Mastercard

IThe business

Mastercard runs the network that connects a card, a store and two banks. It authorises, clears and settles transactions between the bank that issued the card and the bank that serves the seller. It does not issue cards, extend credit or set what cardholders and sellers pay; those belong to its customers.³ In 2025 it switched 175.5 billion transactions, up 10%, and carried $10.6 trillion of gross dollar volume, up 9% in local currency. At 30 September 2026 its customers had issued 3.8 billion Mastercard and Maestro cards.³⁵¹ About 71% of 2025 net revenue came from outside the United States.³

Net revenue was $36.3B in the twelve months to September 2026, up 15.3% C (calculated), and has grown 13.2% a year since 2016. The operating margin rose from 53.5% in 2016 to 57.6% in 2025 and 58.6% in the TTM C (calculated).^3525326 Capital spending including software was 3.7% of net revenue in 2025 and stock pay 1.8% C (calculated).³

How Mastercard is paid

Mastercard reports two kinds of revenue. The payment network bills customers its standard pricing, which the company calls assessments: a fee on dollar volume, a fee on each switched transaction and a charge on cross-border volume. It then pays rebates and incentives back to customers for volume thresholds and for new and renewed contracts, and records them against revenue. Value-added services and solutions, the second line, hold security, data and analytics, processing and marketing services, after any rebates attributed to them.³

In the TTM assessments were $44.7B and rebates and incentives $23.6B, so the payment network kept $21.1B; services added $15.2B, for net revenue of $36.3B. The rebates were 52.9% of assessments C (calculated).³⁵²⁵³

Cross-border is the fastest-growing part of the price list. Cross-border assessments were $13.8B in the TTM, 31% of the total, and grew 22% a year from 2022 to 2025 against 8% for domestic assessments; from the 2020 trough the rates are 28% and 11% C (calculated).^3525322 Cross-border volume grew 11% in the September 2026 quarter, down from 15% in the September 2025 quarter.⁵¹⁸

What Mastercard earns on each $100

Net revenue per $100 of gross dollar volume rose from 22.3 cents in 2016 to 30.8 in 2025 and about 32.0 in the TTM C (calculated).^352567512628 Mastercard changed its revenue categories in its 2022 10-K, so the rebate figure is total rebates before 2022 and payment-network rebates after it. Volume includes cash withdrawals, which were 16% of gross dollar volume in the first nine months of 2026 and grew 0.4% in the September quarter, against 8.9% for purchase volume, so the shift from cash to cards is still a source of growth.⁵¹²²

From 2022 the series can be split into what the network keeps and what services add. The standard price per $100 of volume rose from 32.77 cents in 2022 to 37.62 in 2025. Customers took 4.08 cents of the 4.85 cent rise as rebates, 84% of it, and the network kept 0.76. Services added 2.89 cents, so they supplied 79% of the 3.65 cent rise in net revenue per $100 C (calculated).²²³ By the TTM the network kept 18.6 cents and services 13.4, of net revenue of 32.0 cents.⁵²⁵³

Customers took 84% of the rise in the standard network price from 2022 to 2025, and services supplied 79% of the rise in net revenue per $100 of volume.

Two cautions apply to the split. Some services are sold with the network and priced against it, so the split is an upper bound on services and a lower bound on the network.³ The standard price per $100 is also a yield, not a list price: it includes cross-border volume, which grows faster than the total, switched transactions, which grew 10% in 2025 against 9% for volume, and the effect of currency.³ The finding holds over those effects, because the rebate share rose in every year from 2022 to the TTM C (calculated).^2235253

I think this is the central fact about the business today. The 84% describes Mastercard’s pricing power better than the usual phrase does, and it sets the three drivers of my valuation: volume, the share of the standard price that customers take back, and the growth of services.²²³⁵²

The share rose from 46.4% of assessments in 2022 to 52.9% in the TTM C (calculated). In the first nine months of 2026 assessments grew 16.0% and rebates and incentives 20.9%, so payment network net revenue grew 10.9% C (calculated). The gap narrowed in the third quarter: assessments grew 14.7% and rebates 18.2%, 3.5 points apart against 5.7 points in the first half C (calculated).^52532251

Who pays, and where

The five largest customers generated about $6.9B, or 21%, of 2025 net revenue, and none exceeded 10%. The United States was 29% of 2025 net revenue, and revenue grew 15.6% in the Americas and 15.1% in Asia Pacific, Europe, the Middle East and Africa in the TTM C (calculated).³⁵²⁵³ Mastercard reports one segment and no profit by line of revenue.³

Cycle or utility

Card volume tracks spending and cross-border volume tracks travel, so Mastercard has some cycle in it. In 2020 volume was flat in local currency and cross-border volume fell 29%. Net revenue fell 9.4%, operating income 16.4% and net income 21.0%, and the operating margin went from 57.2% to 52.8% C (calculated).³²⁴²⁵ The next year net revenue grew 23.4% C (calculated), and cross-border volume grew 32% in 2021 and 45% in 2022.²³²² The test shows a business that bends hard when travel stops and recovers within a year.

IIThe moat

I think Mastercard’s moat is a network effect, reinforced by scale in processing and by the switching costs of banks that have built on it. The moat is wide in volume and narrowing in price, and the growth in what Mastercard keeps comes from services, whose moat is thinner than the network’s.

