITDA = ~5.3x. A no-growth, infinite-life FCF asset deserves roughly 8–10x EBITDA (a 10–12.5% unlevered FCF yield for a low-volatility cash stream). At 5.3x, the market is not pricing zero growth — it is pricing persistent EBITDA decline. A crude test: if a fair multiple for a stable broadband cash machine is ~8x, then 5.3x ÷ 8x implies the market discounts roughly a 2.5–3.5% perpetual EBITDA decline rate. Put differently, the price embeds the bear case that broadband EBITDA erodes ~3%/year forever. That is an aggressive assumption for an asset that still grew EBITDA +0.6% in FY25, is the low-cost high-speed provider in most of its footprint, and has a credible mobile-convergence offset. The market is underwriting terminal decline; the data so far shows flatness.

The capex cliff — the asymmetry the multiple ignores. The valuation hinges on free cash flow, not EBITDA, and FCF is set to inflect. FY2025: OCF $16.08B − capex $11.7B = FCF $5.0B. As the network-evolution build completes, management guides capex to <$8B by 2028. Holding EBITDA merely flat, that ~$3.7B reduction flows almost entirely to FCF:

• FY2025 FCF: ~$5.0B → 2028E FCF (flat EBITDA, capex <$8B, lower cash interest): ~$11–13B

On a post-Cox as-exchanged count of ~179M, that is roughly $60–70 FCF/share against $132 — a ~3.5–4x P/FCF / >25% FCF yield on 2028 numbers. This is the bull’s entire case: a price that looks like terminal decline attached to a business about to roughly double free cash flow purely from capex normalization. The market multiple prices the EBITDA line; the value is in the capex line.

Share-count / NCI dilution — getting the denominator right. Correct as-exchanged base: standalone Class A ~127–142M (incl. A/N 15.5M units) → Liberty buy-in net ~−7M → Cox +~33.6M Charter Holdings units → ~179M as-exchanged at close. The $6.0B Cox convertible preferred (6.875%, conversion at $477.41) is not dilution at $132 — it is deep out-of-the-money and behaves as a fixed, debt-like claim; model it as a preferred obligation, not shares, unless the stock more than triples.

What a fair multiple is. For a low-/no-growth but cash-gushing, heavily-levered asset, the right frame is FCF yield and deleveraging optionality, not EBITDA growth. If 2028 FCF reaches ~$11–13B and the market grants even 6–7x P/FCF (still cheap, 14–17% yield) for a stabilized-but-not-growing cash machine, the equity math improves dramatically as debt paydown (target 3.5–3.75x) transfers EV from creditors to equity. The bear counter: if EBITDA declines per the embedded expectation, the same leverage means equity absorbs the entire drop, the FCF cliff benefit is partly offset by shrinking EBITDA, and the multiple stays at 5x — the Altice path. The decisive variable is not the multiple; it is the EBITDA trajectory, which determines whether the capex cliff is harvested by equity or by creditors.

Valuation scenarios

All assume Cox + Liberty close; as-exchanged ~179M shares; $6.0B Cox preferred treated as a debt-like fixed claim. Combined CHTR+Cox EBITDA base ~$28B. Illustrative analysis of embedded expectations — NOT a forecast or price target.

Reading the table: the spread is driven almost entirely by the EBITDA trajectory (decline vs flat vs growth) — itself a proxy for whether broadband subscriber losses stabilize. The capex cliff is largely common to all three (a near-certainty of the build cycle), which is why even the bear case produces healthy FCF/share. The debate is therefore not “will FCF rise?” (it almost certainly will) but “will the rising FCF be captured by equity or bled away by EBITDA decline into the $94B debt stack?” That single question is what the 5.3x multiple is voting on.

