Best Broadband Quarter in Eight Years, Overshadowed by a Flat EPS Line and a $23 Billion Spectrum Bill: Maintaining AT&T at Hold

Results vs. Consensus

Q3 2025 Scorecard

Year-Over-Year Comparison

Sequential (QoQ) Comparison

Quality of Print

Revenue: The 1.6% top-line growth is a deceleration from Q2's 3.5%, but the mix is intact: Mobility service revenue up 2.3% (dampened ~60 basis points by a $90M one-time administrative-fee benefit in the prior-year quarter), Consumer Wireline up 4.1% on 16.8% fiber growth, and Business Wireline still the drag at down 7.8%. The slight consensus miss is a headline event, not a quality event; the growth engines grew and the legacy declined, exactly as modeled.

Margins and EPS: Adjusted EBITDA grew 2.4% with 30 basis points of margin expansion, but adjusted EPS was flat at $0.54 as higher depreciation and the below-the-line items offset the operating growth. The GAAP $1.29 is inflated by the DIRECTV sale gain and should be ignored. The cleaner story is cash: $10.2B of operating cash flow and $4.9B of free cash flow after a $400M pension contribution, with management reiterating a low-to-mid $16B full-year free-cash-flow figure despite a ~$500M Q4 drag from out-of-pattern legal settlements accrued in Q3.

The KPI standout: The broadband result is the number that matters most for the long-term thesis. Over 550,000 total broadband net adds is the best in eight-plus years, and it came with a 65.8% jump in Consumer Wireline operating income and 350 basis points of EBITDA-margin expansion. If there is one figure to carry out of this quarter, it is that the fiber-plus-fixed-wireless broadband machine is operating at a level AT&T has not seen since before the WarnerMedia era.

Segment Performance

Communications Segment Overview

Mobility — Volume Holds, Margin and ARPU Drift by Design

Mobility delivered 405,000 postpaid phone net adds (up slightly year over year) with service revenue up 2.3% and total revenue up 3.1%. But the margin and ARPU lines both softened: EBITDA margin contracted 40 basis points to 44.7%, postpaid phone churn rose 14 basis points to 0.92%, and ARPU fell 0.8% to $56.64. Management was explicit that the ARPU decline is deliberate, a function of leaning into underpenetrated segments (such as a 55-and-older plan) and converged customers who take a discounted wireless line while generating broadband revenue that lands in Consumer Wireline.

"I would describe what's going on in ARPU more as a feature, not a bug. When we talk to you about the fact that we're underpenetrated in certain segments… and… our desire to push convergence, which at the front end… means that we give the customer a square deal and a lot of value." — John Stankey, Chairman & CEO

Assessment: The "feature, not a bug" framing is analytically fair, converged customers churn less and carry higher household lifetime value, but it also means Mobility's reported metrics (ARPU, segment margin) will understate the true economics for as long as AT&T is investing in the convergence mix shift. For a valuation-sensitive investor, this converts Mobility from a clean margin-expansion story into a "trust the household economics" story, which is harder to underwrite and is a core reason we hold rather than upgrade.

Consumer Wireline — The Best Quarter of the Cycle

Consumer Wireline is where AT&T is winning outright. Revenue grew 4.1% on 16.8% fiber revenue growth, EBITDA grew 15.1% with 350 basis points of margin expansion to 36.3%, and operating income jumped 65.8%. The unit added 288,000 fiber and 270,000 Internet Air subscribers, its best combined broadband quarter in over eight years, and crossed 10 million total fiber subscribers. Fiber ARPU rose 4.4% to $73.48. Convergence within the fiber base reached 41.5%, up 180 basis points, one of the largest gains in three years.

"We achieved our highest total broadband net adds in eight years… This includes a major milestone by reaching over 10 million premium AT&T Fiber subscribers, more than doubling our fiber customer base in less than five years… and the train keeps rolling." — John Stankey, Chairman & CEO

Assessment: This is the engine of the entire thesis, and it is running at full speed. The 350-basis-point margin expansion validates the fiber-scale-plus-copper-retirement operating-leverage story more decisively than any quarter to date. The one caution is seasonality: management flagged that Q4 fiber adds will step down on holiday seasonality, and the prior-year Q4 was flattered by post-work-stoppage pent-up demand, so the sequential comparison will look softer.