A card is worth carrying if sellers take it, and a seller takes it if people carry it. Mastercard has 3.8 billion cards, enables transactions in more than 150 currencies and more than 220 countries and territories, and switches more than 70% of all transactions on Mastercard and Maestro cards, including nearly all cross-border ones.³⁵¹ A rival has to bring issuers, acquirers, sellers and cardholders at once. In calendar 2024 Mastercard carried 33.7% of the payments volume of the five networks in the Nilson chart in Visa’s 10-K, against Visa’s 56.5%, and no third network in the chart is above 7.4%; the chart leaves out UnionPay and local networks C (calculated).³⁷

The volume keeps coming. Gross dollar volume grew 9% to 14% a year in local currency in every year from 2016 to 2025 except 2020, when it was flat, and 2021, when it grew 21%, and the operating margin was between 52.8% and 58.6% in every year but 2018, when a $1.1B litigation provision took it to 48.7% C (calculated).^35253222426

The evidence against the moat is in the contracts. Network contracts generally run up to ten years and many relationships are not exclusive. The 10-K says:

Our customers can reassess their future commitments to us

Evidence it is widening

Services, tokens and other rails. The table gives the services data: $15.2B in the TTM, up 21.7%, and 41.9% of net revenue against 35.4% in 2022. About 40% of 2025 transactions were tokenized, which ties a card to a device or a seller through Mastercard’s systems and should make the banks and sellers that adopt it harder to move.³⁵² Mastercard runs account-to-account infrastructure of its own, Vocalink in the United Kingdom, Nets in Denmark and the Move platform, and paid $1.4B in cash for the stablecoin infrastructure firm BVNK, a deal that closed on 3 August 2026, with up to $0.3B more if targets are met; it reports no revenue for any of these.³⁴⁷⁵² Mastercard is buying the new rails.

Evidence it is narrowing

Capital One. Capital One owns Discover and has moved its debit cards from Mastercard to the Discover network. PYMNTS reported on 21 July 2026 that Capital One said the migration was complete and that it is testing credit cards on Discover.⁴⁶ Mastercard’s U.S. debit volume grew 1.4% in the September 2026 quarter against 10.1% for U.S. credit and 9.6% for debit outside the United States.⁵¹ In the first three weeks of October, the chief financial officer said, U.S. switched volume improved partly because the Capital One debit migration was being lapped.⁶⁰ Neither source gives Capital One’s reasons or the volume moved, and the proxy lists Capital One among Mastercard’s 2025 portfolio wins and renewals.¹² One customer moving its debit portfolio to a rival network it owns is a live test of switching costs.

New rails. The 10-K lists real-time payment systems such as Pix, UPI and FedNow, government-backed networks, wallets and stablecoins as alternatives, and says real-time schemes with lower or subsidised pricing add pressure on pricing; Mastercard gives no volume figures for any of them.³ A size for the attack: if cross-border pricing fell by a quarter, assessments would lose $3.4B, 7.7% of the TTM $44.7B, and at the 52.9% rebate share the network would keep $1.6B less, 4% of net revenue C (calculated).⁵²⁵³

The law. Mastercard earns its fees from assessments and does not receive interchange, so the law reaches it through the incentives that issuers demand and through the volume that moves when routing rules change.³ The November 2025 settlement cuts average U.S. credit interchange by 10 basis points for five years and lets sellers decline premium and commercial cards.¹⁷ The Credit Card Competition Act of 2026, which the President endorsed in January (Payments Dive), has had no committee action since 13 January, in the bill record to 10 August.⁴³⁴⁴ The valuation section sizes a routing mandate.

The moat is wide, and on my split nearly all of the rise in the standard price has gone to customers.

IIIThe competitive advantage

Mastercard’s advantage over Visa is growth and breadth, and its disadvantage is scale. It grew net revenue 15.3% in the TTM, to $36.3B, against Visa’s 14.8%, to $45.9B. Services were 40.6% of its 2025 net revenue against 27.3% of Visa’s C (calculated).^35253375440 Visa has the volume, the higher margin and the lower rebate bill, and the market prices the two alike, at 33.3 times TTM GAAP earnings each C (calculated).^55352535457

Most of the margin gap is rebates

Visa’s lead in margin is 2.6 points on GAAP figures, 61.2% against 58.6%, and 5.6 points before litigation provisions, 65.1% against 59.5% C (calculated). Operating costs before litigation are 24.5% of revenue before rebates at Mastercard and 25.0% at Visa, so close that on that base the gap is mostly the rebate share by construction. That is an identity and says nothing about cause. On net revenue the cost ratios are 40.5% at Mastercard and 34.9% at Visa. Mastercard’s larger services mix should raise its cost ratio, and Visa’s is 24.1% without $563M of severance booked in the June 2026 quarter.⁵⁴ Mastercard gives back 39.4% of its revenue before rebates and Visa 28.5%, and Mastercard’s share is a lower bound because it does not disclose rebates attributed to services.³³⁷

Arithmetic only: if Mastercard gave back Visa’s share and its costs stayed the same, net revenue would be $42.8B and the operating margin before litigation 65.7%, against Visa’s 65.1% E (estimate).^352533754 Part of the gap may be the mix of issuers, scale tiers and contract timing, which neither company discloses. One explanation is that the second network pays more to the banks that choose which network goes on a card. Each point of its standard network price that customers take back costs $0.45B of operating income, or $0.41 a share after tax, 2.1% of TTM adjusted earnings C (calculated).³⁵²⁵³