12. Variant Perception

Consensus belief. The market and most of the sell-side treat US cable as a structurally impaired, ex-growth utility whose broadband franchise is in permanent slow decline as FWA and fiber erode share and pricing power. Charter specifically is seen as a levered version of that decline — a melting ice cube where ~$94B of debt sits on a thin, shrinking equity sliver, recently de-rated as Berkshire exited and four straight quarters of sub losses confirmed the trend. Hence 5.3x EBITDA and the decade-low multiple. The street price target (~$265) implies they think it overshot, but the narrative is bearish.

The strongest bull case. The market is conflating a stalling income statement with a rising free-cash-flow statement. Three things are true at once that the 5.3x multiple ignores: (1) the capex cliff roughly doubles FCF by 2028 mechanically, regardless of EBITDA growth, taking FCF/share toward ~$60 against a $132 price (>25% yield); (2) broadband EBITDA is far more stable than the unit losses imply — ARPU/mix still grows revenue, video losses are shrinking, mobile compounds at 17%, and ACP/cyclical drags are washing through; and (3) deleveraging transfers enterprise value from creditors to equity — in a 6:1-levered structure, paying debt from 4.15x to 3.5x is a large wealth transfer to the thin equity. Add disciplined Cox consolidation at ~6.4x (paid in above-market paper) and insider buying, and you have an asymmetric deep-value setup: you are paid >25% FCF yield to wait for stabilization, with large upside if the multiple merely normalizes toward a stable-utility 7–8x.

The strongest bear case. Broadband is in secular, not cyclical, decline, and the leverage makes that fatal for equity. FWA’s ceiling keeps rising (carriers keep finding spectrum), fiber overbuilds past 55% of the footprint with a permanently superior product, and DOCSIS 4.0 fails to stop the share/ARPU bleed. EBITDA declines 2–3%/year; the capex “savings” are partly consumed by defensive spend and partly offset by the shrinking EBITDA; and as the 5% bonds mature, refinancing at 7% consumes a rising share of a falling EBITDA. The equity — a 6:1-levered residual — compounds downward, exactly as Altice’s did (−96%). The buybacks at $225 prove management can’t tell structural from cyclical cheapness, and the “cheap multiple” is the market correctly pricing a melting asset. Berkshire saw it and left.

The 3–5 assumptions that matter most:

1. Terminal broadband EBITDA direction (flat vs −3%/yr) — the master variable; everything else is detail.
2. Does FWA plateau by ~2027? (capacity ceiling binds → cable bleed slows) — the swing input to #1.
3. Does the capex cliff materialize to <$8B and stick? — the magnitude of the FCF inflection.
4. Does DOCSIS 4.0 hold share in mature fiber-overlap markets to a high floor? — the durability of the moat.
5. Refinancing cost trajectory — whether deleveraging outruns the upward repricing of the debt stack.

Falsification tests. Bull is falsified if: broadband net losses accelerate through 2026–27 (beyond ACP/seasonal explanation) AND broadband ARPU turns negative — proving secular decline with no pricing offset. Bear is falsified if: broadband net adds stabilize toward flat for two-plus consecutive quarters AND FCF steps up visibly as capex falls — proving the melt was cyclical/ACP-driven and the cash inflection is real. The next 3–4 quarters of subscriber and ARPU prints, plus the 2027 capex trajectory, will largely settle the debate.

13. Fact vs. Interpretation Table

14. Open Questions

1. What is Charter’s exact % of footprint currently fiber-overbuilt and FWA-exposed? Management says “limited”/“above-competitor share in mature overlap” but gives no clean disclosure. The single biggest gap in assessing moat durability.
2. Does DOCSIS 4.0 actually hold share where fiber has been present 3+ years? Management’s “high floor” claim is its own framing of its own data; independent confirmation is lacking.
3. Will the capex cliff stick at <$8B, or will defensive/competitive spend keep it elevated? The guide is not contractual and FY25 capex still rose.
4. What is Cox’s precise 2025/2026E EBITDA (to pin the paid multiple exactly) and the final close-date as-exchanged share count?
5. Refinancing trajectory: how much of the $94B stack reprices to ~7% by 2028–30, and does deleveraging outrun it?
6. Short interest: the 35.8%-of-float feed reading is almost certainly garbled — actual short interest needs verification (typical cable short interest is low-single-digit %).
7. Will management pause buybacks to prioritize deleveraging post-Cox, or repeat the $225 mistake?

