Mastercard MA

Reference table

1. What the business actually is

The 10-K filed 2026-02-11 states it directly:

> "Mastercard is a technology company in the global payments industry... Through our unique and proprietary global payments network, we switch (authorize, clear and settle) payment transactions. We have additional payments capabilities that include automated clearing house ('ACH') transactions (both batch and real-time account-based payments). Using these capabilities, we offer consumer and commercial payment products, capture new payment flows and provide services and solutions."

Segments: there is exactly one, and the filing names it — "one reportable operating segment, 'Payment Solutions.' The Payment Solutions segment derives its revenues from a wide range of payments solutions provided to customers... All of the segment's activities are interrelated." The vendor's seg_prod block is therefore a disaggregation of revenue by category, not a segment table.

Revenue by category — and this is the entire investment case in one table ($M):

Value-added services and solutions has gone from 35.4% of net revenue to 40.6% in three years, growing at almost twice the rate of the network it sits on. The vendor block does capture this split for FY2022–FY2025 — but note that the FY2018–FY2021 rows use an entirely different taxonomy (Domestic Assessments, Cross-border Volume Fees, Transaction Processing, Other Revenues, Rebates), so the series has a definitional break at FY2022 and cannot be trended across it. Both taxonomies are internally consistent — FY2021's five lines sum to $18.884B and FY2022's two sum to $22.237B, each matching reported revenue exactly — they simply are not the same disclosure.

What is in each line, from the 10-K. The payment network is "domestic assessments... charges based on activity related to cards that carry the Company's brands where the merchant country and the country of issuance are the same", plus cross-border assessments and transaction-processing assessments — all stated net of rebates and incentives, which is the crucial accounting point taken up in Section 3. Value-added services and solutions is "security solutions, consumer acquisition and engagement services, business and market insights, digital and authentication, processing and gateway and other solutions."

Growth drivers for the June 2026 quarter, straight from the 10-Q filed 2026-07-30:

• "Net revenue increased 14%, or 12% on a currency-neutral basis... attributable to growth in both our payment network and value-added services and solutions."
• "Net revenue from our payment network increased 10%, or 8% on a currency-neutral basis... primarily driven by growth in domestic and cross-border dollar volumes and an increase in the number of switched transactions, reflecting growth trends across all of our key drivers."
• "Net revenue from our value-added services and solutions increased 20%, or 18% on a currency-neutral basis... driven primarily by (1) growth in our underlying key drivers, (2) our security solutions, consumer acquisition and engagement services, digital and authentication solutions and business and market insights and (3) pricing."
• Gross dollar volume: "For the three and six months ended June 30, 2026, GDV on a U.S. dollar-converted basis increased 9% and 10%, respectively, while GDV on a local currency basis increased 8% for each of the periods." For the full year 2025: "GDV on a U.S. dollar-converted basis increased 8.7%, while GDV on a local currency basis increased 8.6%."

Geography — and the vendor's table is not comparable across years. The FY2025 rows read Americas $14,044M / International Markets $18,747M; the FY2024 rows read North America $12,375M / International Markets $15,792M. The 10-K's disaggregation footnote resolves it: "Americas includes the United States, Canada and Latin America." The reporting region was re-cut, so the 13.5% "growth" in the Americas line is partly a redefinition, not a business event. The break is even more obvious a year earlier: North America went from $8,359M (FY2023) to $12,375M (FY2024), which is +48% and cannot be organic. We do not draw any geographic growth conclusion from this table. What we do use is the 10-K's own statement that "No individual country, other than the United States, generated more than 10% of net revenue in any such period."

Customer concentration, which Visa's filings do not disclose in the same form: "A significant portion of our net revenue is concentrated among our five largest customers. In 2025, the net revenue from these customers was approximately $6.9 billion, or 21%, of total net revenue. The loss of any of these customers or their significant card programs could adversely impact our revenue." Twenty-one percent of revenue from five relationships is a genuine risk factor and it belongs in the risk score.

Scale of the workforce, for comparison. The profile reports 39,800 employees against Visa's 34,100, on net revenue of $32.8B against Visa's $40.0B — roughly $824,000 of revenue per employee against Visa's $1,173,000. That is a 30% gap, and it is the one place where the bearish knowledge-base claim about "large-headcount enterprises" becoming defensive utilities has a number attached to it. The services business is more labour-intensive than the network; that is the price of the mix shift.

2. The mix shift — Mastercard is already two-fifths a services company

This is the most important analytical fact about Mastercard and it is the main reason to prefer it to its twin.

The 10-K's own account of 2025: "Net revenue from our value-added services and solutions increased 23%, or 21% on a currency-neutral basis... which included a 3 percentage point increase from Acquisitions. The remaining increase was driven primarily by (1) growth in our underlying key drivers, (2) our security and digital and authentication solutions, and consumer acquisition and engagement services, (3) pricing and (4) our business and market insights."

Three things follow, and one of them is uncomfortable.

First, the mix shift is genuinely re-rating-relevant. A network that switches transactions is valued on volume and price; a security-and-analytics business attached to that network is valued on recurring software-like economics. At 40.6% and rising roughly 200 basis points a year, Mastercard crosses 50% around 2030 on the current trajectory. That would change what the terminal multiple should be, and it is the single most credible route to a higher multiple than the one the market pays today.

Second, the operating leverage is real and reconciles. The 10-K: "Operating expenses increased 10% in 2025 versus the prior year. Adjusted operating expenses increased 14%" against revenue up 16%. Operating margin went from 55.3% (FY2024) to 59.5% (FY2025). Adjust both years for the litigation provision — $680M in 2024 and $504M in 2025 per the 10-K — and the margin goes from 57.7% to 61.0%, a 330-basis-point genuine expansion. For contrast, Visa's operating margin fell 570 basis points in its fiscal 2025 on a $2.2B litigation accrual. Mastercard is expanding margins while its twin is absorbing legal costs; that is the cleanest single comparison between them.

Third — and this is the uncomfortable part — Mastercard's services growth is now SLOWER than Visa's. Mastercard's value-added services grew 20% in the June quarter (18% currency-neutral). Visa's grew 33% in the same quarter, from $2.8B to $3.8B, off a base roughly a third smaller. Mastercard is further along the mix shift; Visa is moving through it faster. A reader deciding between the two on the services story alone should know both facts. Our reading is that Mastercard's position is more valuable today (40.6% versus roughly 27% of revenue) and Visa's momentum is better, and that neither difference is large enough to settle the choice on its own.

3. Growth, rebates and margins

Annual history:

Diluted shares have fallen from 1.006 billion to 883 million — 12.2% — in six years. That is roughly 2.2% a year of pure per-share accretion, and it is a material component of the EPS growth rate.

Quarterly, and the honest deceleration:

As-reported net revenue growth has decelerated in five consecutive quarters: 16.8%, 16.7%, 17.6%, 15.8%, 14.1%. We are not going to smooth that over. Two mitigations are legitimate. First, currency: the 10-Q reports the June quarter at "14%, or 12% on a currency-neutral basis" and the half at "15%, or 12% on a currency-neutral basis" — the underlying rate was flat at 12% across both periods, so the headline deceleration is partly a translation effect running the other way. Second, mix: operating margin in the June quarter was 60.2%, the highest in the nine-quarter series, so profit growth outran revenue growth. Net income rose 18.6% year over year in the quarter against revenue up 14.1%.

Rebates and incentives — the number that decides how much of the volume Mastercard keeps. From the 10-Q: "Net revenue from our payment network included $5,997 million of rebates and incentives provided to customers, which increased 22%, or 20% on a currency-neutral basis... primarily due to an increase in our key drivers as well as new and renewed deals." Over the half it was $11,636 million, also up 22%. For all of 2025 the 10-K puts it at $20,522 million, up 16%.

Mastercard returns slightly more than half of gross payment-network revenue to its customers as rebates and incentives. (A definitional warning: this is not directly comparable to Visa's 28.3% "client incentive rate", which is struck against Visa's total gross revenue including its services lines rather than against the network alone. The two disclosures use different denominators, and any cross-company incentive comparison that ignores that is wrong.) What matters is the trend: rebates grew 22% in the quarter against net revenue growth of 14%. That gap is the price of retaining issuers in a competitive market, and it is the single most watchable line in the quarterly disclosure.