Business Wireline — Still Declining, but the Growth Sub-line Is Accelerating

Business Wireline revenue fell 7.8% and EBITDA fell about 13%, with an operating loss of $354M. The encouraging detail is beneath the headline: fiber and advanced-connectivity service revenue grew 6.0%, an acceleration from 3.5% in Q2, as management reinvests cost savings into the growth sub-line. Legacy and transitional services continue to decline 17.3%.

Assessment: The acceleration in advanced-connectivity revenue (from 3.5% to 6.0% growth) is the first tangible sign that the Business Wireline reinvestment is producing a growth offset rather than just a slower decline. It remains a managed-decline segment, but the trajectory of the growth sub-line is the metric to watch for evidence that the enterprise unit can eventually stabilize.

Key Operating Metrics

Key Topics & Management Commentary

Overall Management Tone: Assured and internally focused, with a recurring "we have all the assets we need; now we execute" refrain. Management repeatedly pivoted questions about external threats (industry overbuild, competitor promotions, LEO satellites, CEO succession) back to AT&T's own controllable execution, and declined to engage on succession entirely. The confidence is grounded in the operating results rather than promotional, but the tone also carried a note of deliberate insulation from the strategic questions the market is asking about the $23B spectrum commitment and the leverage path.

1. Record Broadband and the 10 Million Fiber Milestone

The quarter's defining achievement was operational: over 550,000 total broadband net adds, the most in more than eight years, and passing 10 million fiber subscribers. AT&T has more than doubled its fiber base in under five years and nearly tripled quarterly fiber revenue over the same span. The company passed more than 31 million total fiber locations and reiterated the path to 60 million-plus by 2030.

"We had another solid quarter… over 550,000 new subscribers to our most advanced broadband services… This resulted in our highest total broadband net adds in more than eight years." — John Stankey, Chairman & CEO

Assessment: This is unambiguously bullish for the long-term thesis and confirms the fiber flywheel is not just intact but accelerating. It is also the single strongest argument the bulls have for looking through the flat EPS line to the subscriber base being built. Our reservation is timing, not direction: these adds convert to earnings over years, and the stock has to be valued on the earnings the base produces, not just the base itself.

2. The $23B EchoStar Spectrum Acquisition

AT&T detailed the EchoStar transaction announced in August: roughly 30 MHz of nationwide 3.45 GHz mid-band and 20 MHz of nationwide 600 MHz low-band spectrum for approximately $23B in cash, closing in the first half of 2026 and financed with cash on hand plus incremental borrowings. It has already begun deploying the 3.45 GHz spectrum under a short-term lease and expects mid-band coverage across nearly two-thirds of the U.S. population by mid-November, positioning Internet Air to expand faster in 2026.

"The EchoStar spectrum we agreed to acquire will improve our 5G wireless performance in a cost-efficient manner while allowing us to grow Internet Air at a faster pace… we expect these mid-band licenses will be deployed in cell sites covering nearly two-thirds of the U.S. population by mid-November." — John Stankey, Chairman & CEO

Assessment: Strategically sound, financially a two-year drag. The spectrum genuinely strengthens the network and accelerates Internet Air, and buying capacity is cheaper than building it. But the deal takes net leverage to roughly 3.0x, is neutral to adjusted EPS and free cash flow for 24 months, and only turns accretive in year three. This is precisely the unbudgeted spectrum call we flagged as a risk at the Q2 initiation, and it converts the near-term capital-allocation story from "growing returns" into "digesting a big acquisition." It is the central reason we do not upgrade on the strong operating quarter.

3. Convergence Deepens: Fiber and Fixed Wireless Both Over 50% Attach

The convergence rate within the fiber base reached 41.5% (up 180 basis points), and management disclosed that more than half of Internet Air subscribers also take AT&T wireless, with fiber-household mobility attachment now above 50% even before meaningful marketing behind the newly acquired spectrum. Converged customers carry the lowest churn and the highest lifetime value in the base.

