Business

Business — AT&T Inc. (NYSE: T)

AT&T is the third of the three US national wireless carriers and one of the country's largest fiber-broadband builders: FY2025 revenue of $125.6B, operating income of $24.2B, and 120 million mobility subscribers, run over an FCC-licensed spectrum-and-fiber network [1] [2]. It clears both universe tests — a US-listed common stock at roughly $165B of market value — and it is neither an auto maker nor a high-multiple market darling. China dependence is effectively absent.

The universe screen

Ruchir's system only looks at US-listed stocks (or European companies with a US-listed ADR) above $10B of market cap. AT&T passes both, plainly.

Sources: market cap from fit_features.market_cap_usd (7.179B shares at the $22.96 close of 23 Jul 2026); EPS of $3.04 from the FY2025 10-K Consolidated Statements of Operations [3].

• Geography and instrument (U1): AT&T Inc. is a Delaware-incorporated holding company whose common stock trades on the New York Stock Exchange under the ticker T; it is a domestic US issuer, not a Chinese company and not an ADR [4]. The screen is clean.
• Market cap (U2): at the 23 July 2026 close of $22.96 on 7.179 billion shares outstanding, market value is about $164.8B — more than sixteen times the $10B line [5]. Third-party market data over 22–24 July 2026 put the figure in a $150–165B range; either way the test is passed comfortably.

Market cap is fit_features.market_cap_usd; share count from the FY2025 10-K, price from the daily feed dated 23 Jul 2026. Listing venue cross-checked against current market data (NYSE: T), July 2026.

What the company is, and how it makes money

AT&T sells connectivity. It carries voice and data over a nationwide wireless network and delivers internet over a fiber and copper wireline network, charging consumers and businesses recurring monthly service fees plus one-off equipment (handset) sales. Two sentences for a cold reader: AT&T is a US telecommunications carrier that earns most of its money selling wireless phone service to 120 million subscribers and fiber broadband to households and businesses. It is one of three national mobile networks in a capital-heavy, FCC-licensed industry, and its revenue is almost entirely earned in the United States, with a small wireless operation in Mexico.

The company reports two segments — Communications and Latin America. Communications supplied about 97% of 2025 segment operating revenues and substantially all segment operating income; Latin America (wireless service in Mexico) was roughly 3% of revenue and under 1% of operating income [6] [7]. Communications breaks into three business units:

• Mobility — nationwide wireless service and handsets; the profit engine.
• Business Wireline — fiber, ethernet, IP voice and managed services to businesses, alongside a shrinking book of legacy copper-based voice and data.
• Consumer Wireline — residential broadband (fiber and the AT&T Internet Air fixed-wireless product) plus legacy home phone.

Source: FY2025 10-K, MD&A Segment Results — Mobility, Business Wireline and Consumer Wireline detail [8]. Communications (Mobility + Business Wireline + Consumer Wireline) totals $120.9B; consolidated revenue of $125.6B includes Latin America and Corporate & Other eliminations.

Mobility dominates: it earned $89.5B of revenue in 2025, up 5.0%, on 120 million subscribers — 91 million postpaid (74 million postpaid phone), 18 million prepaid and 11 million reseller — with postpaid ARPU still rising and monthly postpaid-phone churn in the low single digits, per the segment operating metrics [9] [10]. Consumer broadband is the growth story management is investing behind: 10.4 million fiber connections (up 1.1 million in the year) inside 16.0 million total broadband connections [11]. The two segments split the map cleanly: the United States for Communications, Mexico for Latin America (about 24.7 million subscribers under the AT&T and Unefon brands) [12].

Scale

Source: FY2025 10-K — revenue, operating income and net income from the Consolidated Statements of Operations [13]; headcount of approximately 133,030 from the Human Capital disclosure [14].

Total operating revenue was $125.6B in 2025, up 2.7%, after two essentially flat years — $122.4B in 2023 and $122.3B in 2024 [15]. Earlier years are not comparable: AT&T spun off WarnerMedia in April 2022 and separated its DIRECTV video business over 2021–2025, so pre-2023 consolidated revenue included businesses the company no longer owns [16]. On the continuing base, revenue is flat-to-rising, not declining — the three-consecutive-years-of-decline disqualifier that the Durability tab tests is not triggered at the consolidated level.

Roughly 43% of employees are represented by the CWA, IBEW or other unions, with several contracts expiring in 2026 — a recurring source of operational and cost risk [17].