Operating expense drivers, from the 10-Q: "For the three months ended June 30, 2026, operating expenses increased 10%... Adjusted operating expenses increased 11%, or 10% on a currency-neutral basis" — below revenue growth in both quarters, which is where the margin expansion comes from. General and administrative rose 11% in the quarter "primarily due to higher personnel and data processing costs to support the continued investment in our strategic initiatives across payment network and value-added services and solutions, as well as fulfillment costs to deliver marketing services." The six-month figure includes a $202 million restructuring charge ($158M after tax, $0.18 per diluted share) taken in the first quarter, which the filing says is "primarily intended to enable reinvestment to support the realization of our long-term growth opportunities."

The two earnings numbers — and Mastercard's gap is half of Visa's. Consensus estimates are non-GAAP, excluding "the impact of gains and losses on our equity investments, Special Items (which represent litigation judgments and settlements and certain one-time items) and the related tax impacts." The June quarter's adjustments were small: net losses of $2M on equity investments ($0.01 per share) and a litigation charge of $82M ($59M after tax, $0.07 per share).

Mastercard's GAAP-to-non-GAAP gap is 3.2% of trailing earnings; Visa's is 8.0%. The reason is the size of the litigation provision relative to earnings: Mastercard recorded $504M (2025), $680M (2024) and $539M (2023), against Visa's $2.2B in fiscal 2025 alone. Less of Mastercard's reported profitability depends on excluding recurring legal costs, and that is a real quality distinction between two otherwise near-identical franchises.

4. Balance sheet and capital returns

FILING OVERRIDES VENDOR — the debt figure is $5.6 billion higher than the screen shows.

The 10-Q is explicit: "Our total debt outstanding at June 30, 2026 and December 31, 2025 was $24.6 billion and $19.0 billion, respectively. At June 30, 2026, $2.5 billion of our total debt outstanding is payable within 12 months." The increase is fully explained: "In June 2026, we issued $5.000 billion principal amount of notes... The net proceeds... were $4.978 billion", plus $710M of commercial paper. The 8-K filed 2026-06-08 gives the composition: $500M of floating-rate notes due 2028, $1,250M of 4.325% notes due 2028, $1,150M of 4.425% notes due 2029, $1,350M of 4.600% notes due 2031 and $750M of 5.000% notes due 2036.

The vendor's FY2025 total debt of $19.000B is itself confirmed by the filing ($19.0B carrying value at 2025-12-31), so the vendor is accurate as of its own date and simply two quarters stale. We use $24.6B.

Net debt, recomputed and then estimated. On the vendor's own FY2025 data, netDebt of $8.434B is total debt less cash and cash equivalents only ($10.566B), discarding $332M of short-term investments; correctly computed FY2025 net debt is $8.102B. Rolling the debt side forward to the filing's $24.6B and holding cash and short-term investments at the FY2025 level of $10.898B implies net debt of roughly $13.7B at 2026-06-30 — about $5.6B more than any screen shows. (The 10-Q's liquidity table did not survive text extraction, so the cash side of that estimate is not independently verified and it is labelled an estimate throughout.) Even at $13.7B, net debt is under one year of net income and interest coverage is 27.9x. This is a leverage change, not a leverage problem.

Why book value, price-to-book and debt-to-equity are meaningless here — and we will not quote them as if they were not. Mastercard's stockholders' equity is $7.737B on a $500.9B market capitalisation. The reason is on the balance sheet in plain sight: retained earnings of $85.035B against treasury stock of −$83.224B. Roughly three decades of buybacks have consumed the book. The consequences:

The vendor's rating model gives Mastercard a B with an overall score of 3, with sub-scores of 1 for debt-to-equity, 1 for price-to-earnings and 1 for price-to-book. Two of those three — debt-to-equity and price-to-book — are the model penalising a capital-return policy it has mistaken for leverage. We disregard both sub-scores and note the model's own discounted-cash-flow score of 4, its highest, as the more informative reading.

Capital returns, and how they were funded — the finding of this section.

Mastercard returned $10.4B against $6.8B of operating cash flow in six months and borrowed to close the gap. The 10-Q confirms the mechanism without editorialising: "Net cash used in financing activities decreased $953 million... primarily due to higher cash proceeds received from debt issuances and lower repayments of debt in the current year, partially offset by higher cash paid for repurchases of our Class A common stock and dividends."

Two qualifications, both fair. First, this is not new for the industry — buying back stock with cheap debt is standard for asset-light, cash-generative franchises, and at 27.9x interest coverage Mastercard has enormous room. Second, the repurchase price was good: $8.9B for 17.6M shares is roughly $506 a share, 11.5% below today's close of $571.10. Management bought well.

But the runway is thinner than Visa's. The remaining authorisation of $7.8B is 1.6% of market capitalisation, against Visa's $28.4B at 4.1%. Mastercard re-authorises annually — $12.0B in December 2024 and $14.0B in December 2025, with the 2025 programme becoming effective in March 2026 after the 2024 programme completed — so a December 2026 authorisation is highly likely, but it is not in hand.

Dividends. dividendPerShareTTM is $3.37, a 0.590% yield, with $1.5B paid in the first half. Total shareholder yield is roughly 4.1% — about 3.55% of buyback at the first-half run-rate plus 0.59% of dividend — of which a material portion in the first half was debt-financed.

Leases: checked, and immaterial. The 10-K's lease footnote reports operating lease amortisation of $161M in 2025 (against $145M and $141M in the two prior years), a weighted-average remaining lease term of 7.2 years and a weighted-average discount rate of 3.6%. Right-of-use assets are described as "primarily related to operating leases for office space, automobiles and other equipment." There is no data-centre lease programme, no not-yet-commenced lease commitment, and nothing here that changes a leverage view. $161M of annual lease amortisation against $32.8B of revenue is a rounding error, and we say so explicitly because the same footnote at other issuers in this batch hides tens of billions.

5. Cash flow

The fiscal 2025 buyback and dividend lines are independently confirmed by the 10-K — "We repurchased 21.1 million shares of our common stock for $11.7 billion and paid dividends of $2.8 billion" — matching the vendor's $11.727B and $2.756B. That is a positive verification and it is worth stating, because the equivalent line for Visa did not survive the same test.

One minor discrepancy, flagged not smoothed: the 10-K states 2025 operating cash flow of $17.6 billion; the vendor reports $17.401 billion, which rounds to $17.4B. A $199M gap on a $17B figure is immaterial to every conclusion here, but it is a gap and we do not paper over it. We use the filing's $17.6B where the round number matters and the vendor's precise figure in the table above, labelled.

Capital intensity is trivial but rising. Capex was $371M in 2023, $474M in 2024 and $1.215B in 2025 — a 2.6x increase in two years, now 3.7% of revenue and 5.3% of operating cash flow. The 10-Q confirms the direction: investing outflows increased "primarily due to higher purchases of property and equipment." Free cash flow still converts at 94.7% of operating cash flow, and free cash flow per share is $18.65 for a 3.28% free cash flow yield.

The near-term cash-flow drag has the same cause as Visa's, and the filing names it: "Net cash provided by operating activities decreased $211 million for the six months ended June 30, 2026... primarily due to higher net income after adjusting for non-cash items, more than offset by higher customer incentive payments and cash paid for litigation settlements." The US MDL accrual fell from $637M to $149M over the same six months — that decline is cash leaving the building as settlements are paid. This is a good problem: the litigation is resolving, and the cash cost lands now.

A vendor inconsistency in the cash-flow block: cf_a reports FY2025 depreciation and amortisation of $2.098B while inc_a reports $1.143B for the same year. The cash-flow figure includes amortisation items the income-statement line does not. Neither drives a conclusion here, and the difference is noted rather than reconciled.

