Payments: card networks and card issuers. MA / AXP. Run 23 September 2026. Built from Mastercard 10-Ks FY2023 to FY2025 and 10-Q Q2 2026, American Express 10-Ks FY2023 to FY2025 and 10-Q Q2 2026, both companies' Q2 2026 earnings calls, Mastercard's Goldman Sachs conference remarks (10 September 2026), recent 8-Ks, and cited web sources for regulation. Events swept through 23 September 2026; most recent events checked: AXP CFO update at a 16 September 2026 conference (revenue growth guided to about 10%, Global Business Travel stake sale pending) and Mastercard's close of the BVNK stablecoin acquisition (3 August 2026). Figures in USD as reported; both companies report on calendar years. Not a valuation and not a recommendation.
Three Mastercard cells should carry the next five years, and each has its own structural driver. Cross-border is priced as well as volume-driven: in Q2 2026 cross-border assessments grew 20% on 12% volume growth because of pricing in international markets and mix. Value-added services were $13.3B in 2025 and about 40% of revenue. They are sold into the switched flows Mastercard already carries (about 60% are network-linked) and grow with fraud, tokenization and now agentic-commerce demand. International consumer payments still have cash to displace (roughly 50% of Japanese retail is still cash) and new domestic-switching mandates to serve, such as the UAE (FY2025 10-K; Q2 2026 call; Goldman Sachs conference, 10 September 2026). AXP has one strong cell, US premium consumer: USCS revenue grew at 11.3% a year over 2023 to 2025, and the Platinum refresh pushed US consumer billings up 11.4% in Q2 2026, the fastest since 2018. But that growth is repriced product by product, and US Commercial Services billings grew only 3% in 2025.
An extra Mastercard transaction mostly costs switching capacity: data processing and telecom stayed at 3.9% to 4.0% of revenue from 2023 to 2025, while operating margin rose from 55.8% to 57.6% and reached 60.2% in Q2 2026 (MA FY2025 10-K; 10-Q Q2 2026). Extra AXP revenue brings its own variable cost with it. Rewards, business development and Card Member services went from 41.3% to 42.8% of revenue between 2023 and 2025 and reached 44.7% in H1 2026, when Card Member services rose 49% as the Platinum refresh was funded (AXP 10-Q Q2 2026, computed). The two answers point the same way, but there is one tension to watch. Mastercard's own net price is increasingly handed back: rebates and incentives rose 22% in Q2 2026, while payment network net revenue rose 10%. Services and pricing are covering that gap for now.
AXP's weak point is its balance sheet combined with fixed benefit commitments. It holds $151.8B of Card Member loans and $213.9B of loans and receivables, 41% of billed business comes from small business and corporate clients, and 26% of spend is T&E. In a downturn, discount revenue, credit losses and net interest income ($17.4B) all move against it together, while lounge and statement-credit costs stay fixed for the product cycle. A 10% APR cap has been proposed but not enacted (S.381). If it passed, it would hit that interest income directly (AXP FY2025 10-K; congress.gov). Mastercard's weak point is its net price, and it erodes slowly. The first threat is the rules settlement, preliminarily approved on 9 June 2026 with final approval scheduled for November 2026. It lets US merchants decline premium Mastercard and Visa credit cards and surcharge them. The second is Block and Intuit's opt-out claims of more than $5B. The third is issuers that build or buy their own rails: Capital One's debit portfolio migration off Mastercard was basically complete in Q1 2026 (MA 10-Q Q2 2026; Q2 2026 call; Payments Dive). A recession hits Mastercard only through lower volume. It hits AXP through volume, credit losses and fixed benefit costs within the same few quarters.
