Compare How Big Of A Threat Is Fixed
FWA — Sizing The Threat To Charter's Broadband Franchise
The fiber tab measures the long-duration capacity erosion. The technology-risk tab benchmarks every access technology side-by-side. This tab is the FWA-only sizing: how many customers are on FWA today, how much of Charter's sub loss it has caused, whether the wave is bounded by carrier capacity or structural, and what to watch in the next four prints to resolve it.
Bottom Line
FWA is the single biggest attributable driver of Charter's broadband sub decline in 2024–2026 and — more importantly — the marginal price-setter that broke residential ARPU below zero in Q1 2026. But it is not a stranded-asset event. Three carriers (TMUS, VZ, AT&T) collectively serve roughly 14–15M FWA subscribers today against a New Street capacity ceiling of ~32M (national, not Charter-footprint specific). Verizon's FWA growth has already slowed (1.5M net adds in 2024 → 1.1M in 2025); T-Mobile is plateauing into its own 12M-by-2028 target; AT&T is the late entrant that just refilled the growth pipeline with a $23B EchoStar spectrum acquisition. The honest read is that 2026 is the FWA peak-pressure year; whether 2027 brings the bounded plateau the bulls assume depends on (a) the upper C-Band auction (mandated by July 2027) and (b) AT&T's run-rate after EchoStar spectrum is digested. Charter's defense — Spectrum One bundle, the mobile-attach economics, the $24.99 Spectrum Internet Assist floor — blunts the low-end share-take but cannot reverse the ARPU repricing already in the print.
Big 3 FWA Subs Today (millions)
New Street Capacity Ceiling (millions)
% Of Ceiling Used
CHTR Q1 2026 Internet Loss (k)
CHTR Q1 2026 Residential ARPU YoY
FWA Share Of CHTR Sub Decline (est.)
Both hypotheses partially true. Hypothesis A (bounded share-take peaking 2026, releasing cable in 2027) is supported by the Verizon slowdown, the 32M capacity ceiling, and the New Street base case of "only slightly negative cable broadband net adds in 2026, positive in 2027." Hypothesis B (structural re-pricing of the bottom quartile that compounds through 2027) is supported by Q1 2026's -1.4% ARPU print — the first negative in 20+ years — and AT&T's just-getting-started Internet Air ramp. The resolution timeline is two to four prints; Q2 2026 (July 24) is the first hard test.
1. Scale — Per-Carrier FWA Subscribers Today
Three carriers run FWA at scale. The cleanest disclosed numbers from FY2025 annual reports, plus an industry-tracker (Tefficient) anchor for T-Mobile's last clean disclosure (June 2025):
T-Mobile no longer cleanly discloses an FWA subscriber line item — its factbook reports "5G broadband" inside customers-per-account postpaid metrics (TMUS FY2025 investor factbook, definitions L819). The mid-2025 7.308M figure is the last clean Tefficient industry-tracker anchor; YE2025 is extrapolated. Verizon's 5.7M and AT&T's 1.5M are directly disclosed in the FY2025 10-Ks. Pre-2024 figures are reconstructed from carrier disclosures and industry trackers and should be treated as directional, not precise to the hundred-thousand.
2. Net Adds vs. Cable Losses — The Comparison That Anchors The Debate
This is the single chart the PM should keep at hand. FWA net adds (TMUS + VZ + AT&T combined) plotted against the sum of Comcast + Charter residential Internet net losses over the same eight quarters. The pattern: FWA growth exceeds cable losses in 2024–2025 (so FWA is taking dollars outside cable too — DSL replacement, mobile-only households, rural new connects), but the gap is narrowing as Verizon decelerates and the cable losses widen.
The cable line aggregates Comcast and Charter residential broadband quarterly net adds (negative = losses). FWA quarterly net adds combine reported VZ FWA, AT&T Internet Air, and the implied T-Mobile FWA quarterly addition (TMUS no longer breaks FWA out cleanly post-2024; values are reconstructed from carrier prepared remarks and industry analyst trackers). Treat the line as directionally accurate to within ~10–15% on a per-quarter basis.
Composition mix, not aggregate slowdown. Combined Big 3 FWA net adds are roughly flat at ~3.8M annual in 2025 vs. 3.7M in 2024 — but the composition shifted materially: Verizon decelerated by ~400K and AT&T accelerated by ~400K. This is exactly what the planner brief flagged as the Hypothesis A vs. B inflection: if Verizon's slowdown represents a hard capacity ceiling, the aggregate must roll over in 2026; if AT&T can grow at +900K to +1.2M for several years (which the EchoStar spectrum suggests is possible), the aggregate stays at ~3.5M+ through 2027.