Mastercard pays litigation from cash; under a 2011 sharing agreement it pays 12% of a global interchange settlement with Visa and the banks, or 36% of one with only its issuing banks. Visa funds most of its covered litigation through an escrow that lowers the conversion rate of its class B shares, so its GAAP provision of 3.8% of net revenue overstates what class A holders bear, against Mastercard’s 0.9%.⁵²³⁷⁵⁴

Where Mastercard wins

Mastercard wins outside the United States, where Asia Pacific, Europe, the Middle East and Africa were 57% of 2025 net revenue and Latin America volume grew 13.5% in local currency in the September 2026 quarter, and on growth: from 2017 to 2025 its net revenue grew 12.8% a year and Visa’s 10.2% C (calculated).³⁵¹³⁷ Its services line is described above.

Where it is exposed

U.S. debit. The Capital One shift is a measured loss of volume. A 2023 civil investigative demand from the Justice Department about Mastercard’s U.S. debit program is in the 2025 10-K and the 2025 10-Qs and is absent from the 10-Qs for March, June and September 2026; no filing says why, and I do not treat the omission as a closing.^34103752

Europe and the United Kingdom. The trial court ruled against Mastercard on certain liability issues in June 2025, the appeal hearing is set for February 2027, and unresolved merchant damages claims exceed £0.5B. In October 2026 Mastercard and Visa were served with two proposed U.K. collective claims for merchants, about £1.5B on consumer domestic and cross-border transactions and £0.4B on online cross-border ones.⁵² A U.K. court upheld the Payment Systems Regulator’s power to cap card-not-present interchange between the United Kingdom and the European Economic Area in January 2026 (FStech); no cap level has been set.⁴⁵

IVManagement

Michael Miebach has been chief executive since 1 January 2021. He joined Mastercard in 2010 to run the Middle East and Africa, became chief product officer in 2016 and president in 2020, and earlier spent 13 years at Citigroup and three at Barclays. He is 58 and succeeded Ajay Banga, who led the company from 2010 to 2020.³¹²²⁴ Ling Hai became chief financial officer on 3 August 2026; he has been at Mastercard since 2010, ran Asia Pacific, Europe, the Middle East and Africa from January 2024, and replaced Sachin Mehra, who became chief business officer.¹⁴³¹⁶

Record

From 2019, the last year before the pandemic, to the TTM, net revenue rose 115% and GAAP earnings per share 141%, from $7.94 to $19.11 C (calculated).^2535253 The 2025 to 2027 objectives Mastercard set in November 2024 have been met so far:

Five-year total shareholder return to 31 December 2025 turned $100 into $164.50 for Mastercard, $196.16 for the S&P 500 and $203.47 for the S&P 500 Financials.³¹² The proxy does not explain the gap. The arithmetic does: earnings per share rose 159% from 2020 to 2025 while the year-end price to earnings multiple fell from 56.0 times, on pandemic-trough earnings, to 34.6 times C (calculated).⁴⁹³²⁴ Say-on-pay support was 94.9% in June 2026.¹³ Five of the twelve executive officers in the February 2026 10-K table changed role on 3 August 2026, a vice chair retires in October, and a new controller started the same day.³¹⁴¹⁵

Pay

Miebach’s pay was $35.4M in 2025, up from $30.1M and $25.7M, and 284 times the median employee’s $124,650. The annual bonus pays two-thirds on adjusted net income and one-third on adjusted net revenue, and his equity is 60% performance units, 20% restricted units and 20% options.¹²

The performance units pay on adjusted net revenue growth and adjusted earnings per share growth, averaged over three years, with a relative return multiplier against the S&P 500 that moves them by up to 50 points either way.¹² No return-on-capital measure is in the plan, and the proxy says:

We do not disclose forward-looking targets for our PSUs

The 2023 units scored 92.9% on the financial goals, because growth in 2024 and 2025 was below target, and a return multiplier of 142.0% at the 71st percentile of the S&P 500 lifted the payout to 131.8%: the share price, and not the financial result, put that award above target.¹² The plan says nothing about the capital that customers’ up-front payments tie up, which is capitalized and amortized against revenue, so the measure the plan pays on lags the cash.³⁵²

Ownership and sales

Form 4 filings from October 2025 to 19 October 2026 show open-market sales of about $69.7M by eight insiders and no purchases, all under Rule 10b5-1 plans. Miebach sold 48,628 shares for about $27.9M between 31 July and 5 August 2026, under a plan adopted on 3 November 2025, and 26,400 of them came from options exercised at $227.25 on the same days; McLaughlin sold 34,060 shares for about $18.1M, Mehra 14,710 for $8.5M and Hai 13,288 for $7.2M C (calculated).³¹ At 21 April 2026, before those sales, Miebach owned 53,144 shares outright, 23 times his $1.45M base salary against a requirement of six times, which the proxy says every named executive officer meets.¹² The sales net of the exercises were 22,228 shares, 42% of that holding, and if nothing else changed he still held about 30,900 shares, 14 times his salary at $635.70 C (calculated).¹²³¹

