Damage Math

Damage Math

GM's reported profit collapsed in 2025 — net income fell from $6.0 billion to $2.7 billion and diluted EPS from $6.37 to $3.27 [1]. But the fall was a discrete $7.9 billion EV write-off plus tariffs, not a cut to recurring earning power: adjusted EPS held near $10.60 and is guided to $11–13 for 2026 [2]. The ~29% drawdown has fully reversed; at $90.30 the stock trades ~50% above its pre-event peak. The damage-gap Ruchir hunts has closed.

The near-term hit, quantified

The 2025 hit lived almost entirely on the GAAP line. Reported net income to stockholders dropped to $2.7 billion, its lowest since the pandemic year, and diluted EPS to $3.27 [3]. The single largest driver was the EV strategic realignment: GM recorded $7.9 billion of charges in GMNA in 2025 — $3.2 billion of non-cash impairment plus cash settlement charges — writing down EV-related tooling and equipment to nominal salvage value [4]. The second was tariffs, which compressed the North America profit engine: GMNA EBIT-adjusted fell to $10,452 million from $14,528 million, a margin of 6.8% versus 9.2% [5].

Sources: FY2025 10-K, Consolidated Income Statements [6]; Q2 FY2026 10-Q, segment note [7]; adjusted EPS from consensus (CapIQ).

Source: FY2021–FY2025 10-K Consolidated Income Statements, as reported [8].

The gap between the reported and the underlying number is where the damage math turns. GM entered 2025 guiding EBIT-adjusted of $13.7–15.7 billion, EPS-diluted-adjusted of $11–12, and adjusted automotive free cash flow of $11–13 billion [9]. In April 2025 it cut EBIT-adjusted to $10–12.5 billion, naming a current tariff exposure of $4–5 billion [10]. That is the near-term numerator: a roughly $3.5 billion midpoint cut to one year's operating profit, tariff-driven. GM then delivered $12.7 billion of EBIT-adjusted for 2025 — above the top of that revised range and down only 15% from the 2024 record of $14.9 billion [11]. The recurring earning power did not step down with it — adjusted EPS came in near $10.60, and for 2026 GM guides EBIT-adjusted back to $13–15 billion, adjusted EPS $11–13, and adjusted automotive free cash flow $9–11 billion [12].

Metric	FY2025 guide (Jan 2025)	Revised (Apr 2025)	FY2025 actual	FY2026 guide (Jan 2026)
EBIT-adjusted	$13.7–15.7B	$10.0–12.5B	$12.7B (actual)	$13–15B
Adjusted EPS	$11–12	—	~$10.60	$11–13
Adjusted auto FCF	$11–13B	—	—	$9–11B
Reported net income	—	—	$2.7B	—

Sources: Q4 FY2024 call [13]; Q1 FY2025 call [14]; Q4 FY2025 call [15]; FY2025 10-K [16].

Consensus tells the same story from the sell side: normalized EPS of $10.60 for FY2025 rises to $13.36 (FY2026) and $14.67 (FY2027), and consensus free cash flow clears $9.9–11.7 billion across FY2026–27 — a rising, not a cut, trajectory. This is the feature that breaks the Centene analogy: there, EPS was cut by two-thirds; here, the reported number fell by half while the underwritable number was flat and is climbing.

The price change over the same window

The market's reaction was a genuine drawdown, but a shallow one by Ruchir's standard, and it is long gone. GM peaked at $60.20 on 25 November 2024, troughed at $42.48 on 8 April 2025 — a 29.4% fall over 134 days — and now trades at $90.30, a full recovery charted below. The volume signature was modest: the peak-to-trough leg carried only a 1.6x volume spike over the prior six-month median — pressure, not the emotion-driven capitulation the framework's fear gauge looks for.

Source: daily price history (month-end closes); intramonth peak $60.20 (25 Nov 2024) and trough $42.48 (8 Apr 2025) per the capitulation gauge — derived from fit_features.capitulation_gauge.

The two moves belong side by side. Reported EPS fell about 49% in 2025; adjusted EPS fell roughly 10% against the original guide and was flat year-on-year; peak-to-trough market value fell about 29% — roughly $21 billion, from a ~$68 billion peak to ~$47 billion at the low. Since then the market cap has risen to $87.9 billion — up about $41 billion from the trough and about $20 billion above the pre-event peak. GM's automotive book runs roughly net cash — consensus net cash of about $7.5 billion — so market cap is a fair proxy for automotive enterprise value; the drawdown and recovery are an equity-value story, not a leverage one.