6. Valuation — priced in or room?

At $571.10 (market cap $500.9B):

Peer context, with the same warning as its twin. The vendor peer list is ALLY ($13.7B), AXP ($234.1B), COF ($136.1B), PYPL ($50.1B), SOFI ($24.0B), UPST ($2.9B), V ($690.0B). Ally, Capital One, SoFi and Upstart are balance-sheet lenders and credit underwriters — their earnings are net interest income and their risk is credit risk, which Mastercard does not take. No inference is drawn from them. American Express is a closed loop carrying card-member receivables — adjacent, not comparable. PayPal is a genuine competitor in digital commerce but a different business model. Visa is the only true comparable in the list, and it is compared directly in Section 11. No peer multiples are supplied in the file, so no peer-multiple table is drawn.

6a. What today's price assumes (the inversion)

At $571.10 — 24.8x FY2027E consensus EPS of $22.993 on a non-GAAP basis, or roughly 25.7x on a GAAP-equivalent basis — the price embeds the following falsifiable claims:

• Net revenue compounds ≥12% annually through FY2028, reaching $46.974B. (Consensus-derived: $37.231B FY2026E on 26 analysts, $41.926B FY2027E on 28, $46.974B FY2028E on 25 — deep, consistent coverage across all three years.) The current currency-neutral run-rate is 12%, so this assumes no further deceleration.
• EPS reaches ~$26.72 by FY2028, a 16.0% annual compound. (Consensus-derived; 15 analysts on FY2028.) Roughly 2.2 points a year of that comes from the buyback, so the assumption embeds continued repurchase at scale — against a remaining authorisation of only $7.8B and a first half in which repurchases exceeded operating cash flow.
• Rebates and incentives stay near half of gross network revenue and do not accelerate further. (Our number, derived from the 10-K's $20,522M against $19,476M of net network revenue.) Rebates grew 22% in the June quarter against net revenue growth of 14%; the price assumes that gap narrows rather than widens.
• Value-added services and solutions keeps compounding near 20% and keeps taking share of the revenue mix (consensus-implied; the 10-Q reports +20% for the quarter and +21% for the half). This is the leg that justifies paying a network-plus multiple rather than a network multiple.
• The US and European interchange litigation is resolved for something close to the current accrual. (Our reading, from the 10-Q.) The US MDL accrual is $149M. Block and Intuit are "seeking aggregate single damages in excess of $5 billion"; a separate group of two opt-out merchants seeks over $250M with trial in September 2026; unresolved UK and Pan-European merchant claims exceed £0.5B (~$0.7B); a UK collective action claims over £1B (~$1.3B); and a Portuguese consumer action claims roughly €0.4B (~$0.5B) with trial in October 2026. This is the most fragile assumption in the price, because Mastercard has no retrospective responsibility plan — no escrow, no bank-held share class — and the 10-Q states the accrual "does not represent an estimate of a loss, if any, if the matters were litigated to a final outcome."

6b. The return bridge (why the multiple moves)

Expected return over 12–24 months decomposes as: EPS growth (+16.0% a year on consensus) + multiple drift (we assume a modest EXPANSION, from 24.8x to ~27.0x FY2027E) + shareholder yield (~4.1%, of which about 3.5% is buyback and 0.59% dividend).

Our base case assumes modest multiple expansion, and the reason is that the de-rating has already happened rather than that we expect enthusiasm. Over the last twelve months Mastercard returned 2.0% while EPS grew roughly 16% — the multiple contracted by something close to four turns while the business compounded. At 24.8x forward for a 16% grower, the price-to-earnings-growth ratio is about 1.55, and the vendor's own priceToEarningsGrowthRatioTTM of 1.39 agrees that this is not a demanding setup. Mastercard is not being asked to re-rate to a new level; it is being asked to stop de-rating.

Three forces argue against pushing the base multiple higher than 27x. First, as-reported revenue growth has decelerated for five consecutive quarters, and multiples do not expand into decelerating headline growth even when the currency-neutral rate is stable. Second, the buyback — 2.2 points a year of the EPS growth — now runs partly on borrowed money, with $7.8B of authorisation left and a December re-authorisation not yet in hand. Third, the litigation is under-accrued relative to the claims and Mastercard's shareholders, unlike Visa's, absorb it directly.

Roughly two-thirds of the expected return in the base case is earnings growth and shareholder yield; one-third is the multiple. Base fair value of $621 is 27.0x FY2027E, which is +8.7% plus a 0.59% dividend. If the multiple simply holds at 24.8x, fair value is $571 — today's price exactly. That, as with its twin, is the arithmetic behind the Hold.

The bull case at $700 requires 30.4x FY2027E (26.2x FY2028E) and is still below the street's high target of $735. It needs value-added services to keep compounding near 20%, the litigation to close near the accrual, and the market to pay a services multiple for a business it currently pays a network multiple for. That is the fragile leg.

6c. Variant perception (where we differ, what would surprise)

• We differ from the screens on leverage in both directions, and the net is that risk is higher than displayed. Screens show total debt of $19.0B and net debt of $8.434B. The 10-Q shows total debt of $24.6B at 2026-06-30, and the vendor's netting omits $332M of short-term investments. On the vendor's own cash held constant, net debt is nearer $13.7B. Simultaneously, the screens show debt-to-equity of 4.39x and price-to-book of 89.8x, which overstate risk because equity is a buyback residual. The correct reading is: more absolute debt than shown, and none of the distress the equity ratios imply. Watchable number: total debt at each quarterly print, and whether repurchases continue to exceed operating cash flow.
• We differ on what the twenty-two-point underperformance means. The market spent a year treating Mastercard as a stalled compounder. Over roughly that year EPS grew about 16%, operating margin expanded 330 basis points on an adjusted basis, and the services line went from 38.5% to 40.6% of revenue. The de-rating was not earned by the fundamentals; it was earned by the category — and our own knowledge base explains why, since the most articulate recent claims about Mastercard are about stablecoins disintermediating card networks, not about Mastercard's execution. We think the business is better than its twelve-month price record, and that the price record is a fair reflection of an unresolved structural question. Both.
• We differ from the bull lane on how much history proves. The strongest bullish claims we track are backward-looking — a 33% annual compound since the 2006 listing, an 86-bagger over sixteen years, a 40% return on invested capital sustained for a decade. All true, all verifiable, and none of them is an argument about 2027. One of the same sources says the more useful thing: Mastercard "generates too much cash to reinvest at similar high rates, so its reinvestment leg is weaker — returns via dividends, buybacks, acquisitions." That is exactly what the first half of 2026 looked like, and it is why the exponential score is 4.
• We do NOT have a variant perception on the growth rate. Consensus models 12-13.5% revenue and 16.0% EPS growth; the currency-neutral run-rate is 12% and the buyback supplies the rest. Our numbers are the consensus numbers. Where we have no edge we say so, and a no-edge name at 24.8x defaults toward Hold.
• Positive surprise that would force a re-rate: value-added services and solutions re-accelerating above 25% and passing 45% of net revenue, which would start the market pricing a services multiple. Final approval of the Rules Relief Class settlement — the hearing is scheduled for November 2026 after preliminary approval in June 2026 — combined with a favourable resolution of the Block and Intuit matters, which would remove the largest unquantified liability. A December 2026 buyback authorisation materially above $14.0B. Watchable numbers: the services growth rate (20% in the June quarter), the US MDL accrual ($149M), the remaining authorisation ($7.8B at 2026-07-27).
• Negative surprise that would break the thesis: an adverse outcome in the Block and Intuit matters, where claims exceed $5 billion against an accrual of $149M and where, unlike at Visa, no escrow or bank-held share class absorbs the loss. Rebates and incentives continuing to grow 8 points faster than net revenue. Currency-neutral net revenue growth falling below 10%. Or the loss of one of the five customers that together supply 21% of net revenue. Watchable events: the September 2026 opt-out trial, the October 2026 Portuguese trial, the November 2026 Rules Relief final-approval hearing, the February 2027 UK liability appeal.