The third name in the proxy’s ownership table, after Vanguard and BlackRock, is the Mastercard Foundation, which began a seven-year diversification in March 2024: its stake went from 97.5M shares, 10.5%, in April 2024 to 65.2M, 7.4%, in April 2026, about $41.5B at $635.70 C (calculated).³⁰²⁹¹²

The bench and the board

Edward McLaughlin has run technology since May 2017. Merit Janow is the independent chair, and ten of the eleven board nominees are independent. The charter caps any holder at 15% of Class A or of voting power, and stockholders cannot act by written consent; a shareholder proposal for written consent got 30.1% of the votes cast in June 2026.³¹²¹³

Capital allocation

From 2016 to 2025 Mastercard bought 244 million shares for $69.5B at an average of $285, shares now worth about $155B at $635.70, and the diluted share count fell 17.7%, from 1,101M to 906M C (calculated).²⁸³ It paid $16.8B in dividends over the decade, together with the buybacks 97% of free cash flow C (calculated).³ The price paid matters more than the amount:

On my numbers the buybacks created value through 2024 and were close to value-neutral from 2025; at $635.70 each dollar buys $0.88 of my base value E (estimate). The cash yield on the shares is 2.8%, and the return on my base case at this price is 8.4% a year against the 9.0% I require, so a buyback funded by debt adds earnings per share more than it adds value E (estimate).³⁵²

In the TTM buybacks of $15.5B and dividends of $3.0B were 120% of free cash flow, and debt rose from $5.2B in 2016 to $23.9B at 30 September 2026, with $5.0B of notes issued in June 2026 C (calculated).³⁵²¹⁸ Stockholders’ equity is $6.7B against $95.2B of treasury stock.⁵²

Acquisitions took $10.9B over ten years, among them Nets for about $3.6B in 2021, Recorded Future for $2.7B in December 2024 and BVNK for $1.4B in 2026.^2335247 None of them reports its own return. The record on buybacks is good through 2024 and neutral from 2025; on deals it is unproven, because the results sit inside a services line that grew 22%.

Miebach is paid two thirds on adjusted net income and one third on adjusted net revenue, with no return-on-capital measure, so it pays him to grow earnings per share, which borrowing to buy back shares does.¹² Debt rose from $5.2B to $23.9B while buybacks and dividends took 120% of free cash flow.³⁵²¹⁸ He held 53,144 shares outright in April, 23 times salary, and sold 48,628 in July and August; insiders bought none.¹²³¹ The buybacks bought $2.90 of my base value per dollar in 2016 to 2021 and $0.88 at today’s price.

VFinancials

Mastercard has turned about 100% of its net income into free cash flow over a decade, 91% in the TTM, and its profit is partly hidden by litigation provisions while its cash flow is partly hidden by up-front payments to customers.³⁵² Both effects are in the numbers below.

GAAP operating margin was 57.6% in 2025 and 58.6% in the TTM, after litigation provisions of $504M and $330M. On Mastercard’s adjusted basis, which excludes litigation and one-time items, it was 59.2% in 2025 and 60.8% in the first nine months of 2026.³⁵²⁵¹⁷

Earnings quality

Adjusted earnings per share were $17.01 in 2025 against GAAP $16.52, and in the TTM $19.70 against $19.11, the difference being litigation provisions, losses on equity investments and tax effects, plus in the TTM a $202M restructuring charge from the first quarter of 2026, worth $0.18 a share C (calculated).^3525153 The effective tax rate rose to 19.4% in 2025 from 15.6% in 2024 as the global minimum tax took effect, and the adjusted rate was 20.1% in the September 2026 quarter.³⁵¹

Cash

Free cash flow, which I define as operating cash flow less purchases of property and equipment and capitalized software, was $15.4B in the TTM and 42.5% of net revenue C (calculated). It was 5% below the prior twelve months while net income was 19% higher.³⁵²⁵³ Part of that is a base effect: conversion was 110% of net income in 2025, the highest since 2017, against 91% in the TTM and 103% across ten years C (calculated).^3525311 The 10-Q gives the cause of the lag: higher customer incentive payments, cash paid for litigation settlements and timing of settlements with customers.⁵² Owner free cash flow is free cash flow less stock pay of $0.6B: $14.8B in the TTM and $15.8B in 2025, or $15.2B and $16.1B before interest, the basis of my model C (calculated).³⁵²⁵³

Up-front payments to customers are the line to watch. They are capitalized and amortized against revenue over the life of the contract, so cash leaves first and the revenue hit follows. Customer incentive assets were $11.6B at 30 September 2026 against $10.4B at December, and I do not know how much of the rise is timing. If up-front payments keep growing faster than revenue, the amortization that follows will weigh on net revenue growth, and the rebate share will rise further.⁵² The value does not depend heavily on this input: net reinvestment of 2.5% of revenue instead of 1.5% takes $14 off my base value, to $547 E (estimate). The lag flags earnings quality; it does not drive the value.

The balance sheet

Debt is $23.9B against $10.5B of cash and investments, a net debt of $13.5B after the $1.4B paid for BVNK on 3 August 2026. My model uses $14.1B, which also counts the $0.6B of stock repurchased from 1 to 26 October, the date of the share count. Operating income covers interest 26 times C (calculated), and tangible equity is negative.⁵²⁴⁷ Mastercard guarantees settlement between its customers, with a gross exposure of $94.6B at 30 September 2026, $77.8B after collateral, and losses historically low.³ That is a tail risk the cash flows do not show.