"When you look at… the mobility attachment associated with [fixed wireless]… across our footprint, we are at 50%. We're better than 50%. And that's really before any meaningful marketing put behind it… that's another pool of value that we're really excited about." — Pascal Desroches, CFO

Assessment: The convergence data is the best evidence that the strategy compounds, and the disclosure that attach rates are already above 50% pre-marketing is a genuine positive-skew data point. This is the mechanism that should eventually pull churn down and improve acquisition efficiency, which is management's stated path to Mobility margin expansion. It supports the long-term case; it does not change the near-term earnings math.

4. Wireless Competition and the "Feature, Not a Bug" ARPU

Postpaid phone churn of 0.92% and the 0.8% ARPU decline framed the wireless discussion. Management characterized the competitive environment as persistently elevated, said churn normalized as the quarter exited, and reiterated that it plans for seasonal Q4 switching. The ARPU pressure was defended as a deliberate mix choice, with a December pricing action expected to support Q4 service revenue.

"I actually don't believe many of my peers walk into their job and say my goal is to lose share… I think most CEOs want to win… We've demonstrated that we can be successful against all those tactics." — John Stankey, Chairman & CEO

Assessment: The competitive read is candid, all three carriers are fighting, and AT&T is holding its own. But the combination of rising churn, falling ARPU, and contracting Mobility margin, even if each is individually explainable, adds up to a wireless business that is contributing volume rather than profit growth. This is the maturing-wireless bear point we set at initiation, and it is playing out.

5. "We Have All the Assets": The End of the M&A Chapter

Asked directly whether AT&T is done with inorganic M&A after the EchoStar and Lumen deals, the CEO signaled a decisive pivot to organic execution, framing everything outside the company as "external and distraction."

"We have all the assets in front of us. We've run the plays that we need to run to be successful over the next five years, and everything that's going on outside of our business right now is external and distraction… our job now is to organically invest in this business and make it a better company." — John Stankey, Chairman & CEO

Assessment: This is what shareholders want to hear after nearly $29B of deal commitments: no more large M&A, and a focus on penetrating and monetizing the assets already secured. If credible, it caps the balance-sheet risk and sets up the deleveraging path from 3.0x back toward 2.5x. We treat it as a genuine intent while noting the CEO explicitly declined to say "never."

6. Capital Returns and the Path Back to 2.5x Leverage

AT&T returned $3.5B to shareholders in Q3, including nearly $1.5B of buybacks, keeping it on pace for the $4B full-year target under a $20B 2025–2027 repurchase capacity. Net leverage improved to 2.59x pre-EchoStar; post-close it rises to roughly 3.0x, with management targeting a return to the 2.5x area within about three years, aided by DIRECTV proceeds ($3.8B still to come) and growing free cash flow.

Assessment: The pre-deal deleveraging and the reiterated buyback pace are constructive, but the post-EchoStar leverage bump means the balance-sheet story goes sideways for two-plus years before resuming its improvement. The $20B three-year repurchase capacity is real optionality, but it now competes with the spectrum bill and the Lumen close for the same cash, which tempers how aggressively returns can accelerate near term.

7. Lumen Fiber Deal: Diligence at the Asset Level

Management reiterated confidence in the $5.75B Lumen Mass Markets fiber acquisition (closing in 2026), citing hard-asset-level diligence and Gigapower out-of-region data confirming the converged economics. AT&T will hold the assets in a "NetworkCo" subsidiary and plans to sell a partial stake to a co-investing equity partner within 6–12 months of close. The deal is expected to be immaterial to adjusted EBITDA, EPS, and free cash flow for 12–24 months, then accretive long-term.

Assessment: Bringing in an equity partner for NetworkCo is a sensible capital-light structure that limits the balance-sheet impact of the fiber expansion. The out-of-region Gigapower data supporting the underwriting is a reasonable confidence signal. Like EchoStar, the accretion is a multi-year story; near-term it is neutral at best.

8. LEO Satellites: Complementary, Not a Replacement

On the competitive threat from low-earth-orbit constellations (direct-to-device and fixed broadband), the CEO framed LEO as mostly complementary to AT&T's terrestrial network, citing the physical capacity limits of satellite spectrum versus dense terrestrial cell sites, and pointed to AT&T's AST SpaceMobile relationship as its direct-to-device path.