Synthos fair values

All three anchors are multiples of FY2027 consensus EPS of $22.993 (27 analysts — the deepest EPS coverage in the estimate block), cross-checked against FY2028E of $26.722 and against a GAAP-equivalent FY2027 EPS of roughly $22.23.

• Bear ~$471 — 20.5x FY2027E (21.2x GAAP-equivalent). Cross-check: 17.6x FY2028E; 25.9x trailing GAAP. Note that $471 is within fifteen cents of the 52-week closing low of $471.55 — this is a retest of the year's worst level, not a new regime. The scenario: an adverse outcome in the Block and Intuit matters against a $149M accrual, or rebates accelerating further, or currency-neutral growth slipping below 10%. −17.5%.
• Base ~$621 — 27.0x FY2027E (27.9x GAAP-equivalent). Cross-check: 23.2x FY2028E; 34.2x trailing GAAP. The scenario: revenue compounds at the consensus 12-13%, value-added services keeps taking mix, operating margin holds above 59%, the buyback is re-authorised in December, litigation resolves near the accrual, and the year-long de-rating stops. +8.7%, plus a 0.59% dividend.
• Bull ~$700 — 30.4x FY2027E (31.5x GAAP-equivalent). Cross-check: 26.2x FY2028E. This sits below the street's high target of $735 and above its median of $665. The scenario: services growth re-accelerates and crosses 45% of revenue, the November 2026 and 2027 legal calendar clears, and the market pays a services-and-network multiple. +22.6%.

Base is 8.7% above spot; asymmetry is roughly 1.29:1 to the upside (−17.5% down, +22.6% up) — slightly better than Visa's 1.16:1, and the reason we prefer this name of the two. Against a street consensus of $657.47 (28.6x FY2027E), our base is 5.6% lower. A better business than its twelve-month chart, at a price that is fair rather than cheap. Hold.

7. Knowledge base — 14 usable claims, and a lane that argues with itself

Search pattern and false-positive check. The patterns searched were \bMastercard\b and \bMaster ?[Cc]ard\b. Unlike the search for Visa — where the case-insensitive word "visa" also matches the immigration sense and produced three false positives that had to be identified and discarded — the Mastercard pattern has no homograph problem. Every one of the 21 raw hits refers to the company. That is worth stating explicitly, because the two searches were run the same way and only one needed cleaning.

The one exclusion: a 2025-12-10 claim ("Firms with early stablecoin positioning—Stripe (buying Bridge Oct 2024), Mastercard—are well placed; banks are scrambling to catch up") is tagged to a quarantined-misattribution source and is excluded outright under the standing attribution policy, notwithstanding that it is relevant, recent and bullish. We flag that this exclusion removes a bullish data point; the policy applies regardless of which way the claim cuts.

Stance mix on the 14 usable claims: 7 bullish, 3 neutral, 4 bearish, across 7 distinct sources (jordi_visser 3, we_study_billionaires 3, invest_like_the_best 3, all_in 2, real_vision 1, business_breakdowns 1, lyn_alden 1), spanning 2022-02-09 to 2026-02-27. Fourteen claims from seven sources is a thin lane and the conviction rating says so.

The bull lane — durability, mostly historical:

> 2022-02-09 · invest_like_the_best · bullish · conviction 90

> "Visa and Mastercard have extraordinary, durable margins and returns on capital — you could halve MA's margins twice and still beat the average business."

> 2022-02-09 · invest_like_the_best · bullish · conviction 85

> "Mastercard compounded ~33% annually since its 2006 IPO — an 86-bagger over 16 years — an incredible long-term compounding engine."

> 2025-10-23 · we_study_billionaires · bullish · conviction 75

> "Both sustain ~40% ROIC over a decade, proving self-reinforcing moats—cited as best-quality-idea breakdowns where performance verifies the advantage."

> 2023-05-20 · business_breakdowns · bullish · conviction 82

> "Card networks are an entrenched protocol layer between fragmented issuers and acquirers; a huge, defensible moat despite taking the smallest clip of economics."

> 2026-02-12 · we_study_billionaires · bullish · conviction 70

> "Visa/Mastercard are a rational duopoly that protects industry margins rather than price-warring, expanding the pie by converting remaining cash to digital."

> 2026-02-12 · we_study_billionaires · bullish · conviction 58

> "Mastercard shares Visa's identical toll-booth model with slightly higher growth potential; many investors own both to capture the whole industry."

Two of these check out against the filings and one is stale. The "~40% ROIC over a decade" claim is not only right, it is conservative — returnOnInvestedCapitalTTM is currently 47.8%. The "halve MA's margins twice" observation is arithmetically sound: quartering a 59.4% operating margin leaves 14.9%, which is indeed above most businesses. The "slightly higher growth potential" claim from February 2026 is directionally correct on consensus (16.0% EPS growth against Visa's 13.6%) but is now contradicted by the most recent prints, where Mastercard's June-quarter revenue grew 14.1% and Visa's grew 14.4%. The gap has closed on trailing numbers even as consensus keeps modelling it forward.

And the single most useful bullish claim in the file is the one that qualifies the others:

> 2022-02-09 · invest_like_the_best · neutral · conviction 70

> "Mastercard generates too much cash to reinvest at similar high rates, so its reinvestment leg is weaker — returns via dividends, buybacks, acquisitions."

That is precisely what the first half of 2026 looked like: $10.4B returned against $6.8B generated, with $5.0B borrowed to bridge it, and capital expenditure of 3.7% of revenue. A four-year-old claim describing this year's cash flow statement is the sort of corroboration that earns weight.

The bear lane — structural, recent, and from the highest-skill voice in the file:

> 2026-02-27 · jordi_visser (skill 2.0) · bearish · conviction 80

> "Payment middlemen (fee-on-a-fee) get disrupted by stablecoins; large-headcount enterprises become defensive utilities, not growth businesses."

> 2026-02-21 · jordi_visser (skill 2.0) · bearish · conviction 75

> "Card networks are a deflationary victim; the K-shaped economy will move to stablecoins to save money, cutting out the middleman and stalling Mastercard growth."

> 2025-09-15 · all_in · bearish · conviction 65

> "Incumbents have regulatory muscle and scale but are slow to adopt tech and can't hire top talent, losing to nimble fintechs."

> 2022-09-03 · lyn_alden (skill 1.1) · bearish · conviction 65

> "A dollar stablecoin on Bitcoin/Lightning would be a big threat to Visa and Mastercard's debit networks — permissionless, cheaper, no ~3% fees."

And the same high-skill source, on Mastercard specifically, is complimentary:

> 2025-11-08 · jordi_visser (skill 2.0) · neutral · conviction 78

> "Firms lacking a multi-year stablecoin plan get punished (Fiserv -40%); Stripe's Bridge buy and Mastercard are ahead while banks scramble to compete."

> 2025-09-18 · all_in · neutral · conviction 65

> "Visa/Mastercard are thin-margin (~10bps) tech companies that get stronger if they cut banks out via stablecoin transfers behind the scenes."

> 2025-08-27 · real_vision · bullish · conviction 70

> "Incumbent card networks are building on stablecoin rails, validating blockchain payments as the future settlement layer."

This is the finding, and it is why the net conviction is mixed rather than positive. The single highest-skill source in the file (skill 2.0) appears three times: bearish on the business model in February 2026 at convictions of 75 and 80 — naming Mastercard specifically in the second — and neutral-to-complimentary on Mastercard's execution in November 2025, placing it ahead of the banks on stablecoin positioning. That is not incoherence. It is a voice distinguishing "this company is well run and well positioned" from "this category gets disintermediated," and holding both. We hold both too.