The legal docket

Mastercard accrued $262M for litigation at 30 September 2026, $156M of it for the U.S. interchange cases, against claims that dwarf it: Block and Intuit seek over $5B in single damages, U.K. and European merchants over £0.5B, a U.K. commercial-card collective claim over £1B and two new U.K. collective claims served in October 2026 for about £1.5B and £0.4B.⁵² The 2011 sharing agreement covers an adverse judgment or settlement of the U.S. cases, but the 10-Q gives Mastercard’s share only for settlements. The 36% share of $5B is $1.8B, $1.63 a share after tax C (calculated). I could not confirm that any share applies to a judgment or to the opt-out claims. The whole $5B costs $4.53 a share, 0.7% of the price. Trebled to $15B and paid in full it costs $13.60 a share, 2.1%. The 10-Q gives no estimate of a loss at trial. The docket matters more for what it can do to the price list, through interchange caps and routing rules, than for what it can cost in cash.

VIValuation

First, how this goes badly wrong

Working back from what would most hurt an owner, I find three. Customers keep the price: they took 84% of the rise in the standard price from 2022 to 2025, and if their share keeps rising 1.6 points a year my base value is $511.^3225253 Services stall: on my bear path they grow 14% falling to 3% and the base value is $452.⁵²⁵³ A routing rule passes: the Credit Card Competition Act is filed as S. 3623, and losing 15% of U.S. revenue by year 3 gives $517, or $307 on the bear case.^4435253

An owner takes out less than earnings show. Reported earnings are a 3.0% yield at 33.3 times. Free cash flow was $15.4B in the TTM, 91% of net income, and after $0.6B of stock pay owner cash is $14.8B, 2.7% of the $553.8B market value C (calculated).^5535253 Capital spending is only $1.4B; the real cost of standing still is up-front payments to customers. Net reinvestment, which includes them, ran at about 4.5% of net revenue against 1.5% in my cases. At 4.5% for 15 years my base is $519, not $561.⁵²⁵³

I cannot know how far customers’ share will climb, or how much of the $1.2B rise in incentive assets since December is timing.⁵² So I show a range: rebates alone move my base value from $511 to $608, and the price is above both. The $5B legal claim is $4.53 a share, which matters less.⁵²³

What the price implies

At $635.70 Mastercard trades at 33.3 times its TTM GAAP earnings and 27.7 times 2027 consensus earnings, which is vendor data retrieved in Q4 2026 E (estimate).^553525359 Its market value is $553.8B, and free cash flow is 2.8% of it C (calculated). Visa trades at 33.3 times.⁵⁵⁵⁴⁵⁷

Against its own ten years the multiple is not high. Today’s 33.3 times is below the median of 36.8 times at the ten year-ends, and the free cash flow yield of 2.8% is in line with the median of 2.8% C (calculated).^554935253 The market paid between 34 and 42 times earnings at seven of those ten year-ends. My base value of $561 is 29.4 times TTM earnings, below every year-end but 2016, which is the same fact as a price that needs a rate below the 9.0% I require.

At $635.70 my cash flows return 8.4% a year in the base case, 10.4% in the bull case and 6.5% in the bear case, against a 5.00% ten-year Treasury yield E (estimate).⁵⁵³⁶ The bear case keeps volume growing every year and growth of 3.5% after year 15, so it is a squeeze on price with volume intact; the tails table below tests volume.

Solved as one rate, the price needs owner free cash flow before interest to grow 9.9% a year for 15 years, from $15.2B to about $62B in year 15, with growth of 3.5% after that C (calculated). The start is free cash flow after stock pay and interest, $14.8B, plus $0.37B of net interest after tax (interest expense of $0.81B less investment income of $0.34B); the cases deduct net debt once, so a start after interest would count it twice. My cases are paths, so I show each as the constant rate that gives the same value: 4.1% for the bear case, 8.8% for the base case and 12.3% for the bull case. The base path compounds at 8.3% but is worth the higher constant rate because it grows fastest in the early years, which are discounted least. With growth after year 15 of 3.0% the implied rate is 10.3% and with 4.0% 9.3%; from 2025 owner free cash flow before interest of $16.1B it is 9.3%.³⁵²⁵³

At $635.70 the price needs about 9.9% growth in owner free cash flow for 15 years; my base case is worth what 8.8% a year would be and my bull case what 12.3% would be.

Why the decade will not repeat

Mastercard’s free cash flow grew 16.2% a year from 2016 to 2025 and 11.3% a year from 2019 to the TTM. The price asks for less than that, and my base case for less again. Revenue grew 13.2% a year, and the rest of the 16.2% came from a rising margin, a falling tax rate and rising cash conversion. None of the three can be repeated at that pace: the margin is near the 60% at which my base case starts, the tax rate is heading to the company’s 21% objective, and conversion is about 100%.³⁵²¹⁹

Revenue is the one input that can carry the decade forward, and I give it 8.5% a year over 15 years: 12.6% in year 1, in line with the company’s objective, fading to 5.3% in year 15.¹⁹

The discount rate

I use 9.0%, the rate on my Microsoft page. The capital asset pricing model with a 5.00% ten-year Treasury yield and Damodaran’s 4.20% implied equity risk premium gives 9.04% on Mastercard’s ten-year beta against the S&P 500 (0.96), 8.20% on the five-year beta (0.76) and 7.19% on the three-year beta (0.52) C (calculated).^36484950 Damodaran struck that premium with a 5.29% Treasury yield on 1 October; with it the ten-year beta gives 9.33% and my base value is $525. I take the ten-year window because it is the only one that spans the 2020 shock. It gives the highest of the three rates, 9.04%, which 9.0% rounds, and 9.33% with Damodaran’s own Treasury yield. The verdict depends on the rate: at 8.5% my base value is $624, within 2% of the price, at the five-year-beta rate $669, and at 9.5% $508.

