Ahead of Consensus.
Intelligence across tech and capital markets, for investors, policymakers, and business leaders.
9 minute read
There is a particular kind of earnings report that does not so much surprise investors as correct their premise. AT&T’s second quarter, released July 22, was one of those. The stock had drifted to a 52-week low earlier in the month, weighed down by a narrative that had calcified into consensus: that Starlink’s expanding ambitions would eventually erode the economics of terrestrial wireless, and that AT&T, lacking a satellite partnership of its own, stood most exposed. Wells Fargo had gone as far as initiating coverage with a Sell rating and an $18 price target, a rare bearish outlier among major bank research desks.
What AT&T delivered instead was a quarter that argued, in the plain language of a financial statement, that the underlying business is stronger than the debate around it. Revenue reached $31.6 billion, up 2.3 percent from $30.8 billion a year earlier. Adjusted earnings per share came in at $0.65, ahead of the $0.59 consensus and a marked improvement from $0.54 in the prior year. Adjusted EBITDA rose 5.2 percent to $12.3 billion, and cash from operations climbed to $10.8 billion, funding free cash flow of $4.7 billion. Shares responded with a modest but telling rally, trading up roughly 2.6 percent to $22.84 in the hours after release. It was not a dramatic reversal. It was something more useful to a company mid-transition: a quiet reassertion of credibility.
The most instructive numbers in this release sit inside a segment AT&T calls Advanced Connectivity, the reporting structure it adopted at the start of the year to isolate its 5G and fiber businesses from the copper network it is retiring. Service revenue in that segment rose 5.1 percent to $23.5 billion. Operating income rose 20.3 percent to $7.3 billion. That gap between revenue growth and profit growth is the story. It suggests a business finally harvesting the returns on infrastructure it spent years building, rather than one still absorbing the cost of building it.
The customer numbers back this up with unusual consistency. AT&T added more than 1 million Advanced Connectivity customers during the quarter, split across 367,000 fiber net adds, 279,000 fixed wireless net adds, and 432,000 postpaid phone net adds, with churn held to a tight 0.86 percent. Management described it as the company’s strongest quarter for postpaid wireless account growth in more than three years, and its best combined fiber and fixed wireless quarter on record. Fiber now reaches 38.6 million locations, putting AT&T on pace for 40 million by year end and 60 million by 2030, a target that increasingly leans on the mass markets fiber business acquired from Lumen Technologies, which closed in February.
The figure that ties these threads together is convergence: the 42.5 percent of households taking AT&T’s fiber or fixed wireless internet who also subscribe to AT&T wireless. This is not a vanity metric. Converged customers churn less, cost less to retain, and generate more revenue per household, which is precisely why AT&T has spent the better part of a decade building network reach specifically to sell services together rather than separately. CEO John Stankey called the quarter validation of that approach, arguing the company’s fiber scale and network depth are advantages rivals cannot quickly replicate. It is the kind of claim executives make every quarter. This time, the segment results gave it some weight.
Every transition has a ledger, and AT&T’s shows up most clearly in its Legacy segment, the copper-based voice and data business the company is actively decommissioning. Legacy revenue fell 25.9 percent year over year, and segment operating income dropped to $523 million, down $436 million from a year earlier. This is not decline by neglect. It is decline by design, part of a stated plan to power down the large majority of the domestic copper network by the end of 2029.
That plan explains an otherwise puzzling line in the consolidated results: operating expenses rose slightly to $24.5 billion from $24.3 billion, despite the company shedding an entire category of legacy cost. The increase reflects a one-time asset abandonment charge tied to a reprioritization of spectrum strategy, along with higher advertising spend and integration costs from the newly acquired fiber subscribers, all partially offset by fully depreciated legacy assets rolling off the books. In practical terms, AT&T is running two businesses inside a single income statement: a shrinking, still cash-generative legacy operation on a fixed exit timeline, and a capital-intensive growth business whose profitability is only now becoming visible in the numbers. Investors reading the headline expense line without this context would draw the wrong conclusion.
AT&T’s balance sheet reveals a company still willing to spend to secure its position. Total debt stood at $144.0 billion at quarter end, with net debt of $126.4 billion, and that figure has further to climb. The pending $23 billion all-cash acquisition of spectrum from EchoStar, covering roughly 30 MHz of 3.45 GHz mid-band and 20 MHz of 600 MHz low-band licenses across more than 400 markets, is expected to close by mid-2026, having cleared the FCC with a $2.4 billion escrow condition tied to EchoStar’s Boost Mobile obligations. Management has been candid that leverage will rise before it improves, guiding that net debt to adjusted EBITDA will return to its roughly 2.5 times target only within about three years of the deal closing.
That spending is inseparable from the Starlink question shadowing the stock. Analyst opinion remains genuinely split. Wells Fargo’s Steven Cahall sees AT&T as the carrier most exposed to satellite-based competition, while Bernstein has argued the opposite, that AT&T’s fiber scale gives it a more insulated position than peers. Morgan Stanley, Barclays, and Scotiabank each trimmed price targets in July, to $25, $24, and $29.25 respectively, without abandoning their underlying ratings, a pattern that reads less as conviction than as caution pending more evidence. The EchoStar spectrum, once deployed, is meant to reinforce AT&T’s fixed wireless offering and expand capacity for next-generation applications, giving the company a way to compete on network depth rather than price.
AT&T reiterated its full 2026 to 2028 outlook without alteration, including adjusted earnings per share of $2.25 to $2.35 for 2026, free cash flow rising from $18 billion this year to $21 billion by 2028, and capital investment held near $23 billion to $24 billion annually. It also reaffirmed plans to return more than $45 billion to shareholders over the period, holding its $1.11 annualized dividend while accelerating 2026 buybacks to roughly $10 billion, a pace increase that signals confidence in the durability of current cash generation rather than a defensive gesture.
The honest reading of this quarter is that AT&T has proven its convergence strategy can produce operating leverage, with segment profitability now growing meaningfully faster than revenue. What remains unproven is whether that leverage will outrun the capital intensity of the EchoStar transaction and the still-unsettled trajectory of satellite competition. That is not a question this quarter answers. It is one the market will keep pricing, one quarter at a time, for the remainder of the decade.