American Express (AXP)
Financiero / Pagos premium
The premium closed-loop payments network (sees both sides of the transaction) + lender to affluent customers — an exceptionally high-quality franchise (ROE ~34%, the Platinum brand, Berkshire holding 22%). At ~$337 it trades at ~20× earnings, a full price for a ~12% compounder. Base case 5-year ~$496 (+9%/year): Fairly valued — growth is partly offset by multiple compression; the credit cycle is the risk, the premium fee annuity the engine.
- Price
- $337.39
- Intrinsic value (5y, base)
- $496
- Total annual return (5y)
- 9.4%
- Status (nominal)
- Fairly valued
- Margin of safety
- +20%
The essentials
- The premium closed-loop payments network: it sees both sides of every transaction (issuer + acquirer + network, unlike Visa/Mastercard, which only operate the network) and is also a card lender (loan book ~$152bn). Billed business $1,670bn (+8%), 152.8M cards, ROE ~34%. The spend-centric model (high-spending, good-credit customers) produces structurally low losses (write-off 2.0% principal, delinquency 1.3% best-in-class).
- The quality engine is the fee annuity: net card fees grew +18% ($10.0bn, average fee per card $117 +14%) on the Platinum card refresh and premium-customer acquisition — high-margin recurring revenue growing faster than spend. Millennials/Gen-Z are the largest and fastest-growing cohort. Berkshire Hathaway holds 22% (a 35-year Buffett position, under a passive governance agreement).
- At ~$337 it trades at ~20× earnings (~20× the FY26 guidance of $17.30-17.90) — a full price. Earnings compound ~12%/year (spend +9-10% + buyback ~2.3%/year), but the multiple compresses from ~20× to ~17.5× over five years. Base 5-year ~$496/share → a total return of +9%/year: Fairly valued. An exceptional compounder at a fair price; the credit cycle (it is a lender) is the central risk.
Intrinsic value — two valuation methods
Total return at 5 years: 9.4%/year = 8.0% appreciation + 1.4% dividend. The target price ($496) is ex-dividend; the $27 in dividends collected over 5 years are added separately.
By both methods, the value today (DCF $600 · Multiples $422) exceeds the market price ($337).
Pillars of the analysis
The verdict — today vs 5 years
Today — fairly valued: at $337 trades ~20.0% below its value discounted to today (~$422); the discount is positive but does not reach the margin of safety we require (≥38%).
At 5 years — En valor: the target price ($496) plus dividends yield between the 4% floor and the 10% average return — a reasonable return, though without the margin of a great investment.
The bridge: the return at 5 years exceeds the risk-free rate (4.5%) — but the discount does not reach the required margin of safety (≥38%). To require a 15% annual return, it would need to be bought at ~$264.
Thesis
The business
American Express is a premium closed-loop payments network (sees both sides of the transaction) + a lender to affluent customers — an exceptionally high-quality franchise. The moat is the premium brand (Platinum/Centurion), the two-sided network effect and the spend-centric economics (high-spend customers → structurally low losses, 2.0% write-off). The engine is the card fee annuity (+18%) + spend (+8%) + the loan book (ROE ~34%). Berkshire holds 22% (Buffett, 35 years) — a quality stamp. Lender cyclicality + regulation are the limits.
The valuation
American Express is a financial institution (deposits + debt fund the loan book), so the equity is valued on P/E over net income to common — earnings are levered (already net of funding cost), the P/E delivers the equity value directly, and the funding is not subtracted (it is offset by the assets it finances). Net income to common on a TTM basis is ~$16/share; FY26 guidance is $17.30-17.90. At ~$337 the P/E is ~20× (~20× the guidance) — a full price for a franchise of this quality.
The base case projects earnings compounding ~12%/year (spend +9-10% + the fee annuity + a ~2.3%/year buyback, with stable margins) to ~$28/share in five years, at a 17.5× exit multiple. That multiple sits below the pure networks Visa/Mastercard (~21×, which carry no credit risk) and above the card lenders (Capital One ~10×) — the discount for the credit cycle + regulation. That gives ~$496/share → a total return of +9%/year, with multiple compression (from ~20× to ~17.5×) eating into part of the growth.
The margin of safety
It trades close to intrinsic value, far from the required margin of safety. At ~$337 American Express trades at ~20× earnings — a full price for a ~12% compounder. The verdict is Fairly valued: an exceptional franchise (ROE ~34%, premium brand, fee annuity, Berkshire holding 22%) at a fair price, not a bargain. Earnings growth (~12%) is partly offset by multiple compression from ~20× toward ~17.5× (the appropriate mature level for a business carrying credit risk) → the net return is moderate, carried by premium spend + the fee annuity, not by multiple expansion. The adverse scenario (a recession hits the lender: write-offs rise, spend slows, and the multiple compresses to 13×) takes the value below the price; the favorable one (the fee annuity + affluent customers + international compound faster + a re-rating) is a clear upside. The asymmetry is balanced — it is a quality purchase at a reasonable price, with the credit cycle as the risk to monitor.
What to watch
Three things. The credit cycle (the central disconfirming risk) — American Express is a lender (~$152bn book, ~79% U.S.); write-offs (currently 2.0-2.3%, best-in-class) and delinquency (1.3%) are the metrics to watch: a recession pushes both up and hits earnings. The fee annuity — whether net card fees (+18%) sustain growth through the product-refresh cycle is the most durable quality engine. And regulation — discount-rate caps (already pushed it off the network in the EU/Australia) + potential U.S. interest-rate caps + the Fed's capital restrictions are the structural threat. Berkshire (22%, a passive agreement, votes with the board) is an anchor of stability. If credit holds up and fees compound, American Express at this price is a quality purchase; if the cycle bites, the lender suffers.
Educational / informational. Does not constitute investment advice.