Three cases

I build each case from volume down to owner free cash flow over 15 years, because the network’s runway is long and a ten-year model would put most of the value in the terminal year. Volume, the standard price per dollar of volume, the share of it that customers take back and services each grow on a path I set; the operating margin is before the litigation provision, which I charge as a share of revenue. Tax is 21%, the company’s own objective. Net reinvestment, which includes up-front incentive payments and working capital, is 1.5% of revenue; the TTM figure on the same definition was about 4.5%, against 1.4% in the TTM to June and a negative figure in 2025, because customer incentive payments ran ahead of revenue C (calculated). Acquisitions are 2.5% of revenue, about $1.0B in year 1, close to the $1.1B a year Mastercard spent on average over ten years; the ten-year record is 5.5% of revenue, and the bridge shows that case. The $1.4B BVNK price is already in the 30 September net debt, so year 1 holds a budget for the next deal; removing it adds $1.^352531947 Stock pay is a cost in the cash flows, so I add no unvested awards to the share count; dilutive options and units were 1M shares in the September 2026 quarter, about $0.6 a share.⁵² Growth after year 15 is 3.5% in every case, so the cases differ only in their operating drivers; the years after 15 carry 52% of base-case value.

The base case carries the last three years forward at a slower pace, and the table above gives its drivers. Customers’ share of the standard price climbs a point a year, to 58.6% in year 6 and 60.5% in year 10. Over the first six years they take 92% of the rise in the standard price, against 84% from 2022 to 2025, and the network’s net revenue per $100 grows 0.5% a year, against 1.4%; over 15 years it grows about 1% a year, from 18.6 cents to 21.5. The base is no more generous to the network than the record C (calculated).²²³ Services reach $64B, 51% of net revenue, against 41.9% now. The operating margin before litigation holds at 60.0% for five years and falls to 58.0% by year 15, because services grow faster than the network and carry people and acquisition costs the network does not; Mastercard discloses no profit by line, so this is a judgement.³⁵²

My base year 1 matches the company’s own aim and the market’s: net revenue up 12.6% against an objective of the high end of low double digits, and after-tax operating profit per share, less interest, 2% and 2% below consensus in the next two years at a 21% tax rate, or 0% and 1% below at 20%.¹⁹⁵⁹

The bear case is a squeeze on the price, with a $4B legal loss in year 2. The loss is my estimate: more than the $1.8B that the 36% settlement share of Block and Intuit’s $5B of single damages would be, and less than the $5B paid in full, because I could not confirm that the sharing agreement reaches a judgment C (calculated).³⁵²¹⁹ The network’s net revenue per $100 falls to 16.6 cents by year 10, and net revenue grows 4.6% a year.

The bull case is consensus and then some: its after-tax operating profit per share lands within about 6% of consensus in both years. Net revenue grows 11.3% a year, to $182B, and the value rests 56% on the years after year 15.⁵⁹ Weighted 25%, 50% and 25%, the cases give $572 a share, and weighted 35%, 50% and 15% $522 E (estimate). The weights are my choice and illustrative; the bull case is $272 above the base and the bear $227 below it, so equal weights tilt the mean upward. The price is above both.

What closes the gap

Single changes in my base case do not reach the price except the discount rate and the services path.³⁶⁴⁸ With services on my bull path the base value is $674, and on my bear path $452, the widest swing of any operating input I tested. Rebates matter less than I expected: holding customers’ share at 52.9% lifts the value to $608, and a rise of 1.6 points a year for ten years, the 2022 to 2025 pace, lowers it to $511. Both together, a flat share and the bull services path, give $721 E (estimate).

To reach the price, change one input in my base case:

Services growth decides the value more than rebates do: my bull path adds $113 a share and my bear path takes away $109.

Three items sit outside the cases. I add nothing for the $1.7B of strategic equity investments, about $2 a share, or for BVNK beyond its $1.4B price, which is already in net debt, or for stablecoin and agentic-commerce products, whose revenue Mastercard does not disclose.⁵²⁴⁷ Buybacks funded by debt are already in the net debt. And the values are dated Q4 2026, the price date; dated 30 September, the last balance sheet, my base value is $557. I discount each year’s cash flow at the end of the year; at the middle of the year the base value would be $587, 4.7% higher E (estimate).⁵⁵⁵²

The tails the cases leave out

The bear case is a squeeze on price with volume still growing. The largest policy tail would move volume: a routing mandate such as the Credit Card Competition Act. The U.S. was 29% of 2025 net revenue. If 10% of it were lost from year 2, with every lost dollar falling off profit, my base value would be $531, or $545 if costs fell in line; if 15% went by year 3, $517 E (estimate). Each stays above the margin-of-safety price of $421. A 15% loss on top of the bear case gives $307; replacing the bear case’s $4B legal loss with Block and Intuit’s single damages trebled and paid in full, $15B, gives $298, 29% under that price. These are sizes, not forecasts: Mastercard has not said what a mandate would do to its volume.^3525344

