Damage Math

Damage Math

PayPal's per-share price fell 57% from its January 2025 peak to the February 2026 trough and sits 39% below peak today, erasing roughly $40 billion of market value. Near-term earnings barely moved: FY2026 consensus EPS is flat and revenue still grows about 3.5%. The destruction is in the out-years — 2027–2028 consensus EPS was cut 11%–15%. Whether that reset is temporary or permanent is what decides if a gap exists; the trial panel put the probability it is temporary at 0.38, and split.

The near-term hit — a small numerator

The trigger was the Q4 FY2025 print on February 3, 2026: non-GAAP EPS of $1.23 missed consensus by 4.4%, revenue of $8.68 billion missed by 1.3%, management guided 2026 EPS "flat-to-down versus 2025's $5.31," and the board replaced CEO Alex Chriss with Enrique Lores the same day [1]. None of that is a collapse in current earning power. GAAP net income actually rose from $4,147 million (FY2024) to $5,233 million (FY2025), and diluted EPS from $3.99 to $5.41, on operating income up from $5,325 million to $6,065 million [2].

Source: consensus estimates (FY2026 revenue $34.3B vs FY2025 actual $33.2B; FY2026 normalized EPS $5.31 vs FY2025 actual $5.31; Q4 FY2025 normalized EPS $1.23 vs $1.29 consensus); FY2025 actuals per the FY2025 10-K [3].

The real move in the estimate record is one year further out. Between the 180-day-ago vintage (before the trigger) and now, consensus normalized EPS was cut 10.9% for 2027 ($6.45 to $5.75) and 14.8% for 2028 ($7.33 to $6.25); the revenue line was cut 4.1% for 2027 and 7.9% for 2028. That is the numerator that matters — not the flat 2026, but a lower long-run slope.

Source: consensus estimate revisions, 180-day vintage vs current (data feed: data/sp/estimates.json momentum series).

The price and value change over the same window

Per-share price fell from $91.81 (January 17, 2025) to $39.08 (February 12, 2026), a 57.4% drawdown, and has since recovered to $56.15 (July 24, 2026) — still 39% below peak. Translated to enterprise value: market cap fell from roughly $95.4 billion at peak (1,039 million shares at year-end FY2024) to about $37.8 billion at the trough (968 million shares), and stands at $54.4 billion today [4]. Corporate net debt is modest (consensus FY2025 net debt near $2.0 billion), so EV tracks market cap closely. The buyback retired 7% of the float in FY2025 — $6,052 million of repurchases against $5,564 million of free cash flow ($6,416 million operating cash flow less $852 million of capital expenditure) — so total-cap destruction (43% peak-to-today) slightly exceeds the per-share move because the share base shrank [5].

Source: derived — price series peak/trough/current (drawdown gauge, fit_features.capitulation_gauge) times shares outstanding near each date (fit_features.share_count_trend); share and buyback figures per the FY2025 10-K cash-flow statement [6].

Stated side by side: consensus 2027–2028 EPS fell 11%–15% and the long-run revenue slope fell about 4 points; the market cap fell 43% from peak (57% at the trough). The question is whether a 4-point slope cut can arithmetically justify a 43% value cut.

The NPV arithmetic, in two scenarios

The workings are deliberately simple and reproducible. Take a normalized free-cash-flow base of $6.0 billion — between FY2025 reported FCF of $5.56 billion and the FY2026–2028 consensus band of $5.8–6.5 billion. (On an owner-earnings basis this base is lower: reported FCF does not deduct the $1.0 billion of stock-based compensation charged in FY2025, which the Yield tab carries; the level scales both scenarios below equally, so it does not change the gap logic.) Value a growing perpetuity with the Gordon formula, V = FCF × (1 + g) / (r − g), and ask what the near-term hit does to g.

Temporary scenario. Long-run growth is unchanged; the hit is one-to-two years of flat FCF while management reinvests, then the prior trend resumes. FY2026 consensus FCF of about $5.8 billion sits roughly $0.7 billion below a $6.5 billion trend. Two years of a ~$0.7 billion shortfall, discounted at 9%, destroys about $1.2 billion of NPV; stretch it to three years and it is under $2 billion. Call the temporary NPV damage $1–3 billion.

Permanent scenario. The near-term hit reflects a durable reduction in the long-run growth rate — branded checkout, over half of profit dollars, decelerating from mid-single digits toward 1%–2% and being diluted by lower-yield volume. Model that as a cut in g and read the NPV loss off the table below.

Source: derived — Gordon growth model on a $6.0B normalized FCF base; NPV hit = FCF×(1+g_pre)/(r−g_pre) − FCF×(1+g_post)/(r−g_post). Values are sensitive as g approaches r; an r=8% column (not shown) roughly doubles each figure.

A permanent 1.5–2.0 point growth cut destroys roughly $22–45 billion of NPV, centering near $30 billion at a 9%–10% discount rate. That is comparable to the $40 billion the market cap has lost from peak and within reach of the $57 billion peak-to-trough stress.

Source: derived — price damage from market-cap change above; NPV hits from the temporary and permanent scenarios in this section.

The trial, both sides at their strongest

The temporary-versus-permanent question was argued by two opposing corpus-cited briefs and ruled on by three independent judges.

The ruling is contested. The three judges put the probability the impairment is temporary at 0.38 overall — individual reads of 0.38, 0.38, and 0.62 — with a spread of 0.24 between the extreme seats and a 0.12 order-effect gap between reading sequences. The panel leans toward the permanent reading, but not decisively, and one seat leans the other way. The diagnosis the report carries is that 0.38, not a rounded-up verdict. What would move it toward temporary: online branded checkout returning to high-single-digit currency-neutral growth for two-plus consecutive quarters (first tested at Q2/Q3 FY2026), FY2026 transaction-margin dollars growing mid-single digits against the guided flat-to-down, and 2027 consensus EPS reversing back above roughly $6.45. What would move it toward permanent: branded checkout turning negative, transaction-margin dollars declining year-over-year, or take rate compressing more than 10 basis points on price rather than mix.

Which line broke, and whether it self-corrects

The driver behind the hit is specific: online branded checkout, roughly 30% of volume but more than half of profit, decelerating to 1%–2% currency-neutral growth [17] [18]. The bull mechanism of recovery is repricing and product: management is reinvesting in a checkout experience it concedes was "stagnant and underinvested," and Q1 FY2026 showed a sequential improvement to +2% with the U.S. firming [19] [20]. The bear mechanism against it is structural and embedded: a decade of take-rate compression (to 1.62%) as growth migrates to lower-yield PSP and Venmo volume, and a management team that withdrew its own multi-year outlook because the underlying assumptions did not hold [21] [22]. Both mechanisms are live in the record; the Durability tab weighs which dominates, and the Dislocation tab anatomizes the drawdown itself.