Visa Inc. NYSE: V

Strip out the shrinking share count and $402.61 asks Visa's total owner earnings to compound +17.0%/yr for a decade — a pace Visa has never sustained (its best four-year run, FY2021→FY2025, delivered +10.3%/yr, off a pandemic-depressed base). Credit the shrink Visa has actually demonstrated — diluted share count down 6.33% in FY2025 alone, $21.4B of repurchases in the TTM — and the required pace falls to +9.6%/yr, below the delivered record. The entire investment question is therefore one question: does the toll — net revenue per dollar of payments volume, 28.2 basis points in FY2025 — hold while stablecoins, account-to-account rails, and a settlement that ends honor-all-cards all push on it? That is the contested driver, and the Oct 27 print carries its first dated test.

🐂 Bull case. The toll booth compounds on its own: payments volume +8–10%/yr in constant dollars, cross-border ex-intra-Europe +12%, value-added services (VAS) +34% in Q3 FY2026, and a management team that guides low-double-digits and then raises. At the credited bar, the buyer needs less than the demonstrated past.

🐻 Bear case. Every leg of the yield stack is now contested at once — interchange capped by settlement, debit routing in court until ~2028, merchants able to refuse premium cards, and bypass rails growing off a small base at network speed. Client incentives already absorb 28.2% of gross revenue and have crept up for three straight years; if the toll compresses 2 basis points, the credited bar stops clearing.

When a cardholder pays, four parties split the work: the merchant's bank (acquirer), the cardholder's bank (issuer), the merchant, and the network in the middle. Visa is only the middle. It sets the rules, runs the authorization/clearing/settlement pipes (VisaNet), and charges its clients — overwhelmingly the banks — service fees on volume and data-processing fees per transaction. The big interchange fee merchants complain about flows mostly to issuers, not to Visa; Visa's own take is the smaller, steadier toll measured here as net yield. Visa lends nothing, holds no card balances, and earns no float — settlement funds pass through matched accounts intraday. On top of the toll, Visa sells value-added services (fraud tools, tokenization, consulting, commercial and money-movement products such as Visa Direct), now the fastest-growing layer: $3.8B in Q3 FY2026, +34% year over year.

FY2023: guided low-double-digit net revenue growth; delivered +11%. FY2024: guided low double digits; delivered +10% GAAP (in range on the company's adjusted basis, as stated on the Q4 FY2024 call). FY2025: guided high-single to low-double digit; delivered +11% and non-GAAP earnings per share +14%. FY2026: guided low double digits in Oct 2025, then raised at Q3 FY2026 — net revenue growth to the low end of the low teens, non-GAAP EPS growth to the low end of the mid-teens. Six-plus quarters of the same pattern, tracked in the desk's append-only guidance ledger. The standing caveat, printed every time: the guided frame is non-GAAP and excludes litigation provisions, which have recurred every year in the window (FY2026 nine-month litigation provision alone: $1,290M).

At Jun 30, 2026 (10-Q): cash $12,359M and investment securities $1,583M against debt of $23,858M (current maturities $2,996M + long-term $20,862M) and preferred stock at a $514M carrying value — net debt $10,430M, about half of one year's owner earnings. Two items are deliberately excluded from the enterprise-value (EV) bridge and named here: the U.S. litigation escrow ($888M restricted cash, down from $2,990M at FY2025 year-end as settlement and B-share payments were made) sits against accrued litigation of $1,274M current; and settlement receivable/payable ($2,400M / $3,277M) plus customer collateral ($4,310M on both sides) are matched operating plumbing, not financing. The deferred client-incentive asset ($2,523M current + $5,884M non-current) is the balance-sheet shadow of multi-year issuer deals — contracted volume, already paid for.

Method, briefly. Solve on EV for the 10-year compound growth in total owner earnings (FCF − SBC) that makes the model's value equal today's EV at a 10% required return, with 2% terminal growth thereafter. EV bridge: market cap $752,881M (1,870M shares × $402.61, as-converted per Note 11 of the Q3 FY2026 10-Q) + debt $23,858M + preferred $514M − cash and investment securities $13,942M = EV $763,311M. Base: TTM owner earnings $20,094M (operating cash flow $22,580M − capex $1,567M − SBC $919M), reported, not normalized — Visa is a fee network with no lending book and no float engine, so cash flow is the right lens and no earnings-per-share-derived base is used. Memo: adding back the Q3 FY2026 $563M severance charge would lift the base to $20,657M and lower the flat bar only to +16.65%/yr — the conclusion does not turn on it. Terminal value is 62.4% of the flat-case value. At the delivered FY2021→FY2025 compounding pace (+10.28%/yr), the competitive-advantage period implied by the flat solve is ~24 years — the market is paying for a very long toll.