The margin of safety

• Price$635.70Close, Q4 2026 [55]
• Base value$561My estimate E (estimate)
• Margin of safety below$42125% under my base value C (calculated)

I set the margin of safety 25% under my base value; my rule is 25% for a business with net debt of 2.5% of enterprise value, interest covered 26 times and no customer above a tenth of revenue, 30% when one cycle or two customers drive results, and 35% when fixed charges such as leases sit ahead of the shares. Below $421, my $334 bear value is 21% lower, my base case would return about 10.6% a year and my bull case 13.1% E (estimate). That is 21.4 times TTM adjusted earnings and 18.3 times 2027 consensus.⁵²⁵³⁵⁹ The price sits 51% above the margin-of-safety price, so the stock would have to fall 34% to reach it. The intraday 52-week low of $464.52 was still above it.⁵⁸

Where this could be in 5 and 10 years

The value at each date is my model’s present value, at 9.0%, of the owner free cash flow after it, less $14.1B of net debt, per today’s 871M shares and in that year’s dollars. Owner cash is free cash flow before interest from October 2026. The return a year counts price growth and that cash, with the exit at my value in 2036 E (estimate).

The base stands at five years if net revenue reaches $61B from $36B, the operating margin stays near 60% and customers take 58% of the standard price. It stands at ten if revenue reaches $92B at a 59% margin, with services at $48B of it. Reinvestment stays at 1.5% of revenue and deals at 2.5%. At 9.0% the value rises about 5% a year, and the rest of the return is the $314 a share paid out.

The bear is a network that cannot raise its price. Revenue is $60B in year 10, customers take 63% of the price and a $4B legal loss lands in year 2. The bull is services compounding at 21% falling to 5%, revenue of $125B and a 62% margin.

The most likely way my base is wrong is services. It needs them to grow from $15B to $48B. If they grow 2 points a year slower, base value in 2036 is $784, not $894, and the return falls from 7.6% to 6.3%. The discount rate matters as much: at 10% the 2036 base is $753.

These are estimates of value, not forecasts of the share price. The market price can sit above or below value for years. The 7.6% is below the 8.4% above because that rate values the cash after 2036 at 8.4%, the rate that matches today’s price, and this one uses 9.0%.

The next best use of the same dollar is the closest comparable on this site, Visa. Its base return to year 10 is 7.1% a year on my Visa page, against 7.6% here, and both sit under the 9.0% I require. Business and price are two questions. The business turns 43 cents of each revenue dollar into free cash flow; the price is 13% above my base value.³⁵²⁵³

VIIThe case for the price

What most people miss

Mastercard’s fees are rising about twice as fast as I assume. In the first nine months of 2026 assessments, the fees before rebates, grew 16.0% and card volume 9.1%, both in dollars. So the fee on each $100 of volume rose about 6.3% C (calculated).⁵²⁵³⁵¹ Part of that is mix, and management expects some of the pricing effect to fade in the fourth quarter.⁶⁰ My base has 3.0% for year 1, my bull 4.5%. Rebates rose 21%, yet the network kept about 1.7% more per $100, against 1.0% in my base year 1 C (calculated).

The chief financial officer said 60% of services are network linked, so much of their growth arrives with card volume. The chief executive said Pinnacle Financial Partners will move its whole card portfolio to Mastercard from 2027.⁶⁰

The bull case

Mastercard takes a small fee on a rising share of the world’s spending, needs little capital, and gets stronger with each new card. Bill Ackman’s Pershing Square made that argument. Its 13F shows 2,124,646 shares worth $1.09B at 30 June, 5.6% of the value it listed, with none in its March filing.⁶²⁶³ Its half-year report for Pershing Square Holdings calls Visa and Mastercard “among the highest-quality businesses in the world”. It says fears over stablecoins, AI shopping agents and U.S. rules are misplaced, and estimates earnings per share growing 18% a year for three to five years, dividend yield included.⁶⁴

Akre Capital has more at stake. Mastercard was 14.3% of its ETF at 30 September, the largest position. Its third-quarter letter says AI shopping agents will need these networks, and that rivals have more reason to partner than to compete. It adds that the networks have only begun to earn money from tokens, which replace card numbers.⁶⁵

MarketBeat and Yahoo Finance reported new targets of $755 from JPMorgan, $750 from KBW and $743 from Goldman Sachs, under my $833 bull value. None gave reasons.⁶⁶⁶⁷⁶⁸ Berkshire Hathaway went the other way: 3,986,648 shares at the end of 2025, none at 31 March 2026. Its filings give no reason.⁶⁹⁷⁰

Against reality

The price is $75 above my base value. One change can close that: services growing 2.1 points a year faster than my base, 18.8% in the year to September 2027.

