Damage Math

The damage and the price

Fiserv's FY2025 adjusted-EPS guidance was cut 16.4% on 29 October 2025 — $10.225 to $8.55 at the midpoint [1] [2]. Market value fell $101.3 billion from the March 2025 peak to 29 July 2026. Capitalised conservatively, the cash the reset removed is worth $3.4bn to $50.7bn. The gap is real and large. The trial put the probability the impairment is temporary at 0.61, and called the result contested.

The near-term hit, quantified

The numerator of this tab is a guidance path, not a modelled estimate. Fiserv guided fiscal 2025 four times, and the four numbers are dated and filed.

memory access out of bounds

Sources: FY2025 outlook as first guided at 10% to 12% and $10.10 to $10.30, 5 Feb 2025 [3]; outlook refined, 23 Jul 2025 [4]; outlook reset to 3.5% to 4% and $8.50 to $8.60, 29 Oct 2025 [5]; FY2026 outlook of 1% to 3% and $8.00 to $8.30, 10 Feb 2026 [6], with the FY2025 adjusted earnings per share of $8.64 [7]. Bars show guidance midpoints; FY2025 actual as reported.

On 5 February 2025 the company guided 2025 organic revenue growth of 10% to 12% and adjusted EPS of $10.10 to $10.30 [8]. On 23 July it refined that to approximately 10% and $10.15 to $10.30 [9]. On 29 October it cut to 3.5% to 4% and $8.50 to $8.60 [10]. Management framed it as a level change rather than a quarter: "Fiserv's growth and margin targets need to be reset," driven by four named factors — slowing cyclical growth in Argentina, recalibration of optimistic growth assumptions, the impact of deferred investments, and deprioritisation of short-term revenue and expense initiatives [11]. Fiscal 2025 then landed at $8.64, above the reset range [12], and fiscal 2026 was guided to 1% to 3% organic and $8.00 to $8.30 [13].

Two hit sizes follow, and they differ:

• Against the guidance in force, the reset removed $1.675 of adjusted EPS from fiscal 2025, a 16.4% cut.
• Against realised earning power, adjusted EPS went $8.80 (FY2024) to $8.64 (FY2025) to $8.15 guided (FY2026) — down 7.4% over two years [14].

Cash fell further than earnings. Company-defined free cash flow went $5,233m in 2024 to $4,435m in 2025 — down 15.3% [15] — and consensus carries $3,880m for 2026, 25.9% below the 2024 level (data/sp/estimates.json, vintage 30 July 2026). Any account of this drawdown that quotes only the 7% EPS decline understates what happened to cash.

The consensus record at the moment of the event is the sharpest single reading. On the Q3 2025 print, adjusted EPS came in at $2.04 against a consensus of $2.646 — a 22.9% miss — and revenue at $4,919m against $5,348m, an 8.0% miss (data/sp/estimates.json:beat_miss). Since the reset, the direction has changed: FY2027 consensus EPS has drifted from $9.15 (31 Jan 2026) to $8.93 now, down 2.3%, while FY2028 has risen from $9.75 to $10.02, up 2.8% (data/sp/estimates.json:momentum). The vendor's revision history in this run begins on 31 January 2026 and covers only FY2027 and FY2028, so there is no pre-trigger consensus vintage for FY2026 to cite; the guidance path above is the primary-document substitute.

Price and enterprise value

memory access out of bounds

Source: daily closing prices, data/prices/daily.json; peak, trough and current dates per fit_features.capitulation_gauge.

The reset day itself: the shares closed at $126.17 on 28 October 2025 and $70.60 on 29 October, on 103.5 million shares against 5.8 million the session before. On the 537,851,887 shares outstanding at 24 October 2025 [16], market value went from $67.86bn to $37.97bn — $29.89 billion in one session. The same day's guidance cut removed $1.675 of adjusted EPS, or about $901m of annual adjusted earnings at that share count. The one-day loss of market value is 33 times the annual earnings amount withdrawn. The volume anatomy of that session and the two legs around it sit in Dislocation.