The credited bar. The credited bar is derived from the flat bar under the desk multiplication form — credited total growth = (1 + flat bar) × (1 − s) − 1 — at the named demonstrated input s = SHRINK_DILUTED = 1 − 2,100/2,242 = 6.33% (filed diluted weighted-average share counts, FY2025 vs FY2024): (1.17035) × (0.93666) − 1 = +9.62%/yr on the model’s unrounded inputs. Any shrink above 6.33% in the grid below is assumed by construction. The 9M FY2026 repurchase pace ($16,430M of Class A repurchases) says the shrink is still running; the May 2026 Class B exchange (Class B shares 125M → 63M) flatters the recent pace and is why the credited bar is anchored to the filed annual rate, not the latest quarter.

Visa's moat mechanics, in order of load-bearing weight. Acceptance ubiquity: ~175 million merchant locations and ~4.9 billion credentials make "not accepted" a merchant's problem, not Visa's. Answered Two-sided contracting: multi-year issuer deals, bought with client incentives and carried as an $8.4B deferred asset, lock the supply side. Answered Rule-setting: Visa writes acceptance, dispute, and routing rules — the asset every legal case on this page attacks. Contested Pricing power: net yield rose 26.5 → 28.2 bps across FY2023–FY2025 even as incentives crept — pricing power is real but increasingly spent on retention. Thin Switching costs: an issuer re-platforming is a multi-year program; a merchant steering at the till is an afternoon's configuration — the two sides switch at very different speeds. Contested The European base rate disciplines the whole debate: consumer interchange there has been capped at 0.2%/0.3% (debit/credit) since 2015, and Visa's European business still operates — caps compress yield; they have not, in that market, dissolved the network.

B+

Ryan McInerney (CEO since Feb 2023, president from 2013) and Chris Suh (CFO since Jul 2023) run a guide-low, deliver-and-raise cadence across the 12-quarter call record read for this edition. Execution is the strong leg: VAS scaled to a $3.8B quarter (+34%), Visa Direct and commercial volumes compounded in the teens, and the network-of-networks build (≈12 billion endpoints across cards, accounts, and wallets) is shipped product, not slide-ware.

Capital allocation is the cap. The shrink is genuinely large — diluted count 2,340M → 2,100M across FY2023–FY2025, TTM repurchases $21,357M, dividend +14% at the FY2025 print — but the timing record is price-insensitive: the pace rose as the price ran from the mid-$200s to a $402.61 close at 98.1% of its 52-week range. That is a policy of buying at any price, graded as such. Candor earns B+: the July 2024 U.S. volume soft patch was named on the call in real time and the settlement rejection was addressed head-on — but the guided frame excludes litigation that recurs annually, so the "clean" numbers always arrive with a GAAP asterisk ($3,100M of FY2025 operating-expense gap). Alignment: say-on-pay passed at 93.28% (Jan 27, 2026 annual meeting); SBC is modest ($919M TTM, 4.6% of owner earnings); the Form 4 census to Oct 16, 2026 (47 filings in the window) shows routine plan vesting and 10b5-1 sales and no open-market insider purchases — alignment by held equity, not by buying.

Leverage — PASS. Net debt $10,430M ≈ 0.5× TTM owner earnings; the debt stack is termed out and the commercial-paper program ($500M outstanding at Jul 28, 2026) is backstopped by the revolving facility.

Moat — FLAG. Honor-all-cards ends by category if the settlement is finally approved; the DOJ debit case seeks conduct remedies. Both are live, dated, and unresolved — the flag is the reason checkpoint 1 is decisive.

Management — PASS. No integrity or disclosure flag; the recurring-litigation exclusion from the guided frame is a candor deduction inside the grade, not a checklist failure.

Concentration — PASS. Revenue is diversified across ~14,500 institution clients and 200+ countries; the U.S. is the largest single market and the locus of the legal risk, which is carried in the Moat flag, not double-counted here.

Bias / valuation discipline — FLAG. The desk's own bias risk is anchoring on the credited bar: it is earned by a demonstrated input anchored to the filed annual rate — the May 2026 Class B exchange flatters the recent pace, which is why the credited bar is anchored to the filed annual rate, not the latest quarter. The flat bar is printed beside it on purpose.

Grading convention: an outcome between a checkpoint's PASS and FAIL lines is MIXED — neither passed nor failed — and the item is re-graded at the following print.