The NPV arithmetic, conservatively

The question the framework asks: under conservative assumptions, how much of the NPV of future cash flows could this hit plausibly destroy, and how does that compare with what the price destroyed? A transparent, no-growth model answers it. Normalized adjusted automotive FCF post-event is about $10 billion (2026 guide midpoint, and consistent with $9.3–10.0 billion reported in 2023–24). Pre-event, GM delivered $14 billion of adjusted automotive FCF in 2024 and guided $11–13 billion for 2025, so the durable level shift the problem introduced is bounded at roughly $2 billion per year (central) to $4 billion per year (against the record year). Discount at 10% — a fair cost of equity for a cyclical automaker — with no perpetual growth.

Source: two-scenario DCF-lite; annual hit and $10B normalized FCF base from FY2024–FY2026 guidance [17] [18]; discount rate 10%, zero growth — computed, workings in text.

The workings are on the page: a temporary hit of $2 billion for three years discounted at 10% is $2B × 2.487 = $5.0 billion of NPV; at $4 billion it is $9.9 billion. A permanent $2 billion level shift is $2B ÷ 0.10 = $20 billion; at $4 billion it is $40 billion. Against the ~$21 billion of market value that actually evaporated peak-to-trough, these NPV losses bracket the reading. At the April 2025 low, the price damage sat at the low end of the permanent range and roughly four times the central temporary case — the market was pricing the hit as essentially permanent. That was the moment the framework's setup could have been live.

At $90.30, consensus forward free cash flow of about $9.9 billion is an 11.3% yield on the current market cap — above Ruchir's 10% moderate-balance-sheet bar, as the Yield tab computes. But the yield is being earned with the fear already gone and the price above its starting point, not at a moment of maximum dislocation — the entry trigger the Dislocation tab tests is no longer live.

The trial — temporary or permanent, presented fairly

Whether the impairment is temporary or permanent was tried by two opposing corpus-cited briefs and ruled on by independent blind judges. Both cases are strong.

The temporary case at its strongest. The 2025 collapse was a discrete, largely non-cash write-off, not lost profitability: the impairment "include[s] the cost of writing down EV-related tooling and equipment to its nominal salvage value," and cash outflow in 2025 was only $400 million [19]. GM states it has "substantially completed the recognition of material cash charges related to our EV strategic realignment" [20] — a terminating charge is the definition of a temporary earnings effect. The retail truck/SUV franchise was untouched, 2026 guidance is back to record-adjacent levels [21], and tariffs — a policy variable — are already partly reversed, with the net 2026 EBIT-adjusted impact guided to $2.5–3.5 billion after a $0.5 billion IEEPA refund [22].

The permanent case at its strongest. A higher adjusted year can coexist with lower NPV when it is achieved by abandoning projects and leaning on mature profit pools. Tariffs reset the cost base — GM "do[es] not expect such actions to fully offset the impact of tariffs in the near term" [23], and the impact recurs at $2.5–3.5 billion into 2026 [24]. The EV assets were written down, not delayed, with a further $3.4 billion of net charges in H1 2026 and additional charges still expected [25]. China was impaired as an "other-than-temporary" loss in value [26], and Cruise's robotaxi option was wound down entirely [27]. The remaining profit pool is more mature and price-sensitive, resting on full-size ICE trucks in a market where incentives may yield "vehicle prices that do not offset our costs" [28].

The ruling. The judges put the probability the impairment is temporary at 0.63, with a per-judge range of 0.44–0.66 and a spread of 0.22 — not flagged contested, but not a consensus either: one judge (reading the permanent brief first) landed at 0.44, leaning the other way. Order stability was reasonable (temporary-first mean 0.63 versus permanent-first mean 0.55, an 0.08 gap). The read the report carries is a moderate lean toward temporary — the discrete EV charge and the fading, partly-refunded tariff both wind down — tempered by the genuinely permanent pieces (China marked other-than-temporary, Cruise exited) that lower NPV regardless of the adjusted-EPS rebound. This diagnosis probability is the trial's, and it stands as the report's; the analysis here only names the drivers.

Which line broke, and whether it self-corrects

Four drivers broke; they do not self-correct at the same rate.

EV realignment is the largest and the most self-correcting: $7.9 billion in 2025 was mostly a non-cash write-down of tooling to salvage plus one-time supplier/JV settlements, and GM calls the material cash charges substantially completed [29] — though it still expects some additional charges, and $3.4 billion more landed in H1 2026 [30].

Tariffs are a policy cost, not a structural one: net 2026 impact of $2.5–3.5 billion after mitigation and a $0.5 billion IEEPA refund [31], fading from the $4–5 billion gross exposure named in early 2025 [32]. It reprices as trade policy settles, but GM concedes mitigation will not fully offset it near-term [33].

China and Cruise are the pieces that do not come back on their own: the China equity interests were impaired as an other-than-temporary loss in value [34], and the Cruise robotaxi option was wound down rather than paused [35]. Their recovery would require new growth GM has not yet disclosed, and their loss is why the 2026 rebound rests more heavily on the mature full-size ICE franchise [36] — a narrower base whose durability the Durability tab weighs.