Mastercard's own 10-K says the same thing in a single sentence. On stablecoins and digital currencies: "Stablecoins and cryptocurrencies may become more popular as they potentially become more regulated around the globe (through laws such as the GENIUS Act, which in 2025 became the first major crypto legislation in the U.S.)... Digital currencies and emerging players (such as crypto natives) have the ability to disrupt traditional financial markets. The increased prominence of digital currencies creates an opportunity for us, but could also compete with our products and services." And on government rails: "Government- and central bank-backed structures (such as the Brazilian Instant Payment System-PIX, FedNow in the U.S. and United Payments Interface (UPI) in India), are increasingly being considered as alternatives to traditional domestic payment solutions and schemes such as ours." Mastercard's answer, from the same filing: "Our solutions enable consumers to use our cards to purchase digital assets and spend those balances across our acceptance network using crypto co-brand cards. Additionally, we support the settlement of stablecoins over our network." Unlike Visa's 10-K, which quantifies its stablecoin settlement at a $2.5B annualised run-rate, Mastercard gives no number. We treat the disclosure as directional only.

Weak non-entity context, retained but unweighted: a 2026-02-21 claim that "Stablecoins are hitting hockey-stick adoption—$11T monthly volume vs MasterCard's $10T yearly" (from the same high-skill source, at conviction 90, and the most aggressive framing of the bear case in the file); a 2026-07-18 claim that "A Stripe/Block/PayPal combo with 600-700M accounts plus stablecoin rails could vertically integrate into a powerful competitor"; a 2026-02-10 claim about "trillions shifting from incumbents (Visa, Mastercard, custodians) to new rails"; a 2025-01-04 claim from gavin_baker that dollar stablecoins "will attack the Visa/MasterCard duopoly"; and, on the other side, a 2025-05-09 observation that hard data from the networks "shows consumer spending remarkably resilient" — which the 8-10% gross-dollar-volume growth in the filings independently confirms.

Why net conviction is "mixed." Seven bullish, three neutral, four bearish, from seven sources, over four years. The bull case is mostly about a decade that has already happened; the bear case is about a decade that has not. The highest-skill voice is bearish on the category and constructive on the company, which is the same conclusion this dive reaches from the numbers. A divided, shallow lane with an unresolved structural question supports a Hold.

8. Data integrity — what we rejected from the vendor file and why

Ten material items. Two are direct filing-versus-vendor contradictions, and in both the filing wins. One is an outright arithmetic impossibility in the vendor file.

1. FILING OVERRIDES VENDOR — total debt is $5.6 billion higher than the screen shows. The vendor's newest balance sheet (FY2025, 2025-12-31) reports total debt of $19.000B. The 10-Q filed 2026-07-30 states: "Our total debt outstanding at June 30, 2026 and December 31, 2025 was $24.6 billion and $19.0 billion, respectively." The vendor's figure is correct as of its own date and two quarters stale; the increase is the $5.0B June note offering (detailed in the 8-K filed 2026-06-08: $500M floating due 2028, $1,250M at 4.325% due 2028, $1,150M at 4.425% due 2029, $1,350M at 4.600% due 2031, $750M at 5.000% due 2036) plus $710M of commercial paper at 3.84%. We use $24.6B. The 10-Q also gives the fair value of that debt as $23.3B, and states that $2.5B is payable within 12 months against the vendor's $749M of short-term debt.

2. VENDOR ARITHMETIC IMPOSSIBILITY — earn_cal reports identical revenue for two different quarters. The vendor's earnings calendar reports revenueActual of $9.277B for the 2026-07-30 print AND $9.277B for the 2026-04-30 print — the same figure, to the dollar, for two consecutive and different quarters. This cannot be true. The vendor's own quarterly income statements contradict it: inc_q reports $9.277B for the quarter ended 2026-06-30 and $8.398B for the quarter ended 2026-03-31. The filing settles it: the 10-Q's narrative gives net revenue up 14% in the quarter and 15% over the six months against 2025 comparatives of $8.133B and $15.383B, which reconciles to $9.277B and $17.675B — i.e. $8.398B for the March quarter. The earn_cal 2026-04-30 revenue figure is a vendor error and is rejected. We use $8.398B. (Note that the same row's epsActual of $4.60 is a non-GAAP figure and is correct as such; only the revenue is wrong.)

3. Book value, price-to-book, debt-to-equity and return on equity are meaningless for this issuer — REJECTED as measures. Stockholders' equity of $7.737B on a $500.9B market capitalisation is the residual of $85.035B of retained earnings less $83.224B of treasury stock. Consequently priceToBookRatioTTM (89.77x), priceToFairValueTTM (89.77x, which is the same number under a different name), debtToEquityRatioTTM (4.39x), debtToCapitalRatioTTM (0.815), financialLeverageRatioTTM (10.28x) and returnOnEquityTTM (232.5%) are arithmetically correct and economically empty. They are not used anywhere in this dive. returnOnInvestedCapitalTTM of 47.8% and interestCoverageRatioTTM of 27.9x are the meaningful substitutes and are used instead. We also disregard two of the vendor rating model's three sub-scores of 1 — debt-to-equity and price-to-book — because they penalise a buyback policy the model has mistaken for leverage.

4. netDebt ignores short-term investments — REJECTED and recomputed. Vendor netDebt of $8.434B is total debt ($19.000B) less cash and cash equivalents only ($10.566B), discarding $332M of short-term investments. Correctly computed, FY2025 net debt is $8.102B. Combined with item 1, the current figure is materially higher: rolling the debt to $24.6B and holding cash at the FY2025 level implies roughly $13.7B of net debt at 2026-06-30 — labelled an estimate, because the 10-Q's liquidity table did not survive text extraction.

5. Enterprise value — a +1.0% residual, immaterial but stated. enterpriseValueTTM of $514.227B against a rebuild from the vendor's own inputs ($500.875B + $19.000B − $10.898B = $508.977B). Well within the 15% withholding threshold. But on the filing's debt figure the true enterprise value is nearer $514.6B, so the vendor number is coincidentally close for the wrong reason. evToEBITDATTM of 23.25x and netDebtToEBITDATTM of 0.604x are built on the stale debt and are not quoted in this dive.

6. inc_q gross margin of 96.7% in the June 2026 quarter is a cost-allocation artefact — REJECTED. The vendor reports gross profit of $8.968B on revenue of $9.277B for the quarter, a 96.7% gross margin, against 75.7% in the prior quarter and a FY2020–FY2025 range of 75.2%–77.9%. This is a cost-of-revenue allocation error, not a business event. Operating margin of 60.2% for the same quarter is plausible, consistent with the nine-quarter series and corroborated by the filing's own operating-margin metric. We use operating margin throughout and quote no quarterly gross margin.

7. est net-income rows are internally inconsistent with est EPS rows — REJECTED. Dividing niAvg by epsAvg gives an implied diluted share count of 882.2M (FY2026E), 903.2M (FY2027E), 889.4M (FY2028E), 898.0M (FY2029E), 898.0M (FY2030E) — a count that rises from 882M and then plateaus, at a company whose diluted share count has gone 1.006B → 927M → 898M → 883M and which repurchased 17.6M shares in the first half of 2026 alone. Implied net margins escalate from 47.1% (FY2026E) to 54.2% (FY2030E) against a trailing 46.3%. All niAvg rows are rejected; forward valuation runs on epsAvg only.

8. est EBITDA and EBIT rows pass the sanity check but are non-GAAP and are used for nothing. FY2026E ebitdaAvg of $24.019B exceeds ebitAvg of $22.878B which exceeds niAvg of $17.538B — the correct ordering, unlike roughly 70% of files in this dataset. But they are struck on the same non-GAAP basis as the EPS rows and are not comparable to the FY2025 actual EBITDA of $20.443B or operating income of $19.514B. No EBITDA multiple appears anywhere in this dive.

9. The seg_geo block is not comparable across years — REJECTED for trend analysis. FY2025 reports Americas $14,044M / International Markets $18,747M; FY2024 reports North America $12,375M / International Markets $15,792M. The 10-K footnote confirms the redefinition: "Americas includes the United States, Canada and Latin America." The FY2023→FY2024 step is even more obviously definitional — North America goes from $8,359M to $12,375M, +48%, which cannot be organic. All three region rows sum correctly to reported revenue in each year, so the totals are sound; only the year-over-year comparison is invalid. No geographic growth conclusion is drawn. The seg_prod block has an analogous break: FY2018–FY2021 use a five-line taxonomy (Domestic Assessments, Cross-border Volume Fees, Transaction Processing, Other Revenues, Rebates) and FY2022–FY2025 use the two-line Payment Network / Value-Added Services split. Both are internally consistent and they cannot be trended across FY2021/FY2022.