"It's really hard to see 40 megahertz spectrum over a 20-mile radius replacing [a cell site's] capacity… I do believe they can be really complementary… When you have dense fiber and you can pick up workloads closer to the customer, you're always going to have a better performing network." — John Stankey, Chairman & CEO

Assessment: The capacity-physics argument is sound and the AST partnership positions AT&T as an integrator rather than a victim of LEO. This is a reasonable framing of a real but often-overstated competitive concern; it is not a near-term thesis driver either way, but it de-risks a tail worry the market occasionally raises.

Guidance & Outlook

AT&T reiterated its full-year 2025 guidance across the board and narrowed the adjusted-EPS outlook to the higher end of $1.97–2.07. The details management added were the mechanical ones investors need to model the fourth quarter: Q4 capital investment of roughly $7.0–7.5B (implying the seasonal back-loaded spend), Q4 free cash flow of about $4B (net of a ~$500M out-of-pattern legal-settlement drag accrued in Q3), and Q4 depreciation and amortization of about $5B, which management flagged as the run-rate to carry into 2026 as certain legacy assets finish depreciating.

Longer-term framework: the 2026–2027 outlook was reiterated, low-single-digit consolidated service revenue growth, 3%-or-better adjusted EBITDA growth, adjusted EPS accelerating to double-digit growth in 2027, and $23–24B annual capital investment. Critically, management said it will refresh the long-term outlook early in 2026 to fold in the EchoStar and Lumen accretion, describing both deals as boosts to organic revenue and profitability. That update is the next real catalyst: it is where the market will learn how much the two deals add to the 2027-and-beyond earnings trajectory.

The unchanged tension: the earnings acceleration is still a 2027 event, and the two acquisitions now sit between here and there as a two-year period of neutral-to-dilutive digestion. The dividend (~4% yield) and the $4B buyback anchor the total return, but the near-term profile is steady rather than inflecting, and the balance sheet takes a step backward before it resumes improving.

Analyst Q&A Highlights

Industry-Wide Fiber Overbuild and the 60-Million Target

The opening question raised the striking arithmetic that announced industry fiber plans roll up to roughly 110 million homes in a country of ~135 million, and asked at what point AT&T investors must worry about insurgents reaching planned passings first. The CEO argued that announced plans are not built plans and that AT&T's scaled build engine, brand, and deliberate market signaling deter overbuilders.

Q: "Our best attempt to estimate the intentions of all the fiber expanders… rolled up is about 110 million in a country with 135 million homes… at what point do AT&T investors have to worry about insurgents getting to some of the homes that AT&T plans to pass before you do?"
— Peter Supino, Wolfe Research

A: "While I know there's a lot of announcements out there that may add to 110 million homes being built, it doesn't mean they're getting built… our job is to remove that friction and be better than everybody else and ensure the 60 million that we're building are in fact the first… when somebody overbuilds a small portion of a metropolitan area, this is a scale business. Having 230,000 homes passed isn't going to cut it."
— John Stankey, Chairman & CEO

Assessment: The distinction between announced and built passings is the right rebuttal, and AT&T's scale and brand are genuine advantages against fragmented overbuilders. But the sheer volume of announced fiber does cap the terminal penetration and ARPU the market will underwrite on the deeper passings, which is exactly the load-bearing assumption we are watching. Management's confidence is credible; the risk is real but not near-term.

Mobility Margin Expansion Against a Persistently Competitive Market

An analyst pressed on how AT&T delivers the multi-year Mobility margin expansion embedded in its guidance if the competitive environment stays as elevated as it has been all year. The CFO grounded the answer in structural efficiency programs (copper retirement, wireless modernization by 2027) and the churn-lowering effect of convergence rather than in a competitive-environment forecast.