3. The Bounded vs. Structural Hypothesis — Capacity Math
The bull case rests on a hard number: New Street Research's December 2025 update puts the Big 3 combined FWA capacity ceiling at 32M subscribers (20.6M consumer + 11.8M business), roughly double the firm's June 2024 estimate of 16M. Doubling the ceiling means the headroom debate has shifted: the question is no longer "is FWA capacity-bound at ~16M?" — it is "is the new 32M ceiling expandable through the next round of spectrum, or is that the wall?"
Capacity ≠ profitable demand. The 32M is a physics/spectrum ceiling, not a forecast. New Street's own commentary projects FWA net adds will slow in 2026 even though the ceiling is far away — because the easy DSL-replacement and rural-unserved pools are largely filled. The marginal next FWA sub costs more to acquire and serves a household closer to the cable price-point. The 32M is therefore better read as "no near-term hard wall" rather than "demand has to fill the bucket."
4. Pricing — The Wedge That Broke ARPU In Q1 2026
The most economically important fact about FWA is not its sub count — it is its $50 retail anchor. Charter's blended residential bill is $119.05; Spectrum's broadband ARPU specifically is $70.72 (Q1 2026); the marginal cable-vs-FWA shopper compares against $50 for T-Mobile 5G Home or Verizon 5G Home. That gap is the lever that pushed residential ARPU to -1.4% year-over-year in Q1 2026 — the first negative residential ARPU print in over 20 years.
The decomposition is the analyst's, not Charter's. Charter disclosed two specific reasons for the soft Q1 ARPU print: the $218M reallocation of programmer-app costs against video revenue, and ongoing video/voice volume decay. They did not break out an FWA-attributable ARPU component. The reading above is industry-pattern reasoning: of the residual after disclosed items, the bulk is most plausibly cable holding the lower promotional tier flat (rather than taking the historical +3 to +5% standard annual price increase) precisely because the FWA $50 anchor exists. Two interpretations are possible: a permissible "ARPU growth is mostly accounting noise" read, or a "the actual pricing-power lever is now genuinely impaired" read. The Q2 2026 print, with the $218M reallocation in the prior-year base, is the first clean test.
5. Where FWA Bites — Customer Tier And Geography
The Q1 2026 transcript was unusually explicit about which customers Charter is losing. Winfrey: "The vast majority of that gross add variance came from the low-income segment." And Q3 2025: "low-income consumers increasingly foregoing fixed broadband for mobile." The combination of FWA at $50 (price floor) and pure mobile-only substitution among lower-usage households is concentrated in the bottom quartile of the cable customer base.
The AT&T Internet Air geofence is the under-discussed structural feature. AT&T's annual report (FY2025 Business L60) is explicit: AIA is offered "in areas where AT&T Fiber is not available." That means AIA does not cannibalize AT&T's own $30B fiber build — it attacks rural and edge geographies, which is precisely where Charter has less fiber overbuilder competition. Where AT&T historically lost broadband customers to nobody (DSL decay), AT&T is now competing for those same households with FWA against Charter cable. This is incremental pressure that didn't exist in 2023.
6. Charter's Defense Stack — Specific To FWA
Charter has built four defensive levers that specifically address FWA (distinct from its fiber defenses). None of them reverse the ARPU repricing; all of them slow the share-take and absorb the bottom-end pressure.
Penetration calculated as Spectrum Mobile lines / total Internet customers, rounded. Per-line and per-customer attach rates would be higher in absolute terms because some households carry multiple mobile lines. The directional read is what matters: mobile attach has roughly tripled in four years and is the largest single retention-side investment Charter has made against FWA and fiber.
7. Hypothesis Test — What Would Confirm Each Side
The honest analytical posture is that both sides have evidence and the resolution will come from the next four prints. Below is the explicit watchlist with the specific numerical threshold that would confirm or refute each hypothesis.
8. What This Tab Does Not Claim
Five honest limits worth flagging before a PM acts on the above.
Bottom line for the PM, restated. FWA is real, it is the largest single driver of Charter's broadband sub decline today, and the ARPU repricing it caused is the principal reason CHTR re-rated from $238 to $149 in May 2026. But it is not stranding the cable plant — it is repricing the low end. Charter's defense stack (mobile bundle, Internet Assist, network evolution) is well-matched to the threat and is showing in the slowing of gross attrition. The hypothesis test resolves in 2-4 prints. Q2 2026 (July 24) is the first hard read — that print, with the video-reallocation drag in the prior-year base, tells you whether the -1.4% Q1 ARPU is the new normal or a one-quarter event.