10. Minor inconsistencies, flagged and resolved. (a) The 10-K states 2025 operating cash flow of $17.6 billion; the vendor reports $17.401B — a $199M gap, immaterial to every conclusion but noted. (b) cf_a reports FY2025 depreciation and amortisation of $2.098B against inc_a's $1.143B for the same year; the cash-flow figure includes amortisation the income-statement line does not, and neither drives a conclusion. (c) quote.yearHigh of $601.77 and yearLow of $464.52 are intraday extremes; tech.hi52 of $598.96 and tech.lo52 of $471.55 are closing extremes — we use the closing series throughout and note the difference once. (d) quote.sma50 of $514.132 differs from tech.sma50 of $517.0202; we use the tech block, which is computed from the same close series as the rest of Section 9. (e) The vendor's implied share count (market cap ÷ price = 877.0M) differs from the count implicit in its own per-share ratios (882.0M) and from the Q2'26 diluted count (883M) — a spread of 0.7%, small enough to be immaterial and noted for completeness. This is a marked contrast with Visa, where the equivalent spread is 3.4% because of the as-converted class B and C structure.

Two positive verifications, stated because they are as informative as the rejections. (i) The vendor's FY2025 buyback of $11.727B and dividends of $2.756B are confirmed by the 10-K: "We repurchased 21.1 million shares of our common stock for $11.7 billion and paid dividends of $2.8 billion." (ii) The vendor's FY2025 total debt of $19.000B is confirmed by the 10-Q's "$19.0 billion" at 2025-12-31.

A note on what the filings could NOT provide. These SEC text extractions preserve narrative prose but drop numeric table cells. The key-driver tables (gross dollar volume, cross-border volume growth, switched transactions), the components-of-net-revenue table, the balance sheet, the liquidity table, the operating-expense table and the share-repurchase table are all present as headings with their footnotes and their numbers absent. Every filing figure quoted in this dive comes from narrative sentences. That is why we can state that gross dollar volume grew 9% on a US-dollar-converted basis and 8% on a local-currency basis, but cannot state the dollar volume; and why cross-border and switched-transaction growth rates are described only by the filing's summary phrase, "reflecting growth trends across all of our key drivers." Stated rather than papered over.

Risk factors: no change disclosed. The 10-Q states — "For a discussion of our risk factors, see Part I, Item 1A - Risk Factors of our Annual Report on Form 10-K for the year ended December 31, 2025" — with no additions or amendments. The risk factors are unchanged from the 10-K, and we checked rather than assumed.

Not defects, correctly reported: currentRatioTTM of 1.061 and workingCapitalTTM of $1.518B are modest but positive and reflect settlement balances. cashConversionCycleTTM of −15.6 days is negative and correct: Mastercard is paid faster than it pays. dividendPayoutRatioTTM of 18.0% is consistent with the dividend and earnings figures. priceToEarningsGrowthRatioTTM of 1.39 is the single most useful ratio in the file for this name and is used in Section 6b.

Non-equity tripwire: checked and clear. MA is Class A common stock on the NYSE, beta 0.729, average volume 3.7M shares, a 52-week closing band of $471.55–$598.96 (a 27.0% range), and a growing rather than fixed dividend. Nothing resembling a preferred or baby-bond profile. (Mastercard does have Class B common stock outstanding, disclosed in the 10-Q's common-stock activity note and voted separately at the 2026 annual meeting per the 8-K filed 2026-06-17; it is a legacy structure of the 2006 listing, is convertible into Class A, and is included in the diluted count. It is not the security this dive covers, and unlike Visa's class B it carries no litigation-absorption mechanism.)

9. Technicals — a repair in progress, with the moving averages not yet agreeing

• Price $571.10, up $0.13 (+0.02%) — as close to unchanged as a session gets — on volume of 3.0M against a 3.7M average.
• −4.7% from the 52-week closing high of $598.96 and +21.1% above the 52-week closing low of $471.55. Maximum drawdown from peak over the trailing year is −4.7%, i.e. the current position: the high was set and the stock has not fallen further from it.
• Above both moving averages — but the moving averages disagree with each other. Price is +10.5% over the 50-day at $517.02 and +8.2% over the 200-day at $528.00. The 50-day is still 2.1% BELOW the 200-day. That configuration — price above both, short average below long — occurs only when a stock has been declining for most of a year and has turned sharply and recently. It is the technical signature of a repair in progress rather than a trend in force, and it is the single most important chart fact on this page.
• RSI 68.16 · MACD +15.38. RSI is the highest in this batch and within two points of the conventional overbought threshold of 70. MACD is strongly positive. Momentum is stretched.
• Relative performance, and the mirror image of its twin:

• A 22.3-point twelve-month deficit is the widest in this batch, and it arrived while earnings per share grew roughly 16% and operating margin expanded 330 basis points on an adjusted basis. The market de-rated the multiple by roughly four turns over a year in which the business improved. That is either a mistake or a verdict on the category; Section 7 argues it is largely the latter.
• Practical read for entry. The whole of the twelve-month gain, and more, arrived in the last three months: +13.1% against a 6-month figure of +2.8%. An investor buying today is buying after the bounce, into resistance 4.7% overhead, with RSI at 68. The 50-day moving average at $517.02 is 9.4% below and is the first meaningful support; the 200-day at $528.00 is 7.5% below; the company's own first-half repurchase price of about $506 is 11.5% below. There is no urgency to buy here and no reason to sell.
• Beta 0.729 — the lowest in this pair. A +13.1% three-month move on a 0.73-beta name is a large move in risk-adjusted terms and is more likely to consolidate than to extend.

10. Insiders — one officer, one day, and nothing else

This table contains no information and we will not pretend otherwise. All eight transactions belong to one officer on one day. The pattern is the textbook exercise-and-sell: 19,800 options exercised at a $227.25 strike, then 18,711 shares sold in seven tranches at $529.45 to $538.09 as the order filled through the book. The net effect on his economic exposure was an increase of 1,089 shares (from 38,739 before the exercise to 39,828 after), because he retained slightly more than the shares needed to cover the strike.

No director bought. No other officer transacted. No position of any size changed hands. For a company with a $500.9 billion market capitalisation, this is noise.

The transaction that actually matters is the company's own. Mastercard repurchased 17.6 million shares for $8.9 billion in the first half of 2026 at an average of roughly $506 — 11.5% below today's close of $571.10 — and continued through 2026-07-27 with a further $0.7B. A company buying $9.6 billion of its own stock in seven months is a far larger and more informative signal than any Form 4 in this table. The qualification, from Section 4: part of that was funded by a $5.0 billion note offering, so it is a levered vote of confidence rather than a self-funded one.

Governance note, from the 8-K filed 2026-06-17: at the 2026 annual meeting on 16 June, holders of 785,352,547 Class A shares elected the board, approved executive compensation on an advisory basis and ratified the auditor; two shareholder proposals — for a right to act by written consent and for cumulative voting in director elections — were not approved. Routine, and noted for completeness.

11. Verdict, kill-criteria and flip conditions

Hold.

Mastercard is the better-priced half of the best duopoly in financial services, and it is still not cheap enough to buy on 2026-08-04.

What is genuinely excellent. Value-added services and solutions at 40.6% of net revenue, up from 35.4% in 2022, growing 20% in the June quarter against 10% for the network — Mastercard is already two-fifths a services company. Operating margin expanding 330 basis points on an adjusted basis in 2025, because revenue grew 16% against adjusted operating expenses of 14%. Net margin 46.3%, return on invested capital 47.8%, interest coverage 27.9x, free cash flow converting at 94.7% of operating cash flow. Diluted shares down 12.2% in six years. Consensus EPS compounding 16.0% a year to FY2028. A GAAP-to-non-GAAP gap of only 3.2%, less than half Visa's, because the litigation provision is a smaller share of earnings. And a twelve-month underperformance of 22.3 points that has already done the de-rating work: this is a 16% grower at 24.8x forward.