1. Decisive Net yield (the contested driver): FY2026 net revenue ÷ FY2026 payments volume. PASS ≥ 27.5 bps · FAIL < 26.5 bps. (FY2025: 28.2 bps.)
2. Net revenue growth, FY2026: PASS ≥ +12% · FAIL < +10%. (Raised guide: low end of the low teens.)
3. Payments volume growth, FY2026 (constant dollars): PASS ≥ +9% · FAIL < +7%. (FY2025: +8%.)
4. Processed transactions, FY2026: PASS ≥ 268B · FAIL < 262B. (FY2025: 257.5B.)
5. Value-added services revenue growth, Q4 FY2026 (as reported): PASS ≥ +25% · FAIL < +15%. (Q3 FY2026: +34%.)
6. First FY2027 guide — net revenue growth, as guided: PASS "low double digits" or better · FAIL "high single digits" or lower.
7. Capital return: Q4 FY2026 repurchases PASS ≥ $4.0B · FAIL < $2.5B; and FY2026 diluted WASO PASS ≤ 2,010M · FAIL > 2,040M.
8. Litigation provision, FY2026 (GAAP): PASS ≤ $1,700M · FAIL > $2,200M. (Nine months FY2026: $1,290M.)

Update protocol: this page is re-scored against the lines above after the October 27, 2026 print; every threshold was set on Oct 17, 2026 and will not be redrawn after the fact.

First standard edition — there is no prior edition to correct. One build-stage correction is recorded in the dossier and summarized here so the record is complete: the TTM client-incentives figure was recomputed at the second-reader pass to $17,442M (28.2% of gross revenue), the only value at which gross − incentives = net revenue ties across all three filed legs; the page carries the corrected figure throughout.

Correction — Oct 17, 2026 (share-count basis). The price strip and enterprise-value (EV) bridge in the original edition used 1,783M shares — the raw Jun 30, 2026 balance-sheet count, with the classes counted 1-for-1. The desk count is the as-converted common count in Note 11 of the Q3 FY2026 Form 10-Q: 1,870M shares (preferred stock excluded; it is carried separately in the EV bridge). Market cap is restated from $717,854M to $752,881M and EV from $728,284M to $763,311M; the flat bar, credited bar, breakeven grid, terminal-value share, memo bar, and competitive-advantage period were re-solved through the governing model and re-printed: flat bar +16.39%/yr → +17.04%/yr; memo (severance add-back) flat bar +16.01%/yr → +16.65%/yr; terminal-value share 61.9% → 62.4%; competitive-advantage period ~23 years → ~24 years. No checkpoint was changed.

Correction — Oct 17, 2026 (credited-bar convention). The credited bar was previously solved directly per share against market cap. It is now derived under the desk multiplication form — credited total = (1 + flat bar) × (1 − s) − 1, at the demonstrated shrink s = 6.33% — and prints at +9.62%/yr on the corrected EV basis (previously +8.83%/yr, on the old count basis and the old convention). No checkpoint was changed.

FY2021–FY2022 gross-revenue/incentive splits and class-level WASO were not re-verified for this edition (the current 10-K comparative tables carry three years); the series marks those cells with dashes rather than estimates. TTM payments volume is not published by the company on a single filed basis, so TTM net yield is not printed — FY2025 yield (28.2 bps) is the last filed annual figure and the checkpoint's reference. The Form 4 census counts filings in the issuer's EDGAR submissions index; it is a filing census, not a transaction-level parse. Settlement and DOJ-case status are as of Oct 16, 2026 and can move before the print.

T1 — Filings & primary records: FY2025 Form 10-K (filed Nov 6, 2025; revenue/incentive tables, EPS-note WASO classes, cover share counts) · Q3 FY2026 Form 10-Q (filed Jul 29, 2026; income statement, cash flows, Jun 30, 2026 balance sheet) · DEF 14A (Dec 8, 2025) · Form 8-K annual-meeting results (Jan 28, 2026; say-on-pay 93.28%) · Form 4 census via EDGAR submissions index (47 filings, Oct 17, 2025–Oct 16, 2026) · Q3 FY2026 earnings release and deck (company, via Form 8-K/IR).

T2 — Transcripts & company communications: 12 quarters of earnings calls, fiscal Q4 2023–Q3 2026 (Motley Fool full-text library; MarketBeat mirror for Q3 FY2026 and Q4 FY2025; IR webcast archive) · Visa announcement of the Oct 27, 2026 print date (Oct 8, 2026, company newsroom/IR).

T3 — Reported (color only, labeled where used): interchange settlement preliminary approval (Jun 9, 2026) and $5.5B damages distribution tally (Reuters/Payments Dive reporting) · DOJ debit-case discovery extension to Jan 29, 2027 (Payments Dive) · price history cross-checks (Finnhub; StockScan). No analyst estimates are used as inputs anywhere.