These six cells cover effectively all of both companies' revenue. Mastercard does not report revenue by product, so its cell sizes use GDV and reported regional revenue. Each score is argued in the tabs below.
| Cell (product x region) | Mastercard | AXP | Why (one clause, sourced) |
|---|---|---|---|
| US consumer credit and charge | 3 | 4 | AXP USCS revenue +11.3% CAGR 2023-25, fee per card $92 to $131, retention flat through a $200 Platinum fee rise; Mastercard US credit GDV +10% in Q2 2026 but premium-card economics exposed to the settlement (both 10-Ks and Q2 2026 calls) |
| US debit | 2 | 0 | Mastercard US debit GDV +1% in Q2 2026 after losing Capital One's debit portfolio (+8% excluding it); AXP has no debit product (MA Q2 2026 call) |
| US commercial and SME | 4 | 3 | Mastercard commercial GDV +11% and commercial cards +14% in 2025 (worldwide; the US split is ND), with new virtual-card wins (Truist); AXP CS billed +3% and proprietary cards 15.4M to 15.3M (both FY2025 10-Ks) |
| International consumer card | 5 | 3 | Mastercard Asia Pacific/Europe/Middle East/Africa revenue $18.7B, +15.9% CAGR, and non-US GDV +10%; AXP ICS +11.6% CAGR at a 12.3% pretax margin, subscale on its own admission (both FY2025 10-Ks) |
| Cross-border travel and e-commerce | 5 | 3 | Mastercard cross-border assessments $12.0B, +19.6% CAGR, priced above volume growth; AXP T&E is 26% of billed business, +8% in 2025, with no FX fee split (ND) (both FY2025 10-Ks) |
| Services, network and merchant | 5 | 2 | Mastercard VAS $13.3B, +19.8% CAGR (about 3 points of 2025 growth acquired); AXP GMNS revenue +2.4% CAGR while the discount rate slid from 2.29% to 2.23% (10-Ks; AXP 10-Q Q2 2026) |
Each score is tied to an exhibit in the three tabs. Scores are per cell and are not summed. The lens verdicts are in the Three Answers above.
Normalization: AXP's segments map directly onto cells (USCS = US consumer, CS = US commercial and SME, ICS = international card, GMNS = network and merchant services). Mastercard reports one segment, with revenue split by category (payment network, value-added services) and by region (Americas; Asia Pacific, Europe, Middle East and Africa). It also says the US was about 29% of 2025 net revenue. Mastercard US and international revenue below are therefore computed from that rounded share and marked approximate. VAS is a separate cut that overlaps both geographies.
Revenue bases also differ. Mastercard reports revenue net of $20.5B of rebates and incentives paid to issuers and other customers in 2025. AXP reports revenue net of interest expense, with cardholder rewards and partner payments below the revenue line. Scale is therefore compared on volume: Mastercard GDV was $10.6T in 2025 against AXP billed business of $1.67T, 6.3 times larger.
| Cell | MA rev (% of total) | MA growth | AXP rev (% of total) | AXP growth | Margin signal / leader |
|---|---|---|---|---|---|
| US consumer credit and charge | US total ~$9.5B (29%), all products (approx.) | US GDV +6%; Q2'26 US credit +10% | $34.8B (48%) | +11.3% CAGR | AXP pretax 19.6%; AXP leads on growth and economics per account |
| US commercial and SME | ND (worldwide commercial GDV $1.4T) | +11% worldwide | $16.9B (23%) | +7.0% CAGR | AXP pretax 21.7%, but growth is mostly interest income (NII +24% in 2025); Mastercard gaining |
| US debit | in US total | +1% Q2'26 (+8% ex Capital One) | 0 | n/a | Mastercard only; losing share to issuer-owned rails |
| International card and cross-border | ~$23.3B (71%) (approx.) | ~+15% (approx.); cross-border assess. +19.6% | $13.0B (18%) | +11.6% CAGR | AXP ICS pretax 12.3%; Mastercard leads clearly |
| Services, network and merchant | VAS $13.3B (41%, overlaps) | +19.8% CAGR | $7.8B (11%) | +2.4% CAGR | AXP GMNS pretax 51.1% but falling (-10% in 2025); Mastercard leads |
Mastercard by region, 2025: Americas $14.0B (+12.3% CAGR 2023-25) and Asia Pacific, Europe, Middle East and Africa $18.7B (+15.9%). AXP by region: US $56.0B (+9.0%) and EMEA, APAC and LACC together $16.5B (+10.2%), before unallocated costs (both FY2025 10-Ks, Note 22 and Note 23).