What holds it at Hold. Base fair value of $621 is 8.7% above spot, against a bear case at $471 — which is, to fifteen cents, the 52-week closing low. The entry is poor: RSI 68.2 after a +13.1% three-month run, 4.7% below the high, with the 50-day moving average still below the 200-day. As-reported net revenue growth has decelerated for five consecutive quarters to 14.1%. Rebates and incentives grew 22% against net revenue of 14%. In the first half Mastercard returned $10.4B against $6.8B of operating cash flow, borrowing $5.0B to bridge it, and total debt rose from $19.0B to $24.6B — a fact no screen shows. The remaining buyback authorisation is $7.8B, or 1.6% of market capitalisation, against Visa's 4.1%. Five customers supply 21% of net revenue. And the litigation is under-accrued against the claims — $149M accrued for the US MDL while Block and Intuit seek "aggregate single damages in excess of $5 billion" — with, critically, no retrospective responsibility plan to absorb any of it.

Pre-registered KILL criteria — what would move this to Avoid:

• An adverse ruling or settlement in the Block and Intuit matters materially above the $149M accrual. Expert reports and summary-judgment briefing run through 2026. This is the single largest unquantified liability on the page and Mastercard's shareholders carry it directly.
• Success for the April 2026 putative class action seeking a declaration that the prospective release in the Damages Class Settlement Agreement — which by its terms runs through August 2028 — does not bar new damages claims. Mastercard and Visa filed a joint opposition; a loss reopens a decade of settlement architecture for both.
• Currency-neutral net revenue growth falling below 10% for two consecutive quarters. It has been 12% in each of the last two periods; the as-reported rate has fallen five quarters running, and if the currency-neutral rate follows, the growth premium disappears.
• Rebates and incentives continuing to grow more than 5 points faster than net revenue for a full year — they grew 22% against 14% in the June quarter. That gap is the duopoly's issuer-side price war showing up in the accounts.
• The loss of one of the five customers supplying 21% of net revenue, or a material repricing of one of those relationships.
• The December 2026 buyback authorisation coming in materially below $14.0B, or buybacks continuing to exceed operating cash flow into 2027 while debt rises further from $24.6B.

Pre-registered FLIP TO BUY (upsize) criteria:

• A price at or below ~$515 — roughly 22.4x FY2027E, essentially the 50-day moving average of $517.02, close to the company's own first-half repurchase price of about $506, and a level from which the risk-reward against a $621 base is better than 2:1. That is the accumulation zone, and it is 9.8% below today.
• Value-added services and solutions re-accelerating above 25% and passing 45% of net revenue, which would justify pricing this as a services business attached to a network rather than the reverse.
• Final approval of the Rules Relief Class settlement at the November 2026 hearing, combined with resolution of Block and Intuit near the accrual — which together would remove the largest open-ended liability and collapse the remaining GAAP-to-non-GAAP gap.
• The 50-day moving average crossing back above the 200-day while currency-neutral revenue growth holds at or above 12% — the technical and fundamental confirmations arriving together.
• Operating margin sustained above 60% for three consecutive quarters, confirming that the services mix carries network-grade economics rather than diluting them.

Mastercard versus Visa — and we prefer Mastercard, narrowly and for one reason. These are the two halves of the same duopoly and a reader deserves a direct answer rather than a shrug. On the numbers as of 2026-08-04:

The reason we prefer Mastercard is narrow and specific: the two trade at within two-tenths of a turn of the same forward multiple while consensus has Mastercard's earnings compounding roughly 240 basis points a year faster, its operating margin is expanding while Visa's is absorbing litigation, and its reported earnings depend less on non-GAAP adjustment (96.8% GAAP conversion against 92.0%). Mastercard is also 4.7% below its high rather than exactly at it, which at least leaves the possibility of buying at a discount to something.

Where Visa is plainly the safer of the two, and it matters. Visa's equity base is $37.9B against Mastercard's $7.7B; Visa's balance sheet can absorb a bad legal outcome and Mastercard's cannot in the same way. Visa has a retrospective responsibility plan — an escrow and a bank-held share class whose conversion rate falls with every litigation deposit, so that US bank holders, not class A shareholders, absorb interchange losses. Mastercard has no equivalent. Its shareholders pay the bills directly, and the largest single claim against it — Block and Intuit's "in excess of $5 billion" — is accrued at $149 million. Visa's remaining buyback authorisation is 4.1% of its market capitalisation against Mastercard's 1.6%, and Visa did not need to borrow to fund the last six months of repurchases. Visa's chart is also cleaner by every measure: a 1.0% maximum drawdown over twelve months, a 50-day above the 200-day, and RSI four points lower.

So: Mastercard is the better buy and Visa is the better hold. Neither is a compelling purchase at these levels, and an investor forced to own only one of them should choose based on whether the question being answered is "what compounds fastest from here" (Mastercard) or "what survives a bad decade with the fewest surprises" (Visa). We would not be embarrassed to own either. We would be embarrassed to chase either today.

Where MA fits in the Synthos Framework Portfolio. The quality-compounder sleeve, as a first-choice accumulation candidate rather than a today purchase. If already owned, a 2-3% weight is well supported by the mix shift and the return on invested capital, and there is no reason to trim. If not owned, the accumulation zone is $505-530 — the 50-day moving average, the 200-day, and the company's own repurchase price all cluster there, roughly 7-12% below today — and a staged entry beginning in that band is the disciplined route in. Note the house standing rule that mega-cap incumbents are guilty until proven innocent on a forward basis: Mastercard proves its innocence on mix (services at 40.6% of revenue), on margin (330 basis points of adjusted expansion), and on price relative to growth (24.8x for 16%), and fails it on entry timing (RSI 68 after a 13% run), on capital discipline (returns exceeding operating cash flow), and on exponential potential — which is a 4, because the margins are already at the ceiling.

Single biggest risk: litigation borne directly by shareholders. Not the everyday provision — $504M in 2025, $680M in 2024, $539M in 2023, all manageable — but the tail. Block and Intuit seek "aggregate single damages in excess of $5 billion" against a US MDL accrual of $149 million, with expert reports and summary judgment running through 2026; a separate group of two merchants seeking over $250M goes to trial in September 2026; over £0.5B (~$0.7B) of UK and Pan-European merchant claims remain unresolved with further claims filed in the second quarter of 2026; a UK collective action claims over £1B (~$1.3B); a Portuguese consumer action claims roughly €0.4B (~$0.5B) with trial in October 2026; and the UK liability appeal is heard in February 2027 with further issues tried in October 2027. The 10-Q states the accrual "represents Mastercard's best estimate of its probable liabilities in these matters and does not represent an estimate of a loss, if any, if the matters were litigated to a final outcome. Mastercard cannot estimate the potential liability if that were to occur." Visa's shareholders have a mechanism that makes banks absorb this. Mastercard's do not, and its stockholders' equity is $7.737 billion.

Most fragile assumption in the price: that the interchange litigation resolves for something close to the $149 million accrued. Everything else in the valuation — 12-13% revenue growth, 16% EPS growth, a 40.6%-and-rising services mix, a 60% operating margin — is well supported by the filings and by five quarters of prints. The litigation is the one variable where the disclosed number and the claimed number differ by more than an order of magnitude, where the company explicitly declines to bound the outcome, and where the structural protection its direct competitor enjoys does not exist. That is not a reason to avoid Mastercard. It is the reason the multiple is 24.8x and not 30x, and it is why a $621 fair value carries a $471 bear case underneath it.