15. What Must Be True (Bull and Bear, each with a falsification test)

For the BULL to be right:

• Broadband EBITDA must stay roughly flat-to-slightly-positive through the transition (units stabilize as ACP washes through and FWA plateaus; ARPU/mix offsets volume).
• The capex cliff must materialize to <$8B and largely stick, delivering the ~$11–13B FCF.
• Deleveraging proceeds toward 3.5–3.75x, transferring EV to equity faster than the debt stack reprices upward.
• Falsification: broadband net losses accelerate through 2026–27 beyond ACP/seasonal explanation and broadband ARPU turns negative — secular decline with no pricing offset.

For the BEAR to be right:

• Broadband enters genuine secular decline (−2–3%/yr EBITDA) as FWA’s ceiling keeps rising and fiber overbuilds past 55% with a superior product DOCSIS can’t match.
• Refinancing at ~7% consumes a rising share of a shrinking EBITDA; coverage compresses below comfort.
• The capex “savings” are partly consumed by defensive spend, muting the FCF inflection; the multiple stays at 5x.
• Falsification: broadband net adds stabilize toward flat for two-plus consecutive quarters and FCF steps up visibly as capex falls — proving the melt was cyclical and the cash inflection real.

The crux: Both cases agree free cash flow rises mechanically from the capex cliff. They disagree only on whether the EBITDA base under that cash flow is stable (bull: equity captures a doubling FCF at a trough multiple) or melting (bear: creditors and decline absorb it, Altice-style). Everything in this report points to that one unresolved variable — the terminal direction of broadband EBITDA — which the next several quarters of subscriber and ARPU data will begin to settle.

APPENDIX A — Standard Diligence Questionnaire

Charter Communications, Inc. (NASDAQ: CHTR) — a supplement to the report above. Grounded in the same evidence base; labels (Fact / Interpretation / Assumption) applied where it matters.

General

What thoughtful questions have other investors asked about this company? The dominant questions cluster around three poles: (1) Is broadband decline cyclical or secular? — i.e., are the four straight quarters of internet losses an ACP-wind-down/housing air-pocket that washes through, or the start of a permanent FWA/fiber-driven erosion? (2) Will the capex cliff actually deliver the FCF inflection? — skeptics note the guide is non-contractual and capex still rose in FY25. (3) Is the leverage safe? — at 4.15x net debt with $94B outstanding and marginal refi at ~7%, bears invoke Altice USA as the template for what happens when a levered cable equity meets EBITDA decline. Sophisticated investors also probe the share-count math (Class A vs. as-exchanged ~179M post-Cox), whether buybacks should pause for deleveraging, and how much of the footprint is fiber-overbuilt (poorly disclosed).

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: neither extreme — earnings are at a plateau with cyclical headwinds (ACP wind-down, weak housing/move activity suppressing gross adds, a 2024-political-ad comp). EBITDA is near an all-time high in absolute terms but growth has stalled to <1%. If housing/move activity recovers and ACP churn fully washes through, current earnings are arguably a modest cyclical low within a flat structural trend.

Driven by the external environment or internal actions? Both. External: FWA/fiber competition, ACP termination, muted household formation. Internal: the late-2024 pricing/packaging relaunch (trading volume for retention/ARPU) and aggressive mobile attach are deliberate strategy choices shaping the mix.

How stable are revenues? Fact: highly stable in aggregate (recurring monthly subscriptions, very low non-pay churn) — revenue moved within a ~$54–55B band for four years. The composition is shifting (connectivity +4.1% vs. video −9.4%), but total volatility is low. This stability is what makes the FCF thesis credible.