Q: "You've mentioned the competitive environment in wireless this year… Your three-year guidance assumes… margins expand next couple of years. Wondering if you could talk a little bit about how you deliver that if… the competitive environment maybe stays this elevated?"
— Benjamin Swinburne, Morgan Stanley

A: "With each passing day, we have less and less copper in the network… we're in mid-flight in modernizing our wireless network. We expect that to be substantially complete by 2027… the more convergence we drive, the overall churn should come down. And as a result, the efficiency of our acquisition spend should also improve."
— Pascal Desroches, CFO

Assessment: Notably, the margin-expansion case rests on cost structure and convergence, not on the competitive environment easing. That is the right way to underwrite it, because management cannot control competition, but it also means the margin expansion is a 2026–2027 back-half story tied to the modernization finishing. For the near term, wireless margins are more likely to be stable-to-soft than expanding, consistent with our Hold.

Wireless Positioning if Q4 Promotions Intensify

With leadership changes at both major competitors, an analyst asked whether AT&T is positioned for a more promotional fourth quarter and how the device-financing cohorts are trending. The CEO pushed back on the premise that a new competitor CEO implies a strategy shift, arguing all carriers already compete to win.

Q: "How would you say the company is positioned if we see higher promotional activity in the fourth quarter given the changes at Verizon and T-Mobile? And maybe touch on the cohorts coming off plan?"
— John Hodulik, UBS

A: "Just because there's a change at the top, I don't know that that suggests to me that there's going to be a 180-degree posture change… our competitors have been pretty aggressive, and they've tried to win… We've demonstrated that we can be successful against all those tactics. And if there's a recalibration… we're going to adjust to that."
— John Stankey, Chairman & CEO

Assessment: The measured answer is reassuring, AT&T is not betting on competitors easing up, and it has held share through a competitive year. But the honest read is that the wireless market is a share-stable, promotionally intense equilibrium in which everyone spends to hold position. That is a fine environment for a disciplined operator to defend its base; it is not one that generates the earnings acceleration a Hold-to-Outperform upgrade would require.

Is AT&T Done With Inorganic M&A?

Given the Lumen and EchoStar deals, the leverage target, the dividend, and the buyback, an analyst asked directly whether AT&T is out of the acquisition game and now a purely organic story. The CEO gave the clearest "we have what we need" answer of the call.

Q: "Is it fair to say that when you say that you wouldn't trade assets with anybody, that you don't need any more assets? That AT&T is out of the M&A acquisition game… and now it's time to build on what you have organically?"
— David Barden, New Street Research

A: "Never going to answer a question absolutely and say never. But… we have all the assets in front of us… our job now is to organically invest in this business… Those plays all sit in front of us and are all contained within the four walls of AT&T, and they don't require uncertain regulatory approvals or difficult external issues or other partners to get it done."
— John Stankey, Chairman & CEO

Assessment: This is the answer that caps the balance-sheet risk, and it matters for the leverage path. If management holds to organic-only, the return to 2.5x leverage and the resumption of return growth become credible. We take it as genuine intent, tempered by the explicit refusal to say "never," and treat any new large deal as the event that would move us to a more cautious stance.

EchoStar and Lumen Accretion Buckets

An analyst asked whether AT&T's confidence in the accretion from the two deals has grown, and to break down the EchoStar value drivers. The CEO detailed capital deferral (denser spectrum defers capacity augments), an improved wholesale relationship (Boost/EchoStar network-as-a-service), and Internet Air scaling.

Q: "Has your confidence around accretion from the Lumen Fiber assets and the EchoStar spectrum licenses increased…? And… the key buckets in terms of the EchoStar spectrum accretion…?"
— Michael Ng, Goldman Sachs

A: "There's the old-fashioned way that accretion is driven… it's going to defer some capital because of the depth we get in the network… We have a better wholesale play… I can see… they're migrating a lot of customers over to our network right now… and then… the scaling that's going on in Internet Air."
— John Stankey, Chairman & CEO

Assessment: The three accretion levers, capex deferral, wholesale revenue, and Internet Air scale, are real and internally consistent, and the early wholesale customer migration is a concrete confirmation. But management would not quantify any of it, deferring the numbers to the early-2026 outlook update. Until that update lands, the accretion is a qualitative promise against a quantified $23B cash outlay, which is why the deal reads as a near-term drag despite its long-term logic.