Market structure — the raw material for durability

US wireless is a facilities-based oligopoly. Three carriers own national networks — AT&T, Verizon and T-Mobile — supplemented by regional providers, resellers (MVNOs) and cable companies reselling wireless over those same three networks [18]. The three carriers are close in scale, which is what an oligopoly looks like from the outside: AT&T at $125.6B of revenue sits between Verizon ($138.2B) and T-Mobile ($88.3B).

Sources: AT&T FY2025 10-K [19]; peer revenue and market cap from each company's latest reported financials and current market data (as reported).

Four structural features make year-10 revenue and cash flow more forecastable than in most industries — the evidence the Durability tab and the jury build on:

• Regulatory entry barriers. A wireless carrier must be licensed by the FCC to transmit on specific spectrum bands in defined geographies; licenses run a fixed 10–15 years and must be renewed, and the FCC controls how much spectrum exists to license at all [20]. A new entrant cannot simply appear; it must win scarce, auctioned spectrum. This is the "regulator does not let a garage startup take share" barrier in Ruchir's framework.
• Capital intensity as a moat. The balance sheet carries $131.6B of property, plant and equipment and $128.1B of FCC licenses, and the company spends on the order of $20B a year on capital [21]. Replicating a national network is a multi-decade, multi-hundred-billion-dollar undertaking.
• Essential product. Mobile and broadband connectivity is close to a utility; low single-digit monthly postpaid-phone churn reflects how sticky the service is once sold [22].
• Long operating history. The legal entity traces to the 1984 Bell System breakup, when it was spun off from AT&T Corp as Southwestern Bell / SBC, later renaming itself AT&T after the 2005 merger — more than four decades as an independent public carrier, on top of the century-old Bell franchise it descends from [23].

Two qualifiers keep this honest. First, broadband is a different structure from wireless: in fixed internet AT&T competes market-by-market mainly against large cable operators (Comcast, Charter) and against the other carriers' fixed-wireless offerings — closer to regional duopolies than a tidy national oligopoly [24]. Second, part of the business is in genuine secular decline: legacy copper voice and data is shrinking every year, and Business Wireline revenue fell 8.4% in 2025 to $17.2B with the segment now running an operating loss [25]. The durable core is wireless plus fiber; the legacy tail is a managed run-off. Whether the growing core outruns the shrinking tail is the question the Durability tab carries.

Leverage is worth flagging here because it colours how essential the balance sheet makes the business: total debt was $136.1B against $18.2B of cash at year-end 2025, so net debt is about $117.9B, roughly 2.6 times an EBITDA of about $45.0B (operating income of $24.2B plus $20.9B of depreciation and amortization) [26]. That is a moderately levered balance sheet, not a fortress and not distressed; the Yield and Self-Help tabs carry what the leverage does to the yield bar and to refinancing.

First-pass exclusion screen

Auto-OEM (X1) — not applicable. AT&T manufactures nothing automotive; it is a telecommunications carrier whose only hardware sales are handsets it resells to its own subscribers [27]. The car-company value-trap screen does not apply.

Darling positioning (X4) — clears. The exclusion targets high-growth names on extreme multiple-to-sales where consensus already owns the story. AT&T is the opposite. At $22.96 it trades at about 7.6 times trailing EPS of $3.04 and roughly 1.3 times sales ($164.8B market cap on $125.6B revenue) — mid-pack among its peers (Verizon ~1.3x, T-Mobile ~2.1x, the cable operators lower) and nowhere near darling territory [28]. The sell side is constructive but not euphoric: 26 analysts carried a consensus "Buy" with an average target near $28.90 in late July 2026. This is a low-multiple, dividend-oriented telecom, not a bottom-left-to-top-right growth story.

Multiples derived from the FY2025 10-K and the 23 Jul 2026 price; peer multiples from current market data and each peer's latest revenue. Consensus rating and target from third-party analyst aggregation, July 2026 (as reported).

China dependence (S1) — effectively absent. AT&T's revenue is earned in the United States and Mexico; the Latin America segment (Mexico only) is about 3% of segment revenue, and no single customer accounts for 10% or more of consolidated revenue [29] [30]. The filings disclose no material China revenue or China-based assets; the only China linkage is ordinary supply-chain exposure to network and device vendors, which the 10-K does not quantify as a dependence. The sensitivity flag is not raised.