Two changes together look likelier: rebates stop at 54% of the standard price for good, and services run about a point a year above my base E (estimate). So if Mastercard meets its high-teens services objective through 2027, keeps going after it, and stops the rebate climb at 54%, with volume and margin on my base, the price is possible.¹⁹

Services grew 19% in the third quarter, on pace, though some came forward from the fourth.⁵¹⁶⁰ Rebates are the hard part. They took 52.9% over the last twelve months, 1.1 points short of 54%. But the third-quarter share rose 1.6 points on a year earlier, and management expects the fourth quarter higher than the third C (calculated).⁵²⁵³⁶⁰ At that pace it passes 54% within a year. It has risen every year since 2022, so the case needs something new. Mastercard gives no rebate guidance, so 54% is my line, not theirs.⁶⁰ Volume and cash are behind my base too: 7% against 8.4%, and cash conversion of 91% against about 100%.⁵¹⁵²⁵³ The objectives end in 2027, and 52% of my base value comes after year 15. Three things have to go right together: services above my base, rebates stopping near 54%, and volume back to my 8.4%. At even odds each, a guess and not a measurement, all three come to one in eight.

What they have to hit

Mastercard publishes the inputs for all five every quarter.

Where I come down

The fee growth moved me most, though not my values, because rebates have taken most of each past rise. Pershing’s stake did not: the shares were about $514 at 30 June, under my $561 base, so it says little about $635.70 C (calculated).⁶² Rebates that stop near 54%, not climbing a point a year, with services above my 16.7% and 15.7%, would raise my values.

VIIIRisks, and what would change my view

1. Rebates and incentives. They took 52.9% of standard network pricing in the TTM, up from 46.4% in 2022 C (calculated).³⁵²²² If customers take back 1.6 points a year for ten years, my base value falls to $511. A large customer that moves volume, or a renewal that resets terms, would show up here first.
2. Services. They supply 79% of the gain in net revenue per $100 and are 42% of revenue. Growth was 22% in the TTM and 19% in the September 2026 quarter, also 19% in constant currency; on my bear path, 14% falling to 3%, the base value is $452.⁵²⁵¹ Mastercard lists technology firms and consultancies among rivals in this line.³
3. Customers moving volume. Capital One shows that a customer can move a block of volume to a rival it owns, and it is testing credit cards on Discover with no volume or date given.⁵¹⁴⁶
4. Law and policy. A routing mandate is the largest tail: losing 15% of U.S. net revenue by year 3, all of it off profit, takes my base value to $517, above the margin-of-safety price of $421, and to $307 on top of the bear case E (estimate). A final-approval hearing on the settlement is set for November 2026, and a U.K. appeal on liability for February 2027.^52174445
5. Cash conversion. Free cash flow fell 5% against 19% higher net income, and TTM conversion is 91% of net income against 98% a quarter ago, because customer incentive payments and legal payments rose again. I still treat it as timing, but it is now two quarters running. If net reinvestment stayed at the TTM level of 4.5% of revenue for all 15 years, my base value would be $519.⁵²⁵³
6. People and execution. A new chief financial officer, a new chief services officer, a new controller, a vice chair retiring in October and the BVNK integration all land within three months.¹⁴¹⁵⁴⁷
7. Currency and supply of stock. Reported growth ran 2 to 4 points above currency-neutral growth in three of the last four quarters; in the September 2026 quarter both were 14%.⁵⁶⁷⁵¹ The Mastercard Foundation holds 65.2M shares, $41.5B, and is selling over seven years.¹²

Mastercard reported the third quarter on 29 October 2026, six days after the price date of the first version of this page, when EDGAR listed no third-quarter filing and the call was set for that day,³²³³ and the thresholds above can now be graded. Net revenue grew 14% to $9.8B, as reported and in constant currency, over the 12% bar. Rebates and incentives grew 18%, under the 25% line, and 3.5 points faster than assessments, against 5.7 points in the first half. Services grew 19% in constant currency, over 15%, and the operating margin before litigation was 60.5%, over 60%, though 59.5% on the TTM. Volume grew 7% in local currency, under the 8% bar and over the 6% line, and cross-border growth slowed to 11% from 12%. Adjusted earnings per share were $5.31 against the $5.14 consensus I noted at the price date.⁵¹³⁵ The cash did not follow. Nine-month operating cash flow was $11.9B against $12.6B a year earlier because up-front customer payments rose by $1.7B, and TTM conversion fell to 91% of net income from 98%. On the call, management said rebates as a share of assessments would be higher in the fourth quarter than in the third.⁵²⁵³⁶⁰ So the quarter confirmed the revenue and services path and left the cash question open. My base value moved from $556 to $561 because a quarter passed, net debt fell and starting revenue came in a little above my path; I changed no driver, and the thesis stands. The price rose 1.9% in the week to results day, to $635.70, so it sits 13% above my base value, against 12% before.¹²⁵⁵

Gaps in the record that move a number: I did not retrieve Mastercard’s own transcript of the third-quarter call, and the guidance I cite comes from a third-party transcript. The chief financial officer said on the call that 60% of services are network linked, without defining the term,⁶⁰ and I did not find a primary source for how much of services revenue is sold only with the network. The final-approval hearing in November has no day in a primary document, and the 10-Q says only that Mastercard has settled with every individual opt-out except Block and Intuit, not what the earlier trial found. The 10-Q does not say whether the 2011 sharing agreement reaches a judgment or the opt-out claims. The Justice Department demand dropped out of the 10-Qs without a reason. I did not update the Treasury yield, the beta windows or the Damodaran premium after the first version; the ten-year yield was 4.98% on 29 October against the 5.00% I used.^1236061