Business

General Motors — what it is, and whether it belongs in the universe

General Motors is a car company — a direct hit on the framework's auto-OEM exclusion, the value trap this reader has paid for before. The universe screens pass cleanly: a primary NYSE listing and a roughly $88 billion market cap, about nine times the $10 billion line. Its market structure is a fiercely competitive global oligopoly with chronic excess manufacturing capacity and no pricing moat, and its China business has turned from profit engine into an equity-method drag.

The universe screen

GM's common stock is registered under Section 12(b) and lists on the New York Stock Exchange under the symbol GM; the company was incorporated as a Delaware corporation in 2009 and is headquartered in Detroit [1]. This is a domestic US primary listing — not a European ADR, not a Chinese ADR — so the geography test (U1) is met without qualification.

Market cap derived from 973 million shares outstanding (FY2025) at the $90.30 close of 2026-07-28; share count per the FY2025 10-K income statement [2].

At $90.30 (2026-07-28) on about 973 million shares, GM is worth roughly $87.9 billion — comfortably clear of the $10 billion floor (U2 met). The market-cap test passes by a wide margin; the exclusion checks below are where GM's fit is decided.

What the business is

GM designs, builds and sells trucks, crossovers, cars and parts, and provides automotive financing. It runs three reportable segments: GM North America (GMNA), its profit engine; GM International (GMI); and GM Financial, its captive lender. Vehicles carry the Buick, Cadillac, Chevrolet and GMC brands, and GM leads the US industry in sales [3]. FY2025 revenue was $167.97 billion, and the company employed roughly 156,000 people — about 88,000 hourly (56%) and 68,000 salaried (44%), with about 47,000 US hourly workers represented by the UAW [4].

Source: FY2025 Annual Report (Form 10-K), Consolidated Income Statements [5]; prior-year figures from reported financials.

Two things stand out from the eight-year record, and both matter to later tabs. Revenue has been range-bound between roughly $109 billion and $172 billion — cyclical, not growing structurally — and FY2025 net income of $2.7 billion collapsed from $6.0 billion in 2024 and $10.1 billion in 2023. The revenue line is stable-to-cyclical; the earnings line is where the recent damage sits, and the anatomy of that fall belongs to the Dislocation and Damage Math tabs, not here.

Segment economics — one engine carries the company

Roughly 87% of GM's vehicle sales are North American, and North America produces almost all of the profit. GMNA generated $10.5 billion of EBIT-adjusted on $154.3 billion of revenue in FY2025 — but that was down 28% from $14.5 billion in 2024, a 6.8% margin against management's stated 8–10% target, with roughly $3.1 billion of tariff cost cited as the swing factor [6]. GMI is small — $13.4 billion of revenue and $0.7 billion of EBIT-adjusted — and GM Financial contributed $2.8 billion of EBIT [7][8].

Source: FY2025 Annual Report (Form 10-K), MD&A segment results [9][10][11]. GM Financial shown at EBIT; segment revenues do not sum to consolidated revenue because of intersegment eliminations.

The economic reality is concentrated: full-size pickups and SUVs sold in the United States and financed through GM Financial. GMI, outside China, is a rounding error against GMNA, and GM's China exposure runs through equity-method joint ventures that sit outside these revenue lines (below).

Market structure — the durability raw material

This is the evidence the Durability tab and the jury lean on, so it is laid out plainly. The US light-vehicle market is a consolidated oligopoly in which GM is the share leader, but leadership is thin and contested, and the industry's own economics work against pricing power.

GM's 17.2% US share (2.85m of 16.6m units) per the FY2025 10-K [12]; competitor US units and shares from the indexed peer filings and industry sales data [13]. Competitor shares are approximate.

Four points, each cited, that a durability jury needs:

It competes on everything except price power. GM's own 10-K describes an industry "characterized by intense competition, evolving regulatory requirements, changing consumer expectations" that "is highly competitive in terms of the quality, innovation, new technologies, pricing, fuel economy, reliability, safety, customer service, and financial services" — an industry that "has historically had excess manufacturing capacity," where "attempts by our competitors to sell more vehicles could have a significant negative effect" on GM's pricing and share [14]. Chronic overcapacity and price-based competition are the antithesis of a monopoly's margin protection.

Share leadership is narrow. GM's 17.2% US share sits barely ahead of Toyota (~15%) and Ford (~13%), with Hyundai-Kia, Honda, Stellantis and Tesla all live in the same market [15]. This is not a duopoly with pricing discipline; it is a seven-way scramble. GM's strength is concentrated in one franchise — US full-size trucks, where it holds about 33% share — not in cars, where its US share fell to roughly 2% [16].

Capital intensity is real, but it is not a moat. GM runs 50 US manufacturing and parts facilities across 19 states, including 11 assembly plants, and announced about $4.0 billion to onshore additional production [17]. Under this framework, capital heft protects an essential if the industry is not oversupplied; here the same 10-K concedes historical excess capacity, so the capital base is a cost to defend, not a barrier that keeps rivals out [18].

Long operating history, discontinuous legal entity. GM has more than a century of brand history but the filing entity was incorporated only in 2009, out of the predecessor's bankruptcy reorganization [19]. The franchise is old; the balance sheet and the going concern are the post-2009 company. Product is essential (personal transportation) and demand is durable; what is not durable is any single automaker's share of it.

Exclusion screen — the checks the corpus settles here

X1 — auto-OEM: a direct hit. GM is exactly the kind of business the framework excludes by name, and does so because these names look cheap on cash-flow yield yet do not compound owner value. GM's FY2025 filing supplies the confirming facts in its own words: a highly competitive industry with historical excess manufacturing capacity, competing on price [20]. This is stated here, plainly, and carried to Fit; it is not softened.

X4 — consensus darling: not a hit. The exclusion of over-owned, extreme-multiple darlings does not fit GM. It trades at roughly 0.5x sales ($87.9 billion market cap on $167.97 billion revenue [21]) and at about a 6x forward earnings multiple against an industry near 20x (consensus data, July 2026) — the profile of an out-of-favor cyclical, not a story stock on a bottom-left-to-top-right chart. The counter-fact, stated fairly: the sell side is not bearish — the average rating is a "Buy" with targets around $100–105 (consensus, July 2026) — so the setup is cheapness the market dislikes owning, not fear the sell side shares. GM is not a darling; whether it is a deserved cheapness is the value-trap question X1 raises and the later tabs test.

S1 — China dependence: material, and now a drag. GM's China business runs through equity-method joint ventures (SAIC-GM/SGM and SAIC-GM-Wuling/SGMW) whose vehicle sales are not recorded in GM's revenue; only GM's share of their results flows through equity income [22]. Those JVs sold 1.88 million units for a 7.1% China share in 2025 (down from 8.4% in 2023) and generated an equity loss of $0.3 billion, including $0.6 billion of SGM restructuring charges [23]. So China is a real sensitivity — a large unit base and a genuine restructuring cost — but it is off-balance-sheet earnings, a shrinking share, and currently a loss rather than a profit pillar. It is flagged, not exclusionary; the listing itself is US, so the Chinese-ADR exclusion does not apply.

Promotional-CEO (X2) and structural-decline (X3) belong to Self-Help and Durability; nothing in the business overview forces either call, and neither is duplicated here.