Mastercard value-added services (mostly organic). VAS grew 23% in 2025, of which 3 points came from 2024 acquisitions, and 18% currency-neutral in Q2 2026. The drivers named were security, digital and authentication, consumer acquisition and engagement, insights, and pricing. It costs little to grow: adjusted opex is guided to low double digits against revenue at the high end of low double digits (Q2 2026 call).
Mastercard cross-border and international (organic, plus BVNK). Cross-border volume grew 15% in local currency in 2025, while assessments grew faster through pricing. Mastercard won several hundred flips and deal expansions in H1 2026, including the renewal of Chase Freedom Flex and exclusivity at Saudi National Bank. BVNK ($1.5B, closed 3 August 2026) adds stablecoin rails for B2B and remittance flows. Its revenue contribution is ND (MA 10-Q Q2 2026; GS conference).
AXP US consumer (organic). This is price and mix. Net card fees rose 18% in 2025 and 15% in Q2 2026 and are guided to exit 2026 in the high teens. The cost of that growth is Card Member services, up 27% in 2025 and 50% in Q2 2026. Management expects that cost growth to slow as it laps the refresh from Q4 2026 (Q2 2026 call).
AXP international (partly acquired from 2026). ICS billed business grew 14% in 2025. Swisscard became wholly owned and consolidated on 12 January 2026, and the proposed TheFork acquisition would add 50,000 European restaurants to its dining network, so part of 2026 ICS growth is bought.
AXP sets both prices, the merchant discount and the cardholder fee, because it runs a closed loop. The merchant price is slipping: the discount rate fell from 2.29% (2023) to 2.23% (Q2 2026) on mix and capped interchange elsewhere, while average fee per card rose from $92 to $131. Mastercard sells through issuers and acquirers but still realizes price. In Q2 2026 domestic assessments grew 10% on 8% GDV growth, and cross-border assessments grew 20% on 12% volume growth. It hands a rising share back, though: rebates and incentives went from $15.2B to $20.5B over 2023-25 (37.7% to 38.5% of gross revenue) and grew 22% in Q2 2026. Five customers generate 21% of net revenue (MA FY2025 10-K; 10-Q Q2 2026).
For payments companies the critical inputs are funding, partners and technology. AXP funds $224.8B of loans and receivables with $152.5B of customer deposits at a 10.5% CET1 ratio. Cobrand partners account for 26% of billed business and 36% of loans, and Delta alone for 13% and 21%. Mastercard needs no credit funding and runs its own switch. Its biggest input costs are people and data processing ($7.3B and $1.3B in 2025). Mastercard is clearly the more resilient; AXP's largest single input dependency is Delta.
In US premium consumer cards, AXP competes with bank issuers (Chase, Capital One, Citi) that run on Visa or Mastercard, and it is taking share there, judging by its fee and spend growth. AXP itself says Visa and Mastercard are larger than it in most countries by purchase volume (AXP FY2025 10-K). Mastercard's rivals are Visa (larger), issuer-owned networks (Capital One now owns Discover, approx., unverified), domestic and sovereign schemes, and new rails including stablecoins. Its response is to become the switch or technology layer inside those schemes, as in the UAE and South Africa (Q2 2026 call).
US consumer: an APR cap would hit AXP's $12.5B of USCS net interest income, while Mastercard's exposure is indirect, through issuer economics and rebates. US network pricing: CCCA routing (reintroduced January 2026, not voted) and the rules settlement fall on Mastercard, while AXP's three-party model sits outside the routing mandate (inferred from the bill's scope). US commercial and SME: AXP carries the credit and Mastercard does not. International: Mastercard faces domestic sovereignty schemes, and AXP faces EU and Australian rules on three-party networks and surcharging. Services: Mastercard's opt-out litigation (Block and Intuit, over $5B claimed, briefing through 2026).
Two-sided ubiquity (High). $10.6T of GDV, 175.5B switched transactions in 2025 and 3.7B cards (FY2025 10-K; Q2 2026 call). Acceptance and issuance at that scale cannot be bought. Rules, tokens and switching (High, but litigated). Tokens were on over 40% of switched transactions in Q2 2026 and contactless on 80% of in-person ones. The franchise rules are exactly what the MDL settlement is changing. Services attached to the flows (Medium-High). About 60% of VAS is network-linked.