What They're NOT Saying

1. Quantified EchoStar/Lumen accretion. Management described the accretion levers in detail but attached no numbers to any of them, deferring the quantification to an early-2026 long-term outlook update. For a combined ~$29B of deal commitments, the absence of a dollar or per-share accretion figure is the single biggest gap in the story.
2. A hard leverage-path schedule. Net leverage goes to ~3.0x post-EchoStar and returns to ~2.5x "within approximately three years," but management did not lay out the year-by-year path or the buyback cadence around it, leaving the trade-off between deleveraging and returns unspecified.
3. Where wireless ARPU and margin actually settle. The "feature, not a bug" framing explains the ARPU decline but sidesteps the question of where ARPU and Mobility margin stabilize once the convergence mix shift matures, which is what would let an investor model the segment rather than trust it.
4. CEO succession. Asked directly about the board's succession plan and whether the CEO would hand off to the president, management declined to engage at all ("we're not worried about your question"), a notable non-answer given the industry-wide leadership churn the questioner referenced.
5. 2026 as a standalone year. Management reiterated the 2027 EPS acceleration and the 2026–2027 framework but gave no 2026-specific EPS or free-cash-flow figure, leaving the near-year, the one that includes the deal-digestion drag, unquantified until the early-2026 update.

Market Reaction

• Pre-print setup: T closed at $26.05 on October 21, entering the print up 14.4% year to date but down 9.9% over the trailing 30 days, having faded from a $28.92 close a month earlier as the market digested the $23B EchoStar commitment and a higher-leverage story. The S&P 500 was up 14.5% YTD, so T had given back its earlier relative lead.
• Reaction session (October 22): the stock gapped down to $24.92 at the open (−4.3%) on the slight headline miss and the flat EPS line, traded a $24.32–$25.83 range, and closed at $25.55, down 1.9% (−$0.50) on the day.
• Volume: 95.5M shares, 2.6x the 30-day average, a heavy-volume repricing session.

The tape tells the story of a market weighing an outstanding operating quarter against a flat earnings line and a bigger balance sheet. Unlike Q2, when the stock recovered intraday to close green, this time the initial gap-down held and the stock closed down 1.9% despite the record broadband result. The difference is context: the stock had already slid nearly 10% over the prior month on the EchoStar overhang, and a slight consensus miss plus a flat EPS line gave the sellers a reason to press.

We read the reaction as fair rather than an overreaction. This was not a quarter where a great result was punished by fickle positioning; it was a quarter where genuinely strong subscriber metrics could not offset the reality that adjusted EPS did not grow and the company has just committed $23B that pushes leverage the wrong way for two years. A stock that trades on cash flow and balance-sheet trajectory reasonably paused on a print that advanced the subscriber base but not the near-term earnings or the deleveraging path. That balance, strong operations, unchanged near-term financials, is why we maintain Hold rather than either upgrade on the KPIs or downgrade on the miss.

Street Perspective

Debate: Does the Record Broadband Quarter Justify Looking Through the Flat EPS?

Bull view: The best broadband adds in eight years, 10 million fiber subscribers, and a 41.5% convergence rate are building a base that will produce years of high-margin revenue. Flat EPS in a quarter with this much subscriber momentum is a timing artifact; the earnings follow the base with a lag, and the operating leverage in Consumer Wireline (350bp of margin expansion) is the proof.

Bear view: A stock is valued on the earnings a base produces, not the base itself, and the earnings line has been flat for a year while ARPU falls and wireless margins compress. The subscriber growth is real but low-ARPU and increasingly discounted; it is not obvious the flywheel converts to the EPS acceleration management keeps promising for 2027.

Our take: The bull case is directionally right on the long term and the bear case is right on the near term, which is the definition of a Hold. We need to see the subscriber base convert to free-cash-flow-per-share growth, not just subscriber growth, before paying up. The early-2026 outlook update is where that conversion gets quantified.

Debate: Is the $23B EchoStar Deal Value-Creative or a Leverage Mistake?

Bull view: Buying nationwide mid-band and low-band spectrum for ~$23B is cheaper than the capex to add equivalent capacity, defers future network spend, accelerates Internet Air, and improves the Boost wholesale relationship. A temporary move to 3.0x leverage for a franchise-strengthening spectrum position is a reasonable trade, and management has a track record of deleveraging.