Outlook for products/services? Internet: flat-to-down units, positive revenue (thinning buffer). Mobile: strong growth (+17–22%). Video/voice: managed decline, now repurposed as churn-reducers. Commercial/B2B: low-single-digit growth, a Cox-enhanced opportunity.

How big will this market be — growing, shrinking, domestic or international? Fact: 100% domestic. The US fixed-broadband market is saturated (~low-90% household penetration), growing only at ~household-formation rates (~1%/yr). Mobile is the one expanding adjacency for Charter (share gains from carriers). Net: a large but no-growth-to-shrinking end market.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Fact/Interpretation: decisively more competitive — for the first time cable faces two credible broadband substitutes (FWA, fiber overbuild) plus a rural LEO flank. Cable’s share of gross adds fell to an all-time-low 41% (Q4’25).

How profitable is the business (ROIC, ROE)? ROIC ~8.6% (modestly above WACC); ROE ~31% (a leverage/buyback artifact, not a quality signal). EBITDA margin ~41.5%. Honest read: a good-but-not-great-returns business whose headline ROE flatters via thin levered equity.

How profitable is the industry — competitors, barriers to entry? Incumbents run 40–53% EBITDA margins on sunk plant. Barriers were historically high (sunk-cost network = near-zero marginal cost on passed homes) but are falling as well-capitalized entrants sink fresh capital. A handful of national competitors (telcos, carriers) now contest a formerly local-monopoly structure.

Can the business be easily understood? Yes — sell connectivity over a network; monetize sunk infrastructure; the debate is purely about competitive trajectory and leverage, not business-model opacity.

Can it be undermined by foreign low-cost labor? No — physical local network; Charter explicitly markets its 100% US-based workforce as a differentiator.

Do brands matter? Modestly. “Spectrum” carries some recognition but broadband is near-commodity; brand is a weak moat leg relative to the cost-to-serve asymmetry.

Nature of competition / switching costs? Competition is on price, speed, and bundle value. Switching costs are low (no data lock-in; professional installs make swaps trivial) — the convergence bundle (mobile + broadband) is the main, partial, churn-reducer Charter has engineered.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Interpretation: the network’s replacement value and the spectrum/CBRS optionality are arguably understated; the franchise rights ($67.5B) and goodwill ($29.7B) are large but reflect past M&A, not hidden value. The mobile business’s strategic option value is not separately capitalized.

Off-balance-sheet liabilities? The Tax Receivable Agreement with A/N (50% of unit-exchange tax-step-up benefits) is a recurring cash leakage; programming and spectrum/MVNO purchase commitments are contractual obligations disclosed in the 10-K. No unusual off-balance-sheet leverage beyond standard operating commitments.

How conservative is the accounting? Standard-to-conservative for cable. Two EBITDA flatters: exclusion of $673M stock comp from Adjusted EBITDA, and heavy capitalization of installation labor into PP&E. Franchises carried indefinite-lived (impairment-only). No aggressive revenue recognition surfaced.

How CapEx-hungry is the business? Very — but temporarily peaked. Capex was ~21% of revenue (FY25, $11.7B), elevated by rural line-extension and the DOCSIS 4.0 network-evolution build. Management guides to <$8B run-rate by 2028 — the central thesis variable.

Capital Allocation & Management

How much FCF, and how is it used? ~$5.0B FCF (FY25), guided to roughly double by 2028 on the capex cliff. Use of cash: ~all discretionary FCF plus EBITDA-driven debt capacity historically routed to share buybacks (~$73.7B since 2016; share count −46%); no dividend. Philosophy: maintain 4.0–4.5x leverage, shrink the share count.

Significant acquisitions recently? Yes — the pending Cox Communications ($34.5B EV, ~6.4x EBITDA, ≥$800M synergies) and the Liberty Broadband buy-in. Both at disciplined prices; Cox paid largely in equity-linked paper struck above market.