The peak-to-current move on the whole enterprise:

memory access out of bounds

Peak market cap: 554,433,824 shares at 18 Apr 2025 [17] times the $237.79 peak close. Current market cap from fit_features.market_cap (549.0m shares, $55.67, 29 Jul 2026). Net debt: total debt of $1,239m current plus $27,758m long-term at 31 Dec 2025, less $798m cash [18] [19]; prior-year comparatives $1,110m and $23,730m from the same debt note, and $1,236m of prior-year cash from the consolidated balance sheet [20].

Equity value fell $101.3 billion, or 76.8%. Enterprise value fell $96.7 billion, or 62.2%. The difference matters: net debt rose $4.6 billion over the window, and equity is now 52% of enterprise value, so a given percentage move in operating value arrives at the equity magnified. Comparing the equity fall with an equity-level NPV estimate is the consistent pairing, and that is what follows — but the leverage is the reason the equity number is the larger one.

Two further readings sit inside the phase table. First, the largest single leg — $237.79 to $126.17, down 46.9% — happened entirely before any guidance change, with the fiscal-2025 range still at $10.15 to $10.30. That leg is a multiple compression from 23.3 times the guidance in force to 12.3 times, on unchanged numbers. Second, the 21.1% decline since the reset has run against a fiscal 2025 that beat the reset range, a fiscal 2026 guide reaffirmed in February and May, and a Q1 2026 adjusted EPS that beat consensus by 13.5% (data/sp/estimates.json:beat_miss). Of the 76.6% total price decline, the portion coincident with the earnings damage is the 44.0% reset day.

The NPV arithmetic

The question is how much of the present value of future cash the reset plausibly destroyed. The workings below are deliberately simple and fully visible.

Assumptions, stated. Cash flow measure: company-defined free cash flow, which is levered (operating cash flow is after interest), so it discounts to equity value and pairs with market capitalisation. Discount rate: 9% on equity, with 8% and 10% shown as sensitivities. Terminal growth in the permanent case: 3%. No terminal-value modelling of the business itself — only of the shortfall, which is what the event changed.

Step 1 — size the annual cash shortfall. Two anchors, because the answer depends on what counts as "before":

• Anchor A, realised. FY2024 free cash flow of $5,233m less FY2026 consensus free cash flow of $3,880m = $1,353m a year [21].
• Anchor B, the withdrawn algorithm. The February 2025 guidance implied adjusted EPS growth of 15% to 17% [22]. Compounding FY2024's $5,233m at 15% for two years, holding conversion constant, gives a FY2026 path of $6,921m. Less the $3,880m consensus = $3,041m a year. This anchor capitalises growth management has since withdrawn, so it is the version most favourable to the permanent reading.

Step 2 — capitalise each anchor two ways. Temporary: the shortfall runs three years and then the prior cash level is restored, so the damage is a three-year annuity. Permanent: the shortfall never reverses and grows at 3%, so the damage is shortfall divided by (9% − 3%), a 16.7 times multiple.

memory access out of bounds

Derived: shortfall times the 9% present-value annuity factor (2.531 for three years, 3.890 for five) for the temporary cases, and shortfall divided by 0.06 for the permanent case. Inputs are the FY2024 and FY2025 free-cash-flow reconciliation [23] and FY2026 consensus free cash flow from the vendor estimates file.

Step 3 — the gap.

memory access out of bounds

Derived from the tables above: price damage from the enterprise-value bridge; NPV damage from the two-anchor grid at a 9% discount rate.

Against the most damaging cell in the grid — a permanent level shift measured against the withdrawn growth algorithm, $50.7bn — the equity price damage of $101.3bn is $50.6 billion larger, a ratio of 2.0 to 1. Against the more defensible permanent case anchored on realised FY2024 cash, $22.6bn, the gap is $78.7 billion, a ratio of 4.5 to 1. The temporary cases leave gaps of $93.6bn to $97.8bn.

Discount rate does not close it. At 8% the permanent anchor-B damage is $60.8bn (gap $40.5bn); at 10% it is $43.4bn (gap $57.8bn). Closing the gap entirely on the anchor-B shortfall requires capitalising $3.04bn a year at 33 times — a spread between discount rate and growth of 3.0%, for example 9% and 6% in perpetuity. On anchor A it requires 75 times.