Premium membership brand plus closed loop (High). Average fee per card rose from $92 to $131 while retention stayed flat through a $200 Platinum fee increase, and 65% of new consumer accounts come from Millennials and Gen Z (Q2 2026 call). Partner and experience web (Medium). Delta, Marriott, Hilton, Accor, Resy and Tock co-fund the value, but contracts are portable. Credit selection (Medium-High). It had the lowest projected card loss rate in the Fed's 2026 severely adverse scenario (Q2 2026 call).
Mastercard: its customers are banks. The top five are 21% of net revenue ($6.9B), and rebates are the price of keeping them. Losses do happen: Capital One moved its debit portfolio away. Mastercard also wins them, with several hundred flips in H1 2026 that management says bring "trillions of dollars in incremental volume" over a decade (Q2 2026 call). AXP: it faces no dominant merchant and owns its cardholders, but its partners are concentrated, with cobrands at 26% of billed business. It has exercised pricing power over cardholders, with fee per card up 12% to 14% a year. On the merchant side it is losing price, as the discount rate is down 6bp in two and a half years.
Mastercard: its inputs are people, technology and data processing, and it needs no funding. Pass-through is strong: operating margin held at 55% to 58% through the 2023-25 inflation period, and adjusted opex is guided to grow a little slower than revenue. AXP: its inputs are funding, partners and technology. Interest expense was $8.2B in 2025, deposit pricing competes with banks, and partner contracts are fixed for years. Pass-through is partial, because fees reprice only as cards renew.
| Cell | Route control | Pocket price | Continuity | Outcome (share / margin) | Confirming KPI |
|---|---|---|---|---|---|
| US consumer credit | AXP goes direct to cardholder and merchant; Mastercard goes through issuers | AXP well above (2.23% discount rate plus $131 fee) | AXP: partner renewals. Mastercard: settlement, CCCA, issuer-owned rails | AXP share up, margin flat as benefits absorb fees; Mastercard share flat, net price under pressure | AXP USCS billed growth vs Mastercard US credit GDV growth |
| International and cross-border | Mastercard through local issuers and domestic switches; AXP proprietary plus network partners, subscale | AXP above; Mastercard priced up in cross-border | Mastercard high; AXP exposed to EU and Australian three-party rules | Mastercard share and margin up; AXP share up from a small base, ICS pretax margin 9.3% to 12.3% (2023-25) | Mastercard cross-border assessments growth; AXP ICS pretax margin |
| Services, network and merchant | Mastercard VAS on its switched base; AXP GMNS partners | Mastercard priced for value (tokens, security); AXP discount rate falling | Mastercard high | Mastercard up / up; AXP flat / down (GMNS pretax -10% in 2025) | VAS growth; GMNS revenue |
Tension: AXP grows fastest in US premium, where its pricing power is strongest, but it also has to spend the most there to keep that power (Card Member services +50% in Q2 2026). Mastercard's growth runs the other way: it gains price at the network and pays some of it back to banks.
1. Balance sheet vs toll road (Major). AXP carries credit and funding: provisions were 7.3% of revenue and interest expense $8.2B in 2025. Mastercard carries neither. This is the model itself and cannot be closed. 2. Scale (Major). Mastercard's GDV is 6.3 times AXP's billed business. Money cannot close that within a decade. 3. Who gets the recycled value (Moderate). Mastercard gives 38.5% of gross revenue to issuers, while AXP gives 42.8% of revenue to cardholders and partners, plus 7.3% in credit costs. AXP narrows this gap only if fee growth outruns benefits after the Platinum refresh is lapped, which management expects from Q4 2026.