Bear view: $23B of cash and debt for spectrum that is EPS- and FCF-neutral for two years is a large opportunity cost that competes directly with the buyback and dividend, sets deleveraging back to 3.0x, and adds integration and deployment execution risk, all for accretion that management has not yet quantified.

Our take: Strategically defensible, near-term dilutive to the equity story. The spectrum strengthens the network, but the two-year neutral-to-dilutive window and the leverage step-up are why the deal, on balance, argues for patience rather than an upgrade. The accretion quantification in early 2026 will settle this debate.

Debate: Can Wireless Margins Expand From Here, or Is This the Plateau?

Bull view: As copper retires and the wireless network modernization completes by 2027, structural cost comes out, and rising convergence lowers churn and acquisition cost. Margins expand in 2026–2027 independent of the competitive environment.

Bear view: Churn is rising, ARPU is falling, and Mobility margin contracted again this quarter. The competitive equilibrium is promotionally intense and share-stable; the cost of holding subscribers keeps climbing, and margin "expansion" keeps getting pushed to next year.

Our take: Lean bear near term, open to the bull case by 2027. The margin-expansion story is entirely a cost-and-convergence story tied to programs that complete in 2027; nothing in this quarter's wireless metrics suggests near-term expansion. We do not underwrite wireless as an earnings engine over the next four quarters.

Model Implications & Valuation Framework

Valuation framework: At the $25.55 reaction close and a ~$2.05 FY2025 adjusted-EPS estimate, T trades at roughly 12.5x this year's earnings with a dividend yield near 4.3%, a modestly cheaper multiple than at our July initiation because the stock has pulled back on the EchoStar overhang while estimates held. That is not expensive, but the de-rating reflects a real change: the balance sheet now carries a $23B spectrum bill and moves to 3.0x leverage, and the near-term earnings are flat. A low-teens multiple is appropriate for a business in a two-year digestion phase with a 2027 earnings inflection ahead.

Price-target framework (12-month):

• Base case: $27–29. ~13x a $2.15 FY2026 adjusted-EPS midpoint, a modest multiple recovery as the market gains confidence in the deleveraging path. With the ~4.3% yield, that is a high-single-digit to low-double-digit total return, broadly market-like. This is the core of the Hold.
• Bull case: $32–33. The early-2026 outlook update quantifies meaningful EchoStar/Lumen accretion, the market re-rates toward 15x on visibility into the 2027 acceleration, and convergence-led broadband margins keep expanding.
• Bear case: $22–23. Wireless margins keep slipping, the leverage path proves slower than the ~3-year target, a promotional Q4 pressures adds, and the multiple stays at ~11x with no re-rating.

The base case implies roughly 6–13% price upside plus the yield, against a bear case of roughly −10–14%. As at initiation, the risk/reward is balanced rather than asymmetric, and the S&P-relative return does not justify Outperform. The stock is a touch cheaper than in July, but the offsetting negative, a two-year balance-sheet digestion, leaves the call unchanged.

Thesis Scorecard Post-Earnings

We grade this quarter against the standing thesis established at the Q2 2025 initiation. The commitments we flagged to watch, accelerating H2 broadband adds, seasonal (not structural) churn, sustained free cash flow, and buyback pace, were largely delivered.

Overall: Thesis unchanged. The bull pillars strengthened on the record broadband quarter and deepening convergence, but the bear pillars escalated in exactly the way we flagged: wireless margins are drifting and the balance sheet took on a large spectrum commitment that consumes the free-cash-flow upside for two years. The net is a wash for the near-term risk/reward.

Action: Maintain Hold. Management executed well operationally and delivered on last quarter's commitments, which keeps us constructive on the long-term franchise. But the EchoStar-driven leverage step-up and the flat earnings line leave the near-term risk/reward balanced. The key catalyst to watch is the early-2026 long-term outlook update, where quantified EchoStar/Lumen accretion and a clearer 2027 EPS-acceleration bridge could support an upgrade. We would move to Outperform on evidence that free-cash-flow-per-share is inflecting and the deleveraging path is credible; we would move to Underperform on a clear deterioration in wireless margins or a slippage in the leverage plan.