The same arithmetic run backwards gives the plainest version. At $55.67 the equity is worth $30.56bn (fit_features.market_cap). Discounted at 9% with no growth, that price implies sustainable free cash flow of $2.75bn a year — 29% below the FY2026 consensus of $3.88bn and 38% below the $4,435m Fiserv actually generated in 2025 [24]. At 3% growth the implied figure falls to $1.83bn, 53% below the 2026 consensus. The yield side of that computation is worked in Yield.

Three qualifications belong next to the gap. The peak multiple of 23 times was set on guidance that management later attributed in part to an Argentine inflation windfall — organic growth excluding Argentina was roughly 6% in FY2023, 6% in FY2024 and 3% year-to-date 2025, against headline rates of 12%, 16% and 5% [25] — so part of the "value" the price gave up was never underwritten by durable cash. That is why both anchors here start from realised free cash flow rather than from the peak market value. Second, the fit_features market capitalisation uses 549.0 million shares; the most recent filed count is 533,253,986 at 1 May 2026 [26] and 2026 guidance assumes a weighted average of approximately 530 million [27], which would put market value at $29.5bn and widen the gap slightly. Third, fit_features cannot compute the framework's adjusted free cash flow at all — stock-based compensation is missing from the cash-flow feed for every year from 2018 to 2025 — so the cash figures here are company-defined free cash flow, not FCF less SBC less average acquisition spend.

Temporary or permanent

The temporary-versus-permanent question was argued by two opposing briefs, each restricted to corpus citations, and ruled on by three judges reading in different orders.

Sources — temporary: adjusted EPS reconciliation [28]; Argentina contribution [29]; One Fiserv investment step [30]; Project Elevate [31]; second-half bridge [32]; trough and Q1 detail [33]; FY2025 cash conversion [34]; 2029 profile [35]; the two post-reset beats from data/sp/estimates.json:beat_miss. Permanent: four causes [36]; ex-Argentina growth [37]; margin path [38] [39] [40]; Clover revenue expectations and the Q4 fee-elimination headwind [41] [42]; One Fiserv risk language [43]; Financial Solutions detail [44]; core survey [45]; Clover litigation [46].

The ruling

memory access out of bounds

Source: ruchir/trial/tally.json, generated 31 July 2026. Panel figure is the recorded p_temporary; per-judge figures as filed.

The panel put the probability the impairment is temporary at 0.61, on a per-judge range of 0.38 to 0.62 — a spread of 0.24, with a simple mean of 0.54. The tally records the result as contested, and the reading order left a mark: the judge who read the temporary brief first landed at 0.62, while the two who read the permanent brief first averaged 0.495, a gap of 0.125. Both readings are live, and this tab does not pick between them.

The tally's own quote checks discipline both briefs. On the permanent side, the Investor Day core-satisfaction exhibit was recorded as discredited as used: the same slide that shows 2.59 for Fiserv-sold DNA also shows 4.03 for partner-sold DNA — the highest score in the survey, above both named peers, on the identical product — which points at a sales-channel failure rather than product decay [47]. On the temporary side, the brief's "two CEOs in 17 months" rebuttal was recorded as partly unsupported and chronologically reversed: Georgakopoulos was named co-president effective 1 December 2025, after the reset, and the Q1 2026 beat was reported on 5 May 2026, six weeks before he became chief executive. The tally also notes that the brief omitted the Jana Partners activist campaign and the departure of the chief executive who presented the 2029 targets one month after presenting them.

The named flip conditions on both sides converge on the same near dates: whether Q2 2026, reported 6 August 2026, is the trough; whether second-half 2026 adjusted revenue reaches the 6%-to-8% bridge rather than landing below 5%; whether free cash flow conversion holds near 90%; and whether FY2027 adjusted EPS guides above $8.30. The timing of those tests is worked in Clock.

Which line broke

Revenue did not break. Fiscal 2025 adjusted revenue was $19.8bn, up 4%, and consensus carries $20.0bn for 2026 — up 1.2% (data/sp/estimates.json). What broke is margin. Adjusted operating margin ran 39.4% in 2024 [48], fell 200 basis points to 37.4% in 2025 [49], and is guided to approximately 34% for 2026 [50]. Roughly 5.4 points of margin on about $20bn of revenue is about $1.08bn of operating income — the same order as the approximately $600m of people-related and $400m of technology investment the company sized and labelled "Fully in Place" [51].

memory access out of bounds

Source: analyst driver consensus, data/sp/va.json, consensus vintage 24 July 2026; 12 to 20 contributing brokers per line. FY2025 figures are the consensus-recorded actuals.

memory access out of bounds

Source: analyst driver consensus, data/sp/va.json, consensus vintage 24 July 2026; 16 to 19 contributing brokers per line on FY2025 to FY2027 and 9 to 10 on FY2028.