Mapping, with R&D ND for both because neither reports it. Mastercard: revenue = net revenue after rebates and incentives. COGS proxy = data processing and telecommunications, a component of G&A. SG&A = total operating expenses less data processing and litigation provisions. EBIT = operating income. AXP: revenue = total revenues net of interest expense. COGS proxy = Card Member rewards + business development + Card Member services + provisions for credit losses. SG&A = marketing + salaries and employee benefits + other, net. EBIT = pretax income. The two companies book customer payments differently: Mastercard nets its rebates out of revenue, while AXP books rewards and partner payments as expenses. That shrinks Mastercard's revenue base and inflates its SG&A ratio, so a gross-basis check is shown below.
| % of revenue, 3y avg | Mastercard | American Express | Gap | What drives it |
|---|---|---|---|---|
| COGS (proxy) | 4.0% | 49.6% | 45.6 pts | AXP pays rewards, benefits and credit losses; Mastercard's marginal cost is switching |
| R&D | ND | ND | ND | Not reported by either |
| SG&A | 37.8% | 31.7% | 6.1 pts | Revenue-base artifact: Mastercard is 22.7% on gross revenue (2025); AXP falling (33.2 to 30.8) |
| Gross margin (proxy) | 96.0% | 50.4% | 45.6 pts | Business model, persistent in all three years |
| EBIT margin | 56.3% | 18.7% | 37.6 pts | Mastercard 58.3% before litigation; both improved over the period |
| Year | MA rev $M | COGS p. | SG&A | EBIT (ex-lit.) | AXP rev $M | COGS p. | SG&A | EBIT |
|---|---|---|---|---|---|---|---|---|
| FY2023 | 25,098 | 4.0% | 38.0% | 55.8% (58.0%) | 60,515 | 49.4% | 33.2% | 17.4% |
| FY2024 | 28,167 | 4.0% | 38.3% | 55.3% (57.7%) | 65,949 | 49.2% | 31.2% | 19.6% |
| FY2025 | 32,791 | 3.9% | 37.0% | 57.6% (59.2%) | 72,229 | 50.1% | 30.8% | 19.1% |
Gross-basis check, FY2025: Mastercard handed back rebates of $20.5B, 38.5% of $53.3B gross revenue. Its SG&A was 22.7% of gross revenue and its operating income 35.4%. AXP's rewards, business development and Card Member services were 42.8% of revenue, plus 7.3% in provisions. Profit per dollar of volume: Mastercard 0.18% of GDV (operating income), AXP 0.83% of billed business (pretax). H1 2026: Mastercard operating margin 59.4%, AXP pretax margin 20.4% (both 10-Qs, computed).
The most persistent difference is the gross-margin proxy: about 46 points in every year from 2023 to 2025. The Step 1 and Step 2 mechanisms drive it, not one-offs. AXP owns the cardholder, the credit line and the reward liability. Mastercard rents its rails to banks that carry all three and pays them rebates to stay. The SG&A line looks like a contradiction, since Mastercard is higher at 37.8% vs 31.7%, but it is an artifact of the revenue base: on gross revenue Mastercard's overhead is about 23%, so this is not a discipline gap. The cost data confirm the power map in both directions. Mastercard's power over the system shows up as a 56% margin. AXP's power over its customers shows up as 4.6 times Mastercard's profit per spend dollar, earned with a regulated bank balance sheet. The one line to watch at AXP is the rewards, partner and benefits ratio, which was 44.7% in H1 2026 against 42.3% a year earlier. At Mastercard it is rebates as a share of gross revenue.
| Metric | Threshold | By when | If it hits, it favors | Where published |
|---|---|---|---|---|
| Mastercard VAS net revenue growth, currency-neutral | 15% or more | FY2027 (Feb 2028) | Mastercard | MA 10-K MD&A |
| Mastercard rebates and incentives as % of gross revenue | 40% or above (vs 38.5% in 2025) | FY2027 | AXP | MA 10-K MD&A (rebates disclosed in text) |
| Mastercard US GDV growth, local currency | Below 4% | FY2027 | AXP | MA quarterly earnings releases |
| AXP rewards + business development + Card Member services as % of revenue | 41.5% or below (vs 42.8% in 2025, 44.7% in H1 2026) | FY2027 (Feb 2028) | AXP | AXP 10-K income statement |
| AXP consumer and small-business net write-off rate (principal only) | Above 2.5% in any quarter | Through Q2 2028 | Mastercard | AXP 10-Q credit statistics |
| Rules settlement final approval and any appeal | Final approval without change (hearing November 2026) | Q4 2026 to 2027 | AXP (narrows its checkout disadvantage) | MA 10-Q legal note; Payments Dive |