Provenance & disclosures

• Traceability: 14 usable entity-level knowledge-base claims name Mastercard Incorporated — from 21 raw hits across all 51,928 records on the patterns \bMastercard\b and \bMaster ?[Cc]ard\b. The pipeline tagged 15 entity-level; 1 was excluded outright under the quarantined-misattribution policy (a 2025-12-10 bullish claim on early stablecoin positioning), and we note that the exclusion removes a bullish data point. Unlike the parallel search for Visa — where the case-insensitive word "visa" also matches the immigration sense and produced three false positives requiring discard — the Mastercard pattern has no homograph problem and every one of the 21 raw hits refers to the company. Six non-entity mentions are retained as weak, unweighted context. Stance mix on the 14 retained claims: 7 bullish / 3 neutral / 4 bearish; breadth 7 distinct sources (jordi_visser 3, we_study_billionaires 3, invest_like_the_best 3, all_in 2, real_vision 1, business_breakdowns 1, lyn_alden 1); range 2022-02-09 to 2026-02-27; net conviction mixed. The highest-skill source in the file (skill 2.0) appears three times — bearish on the business model twice in February 2026 and neutral-to-complimentary on Mastercard's stablecoin positioning in November 2025 — and that split is the reason the net conviction is mixed rather than positive. No management-lane (skill 0.5) claims appear in this file; the house half-weighting rule was applied to management framing wherever company guidance is quoted from the filings. All quoted claims are verbatim from the stored claim text.
• Data as-of: fundamentals — income statement 2026-06-30 (Q2 2026, reported and filed 2026-07-30, five days before this dive); annual balance sheet and cash flow 2025-12-31 (FY2025, 10-K filed 2026-02-11) with total debt superseded by the 10-Q as documented in Section 8 · estimates 2026-08-04 · prices 2026-08-04 ($571.10; 50-DMA $517.02; 200-DMA $528.00; RSI 68.2; MACD +15.38) · knowledge-base claims 2026-08-04. Every figure comes from the Synthos vendor data file for MA or from the SEC filings in filings/MA/; no figure comes from memory, recall or external retrieval.
• Filings consulted (11 targeted searches across three documents): 10-K filed 2026-02-11 (fiscal year ended 2025-12-31) — searched for the revenue disaggregation and its geographic footnote, the segment definition, the FY2025 net-revenue and rebate figures, the operating-expense and litigation-provision drivers, the buyback and dividend disclosures, the lease footnote, customer concentration, and the stablecoin and government-rails risk factors; 10-Q filed 2026-04-30; 10-Q filed 2026-07-30 (the primary source for this dive) — searched for gross dollar volume, cross-border and switched-transaction drivers, the payment network / value-added-services revenue split, rebates and incentives, operating-expense drivers, the debt footnote and total debt, the fair value of debt, the stockholders'-equity and buyback notes, the legal and regulatory proceedings note, liquidity and cash flows, non-GAAP reconciliation, and risk factors; 8-K filed 2026-06-08 (the $5.0B note offering and its full tranche structure); 8-K filed 2026-06-17 (2026 annual meeting results); 8-K filed 2026-07-30 (Q2 2026 earnings release).
• Filing-versus-vendor overrides (two, both resolved in favour of the filing): (1) total debt — vendor $19.000B at 2025-12-31, 10-Q $24.6B at 2026-06-30, a $5.6B increase from the June note offering and $710M of commercial paper; we use $24.6B, and the resulting net-debt estimate of roughly $13.7B replaces the vendor's $8.434B; (2) the March 2026 quarterly revenue — the vendor's earn_cal reports revenueActual of $9.277B for BOTH 2026-07-30 and 2026-04-30, identical to the dollar across two different quarters, which is arithmetically impossible; inc_q and the 10-Q's own growth reconciliation both put the March quarter at $8.398B, and we use $8.398B. Two vendor figures were positively confirmed by the filings: FY2025 total debt of $19.000B (10-Q, "$19.0 billion" at 2025-12-31) and FY2025 buybacks and dividends of $11.727B and $2.756B (10-K, "21.1 million shares... for $11.7 billion and paid dividends of $2.8 billion").
• Other data rejections (detailed in Section 8): priceToBookRatioTTM (89.77x), priceToFairValueTTM (89.77x), debtToEquityRatioTTM (4.39x), debtToCapitalRatioTTM (0.815), financialLeverageRatioTTM (10.28x) and returnOnEquityTTM (232.5%) all rejected as meaningless, because stockholders' equity of $7.737B is the residual of $85.035B of retained earnings less $83.224B of treasury stock — returnOnInvestedCapitalTTM (47.8%) and interestCoverageRatioTTM (27.9x) used instead, and two of the vendor rating model's three sub-scores of 1 disregarded on the same grounds; netDebt corrected from $8.434B to $8.102B on FY2025 data (omits $332M of short-term investments) and then estimated at ~$13.7B on the filing's debt; enterpriseValueTTM carries a +1.0% residual against a rebuild and is built on stale debt, so no EV multiple is quoted; inc_q gross margin of 96.7% in the June quarter rejected as a cost-allocation artefact, with operating margin used instead; all est niAvg rows rejected as implying a rising share count (882M → 903M) against an actual count falling from 1.006B to 883M, and net margins escalating to 54.2%; est EBITDA/EBIT rows pass the sanity check but are non-GAAP and are used for nothing; the seg_geo block rejected for trend analysis because the FY2025 "Americas" region was redefined from "North America" (10-K footnote: "Americas includes the United States, Canada and Latin America") and the FY2023→FY2024 step of +48% is definitional; the seg_prod block has a taxonomy break at FY2021/FY2022 and cannot be trended across it; the 10-K's $17.6B of 2025 operating cash flow against the vendor's $17.401B noted; cf_a D&A of $2.098B against inc_a D&A of $1.143B noted; closing-price 52-week extremes (tech) preferred over intraday (quote); tech.sma50 preferred over quote.sma50; the 0.7% spread between the vendor's three implied share counts (877.0M / 882.0M / 883M) noted as immaterial.
• Extraction limitation, stated plainly: the SEC text extractions preserve narrative prose but drop numeric table cells. The key-driver tables (gross dollar volume, cross-border volume, switched transactions), the net-revenue components table, the balance sheet, the liquidity table, the operating-expense table and the share-repurchase table appear as headings and footnotes with the numbers absent. Every filing figure in this dive comes from narrative sentences — hence percentage growth rates for volume rather than dollar volumes, and no independent verification of the cash balance at 2026-06-30.
• Estimate coverage: FY2026 24 analysts on EPS and 26 on revenue; FY2027 27 on EPS and 28 on revenue — the primary forward anchor and the deepest coverage in the block; FY2028 15 and 25; FY2029 10 and 12; FY2030 6 and 12. No conclusion in this dive rests on the FY2029 or FY2030 rows.
• Peer note: the vendor peer set is ALLY, AXP, COF, PYPL, SOFI, UPST, V. Four (Ally, Capital One, SoFi, Upstart) are balance-sheet lenders and credit underwriters whose earnings are net interest income and whose risk is credit risk — not comparable to a network that takes no credit risk, and no inference is drawn from them. American Express is a closed loop carrying card-member receivables; PayPal is a genuine digital-commerce competitor on a different model. Visa is the only true comparable, and it is compared directly and at length in Section 11. No peer multiples are supplied in the file, so no peer-multiple table is drawn.
• Fair-value caveat: the $471 / $621 / $700 anchors are multiples of FY2027 consensus EPS of $22.993 (20.5x / 27.0x / 30.4x), cross-checked against FY2028E of $26.722 and against a GAAP-equivalent FY2027 EPS of roughly $22.23. Stated arithmetic, not a discounted cash flow. The base case requires roughly 2.2 turns of multiple expansion off today's 24.8x; if the multiple simply holds, fair value is $571 — today's price. The bull case requires 30.4x and is flagged as the fragile leg. The bear case of $471 coincides, to fifteen cents, with the 52-week closing low of $471.55.
• Timing: Q2 2026 was reported 2026-07-30 and the 10-Q filed the same day, five days before this dive, so the most recent quarter is fully reflected. The next print is 2026-10-29, 86 days away — no earnings banner applies, and this dive states that explicitly rather than leaving it inferred.
• Accessibility note: all tables are labelled, all emphasis is carried by text rather than colour, and every figure is stated in full rather than encoded in a chart.
• Not investment advice. Independent research, educational and informational only, never personalised. No recommendation to buy, sell or hold any security is made to any person.
• Version: 2026-08-04-full.