Financial Solutions is the segment that turned negative. Its organic growth goes from plus 1.6% in 2025 to minus 1.1% in 2026, and its operating margin from 45.6% to 41.0% — 455 basis points on about $9.6bn of revenue, roughly $436m. Merchant Solutions decelerates but does not invert: plus 6.1% to plus 3.3%, with margin down 312 basis points on about $10.5bn, roughly $327m. Inside Financial Solutions, Banking is the persistent negative — minus 2.9% in 2025, minus 2.0% in 2026 — and consensus has it recovering only to plus 1.6% in 2027. Inside Merchant, Processing has now run at about minus 4.2% for two consecutive years.

The mechanism that would self-correct. Three of the four components are level effects or dated comparisons rather than lost business. The One Fiserv spend is a one-time step into the cost base, sized and complete, against which the company sets approximately 34% as the leverage baseline [52]; Project Elevate then adds $500m-plus of run-rate cost reduction and 200 basis points-plus of margin by 2029 [53]. The single largest driver of the second-half 2026 revenue reacceleration is a plus-3.5% lapping of a non-recurring headwind — an expiry with a date, not a win [54]. And the Banking attrition is described by management as self-inflicted: it "is the result of actions taken over the last several years and especially around the client service front", with the decision to support all cores having "removed a significant amount of perceived pressure for them to switch" [55]. The core-satisfaction survey supports that reading rather than a product one: 4.03 for partner-sold Fiserv DNA against 2.59 for Fiserv-sold DNA, on the same core [56]. The medium-term plan asks Financial Solutions for only 2% to 4% growth, with total company 4% to 6% [57], and holds free cash flow conversion at approximately 90% with capital expenditure normalising from approximately 8.8% of adjusted revenue toward approximately 8% [58].

The structural case against. Banking organic revenue has been negative in both 2025 and 2026 consensus and returns only to plus 1.6% in 2027; Merchant Processing has been at about minus 4.2% for two years. Core account counts fell 2% year on year in Q1 2026 while overall accounts and positions including Finxact grew 6%, and Financial Solutions segment margin was 38.1% against 47.5% a year earlier [59]. The 10-K states that the expected benefits of the plan may not be achieved on the anticipated timeline "or at all" [60]; Standard and Poor's revised the corporate credit rating outlook from stable to negative on 5 November 2025 [61]; and a securities class action alleges that statements about Clover's growth were false or misleading over a July 2024 to July 2025 class period [62].

Two of the framework's exclusion tests bear on this evidence and neither is met at the company level. Total revenue is not in decline — fiscal 2025 adjusted revenue grew 4% and consensus carries growth in each of 2026, 2027 and 2028 — so the three-consecutive-years-of-high-single-digit-decline disqualifier does not apply, and the segment-level erosion in Banking and Merchant Processing is stated here as what it is: two lines shrinking inside a company that is not. The durability of the franchise on the longer clock is worked in Durability; the balance sheet that has to carry the reinvestment is in Self-Help.

The gap, restated

The near-term hit is real and measurable: a 16.4% cut to the guidance in force, a 7.4% decline in realised adjusted EPS from 2024 to the 2026 guide, and a 25.9% decline in free cash flow from 2024 to the 2026 consensus. Capitalised at 9%, the present value of that shortfall is $3.4bn if it is temporary and lasts three years, and $22.6bn to $50.7bn if it is permanent, depending on whether the pre-reset growth algorithm is treated as value that existed. The price took out $101.3bn of equity value and $96.7bn of enterprise value. The gap between price damage and the most damaging plausible NPV damage is $50.6 billion; against the central permanent case, $78.7 billion. Whether that gap is a mispricing or a correct reading of a franchise still eroding is the question the trial answered at 0.61 probability temporary, contested.