Buying back shares? Aggressively — including $963M at ~$225 in Q1’26 (now $132), a real recent capital mis-timing. The model lacks a circuit-breaker for structural (vs. cyclical) cheapness.

Issuing large amounts of stock to insiders? No outsized insider issuance; comp is equity/option-heavy but at-market. The Cox unit/preferred issuance is acquisition consideration, not insider enrichment.

Compensation policy / motivations of management? CEO bonus keys on Adjusted EBITDA (55%), Revenue (15%), Strategic (30%); LTI ~90% at-market options with price hurdles. No FCF/share or ROIC hard metric — rewards EBITDA defense and absolute stock appreciation. Owner-operator alignment is genuine (Newhouse/A/N, Liberty legacy). Governance flag: 2023 SEC settled action on 10b5-1 buyback controls.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No — Class A common stock (NASDAQ: CHTR); standard 1099 treatment. (Note the partnership-unit/NCI structure at the Charter Holdings level affects share-count math but not the holder’s tax form.)

Dividend policy? None — 100% of returns via buybacks.

How profitable is the business? ~41.5% EBITDA margin; ~9% net margin; ROIC ~8.6%. Cash-generative but not high-returns.

Is net income diverging from cash from operations? OCF ($16.08B) vastly exceeds net income ($4.99B) — normal for a high-D&A network business; the gap is depreciation, not an earnings-quality red flag. The relevant divergence is OCF vs. FCF, driven by the (peaking) capex.

Risks & Downside

What factors would cause the stock to decline? Accelerating broadband sub losses; broadband ARPU turning negative; capex cliff slipping; refinancing at higher rates compressing coverage; Cox integration stumbles; renewed buybacks at the wrong price.

Risk of a catastrophic loss? Low-to-moderate and explicitly leverage-driven: the path is sustained EBITDA decline plus upward refinancing simultaneously (the Altice trajectory). Mitigated near-term by a long, well-laddered maturity profile ($63.5B due post-2030) against $16B OCF and 4.6x interest coverage.

Chance of a total loss? Very low on any reasonable horizon — the debt is long-dated and well-covered; the realistic bear outcome is a value trap / large drawdown, not a zero.

Recent News & Events

Has the business environment changed recently? Yes materially: ACP subsidy ended (mid-2024); AT&T entered FWA; Verizon closed Frontier (Jan 2026, >30M fiber homes); Starlink passed 10M; the FCC approved Charter–Cox (Feb 2026). Berkshire Hathaway exited its CHTR position in Q1 2026.

Significant acquisitions? Cox (pending, summer-2026 close target) and the Liberty Broadband merger — both reshaping the entity to ~38M subscribers / >70M passings.

Change in accounting policies? None material; the “seamless entertainment” programmer-app revenue/cost reallocation is a presentation matter (nets ~$322M within video), not a policy change.

Recent changes — new markets, facilities, management? COO Nick Jeffrey (ex-Vodafone) joining September 2026; ongoing DOCSIS 4.0 network evolution; subsidized rural footprint expansion (+483K passings TTM); late-2024 pricing/packaging relaunch now ~45% penetrated into the base.

APPENDIX B — Source Appendix

Charter Communications, Inc. (NASDAQ: CHTR) — sources consulted, with type and access date (June 2026). Primary sources prioritized; third-party AI/market-data signals reconciled to filings.

Primary — SEC filings (EDGAR, CIK 0001091667)

Primary — management commentary

Market & valuation data (third-party — reconciled to filings)

Industry & competitive research (trade press / sell-side cited)

Analytical frameworks

Note: Management commentary and third-party AI sentiment/valuation signals are treated as hypotheses and reconciled to primary filings throughout. No price target or buy/sell recommendation appears anywhere except the clearly-labeled “Claude’s Take” opening block.