Lubin Investment · Blog

American Express (AXP): Q2 2026 results, my verdict

2026-07-25 ·

AXP: see the full analysis on Lubin Investment

American Express reported earnings per share of $4.53 on July 24, 2026, up 11% and above consensus, but revenue slightly below expectations despite 10% growth. The stock fell more than 5% despite a raised full year guidance. My quality filter validates 6 out of 10 criteria.

A quarter that beats on profit, misses on sales, and splits the market

American Express reported second quarter results before market open on July 24, 2026: earnings per share of $4.53, above the $4.45 consensus and up 11% year over year (from $4.08 a year earlier), on revenue net of interest expense of $19.64 billion, up 10% year over year but slightly below the $19.9 billion consensus. Total card member spending, known as billed business (the standard measure of activity for card issuers), reached $455.8 billion, up 9%, and network volumes reached $516.8 billion, also up 9%.

The segment breakdown shows where the real momentum sits: U.S. Consumer Services revenue grew 11%, Commercial Services (business cards) 7%, and International Card Services jumped 12%, the company's fastest growing segment. Another strong signal: net card fees, the annual fee paid to hold a premium card such as the Platinum, climbed 15%, driven by strong acquisition and retention among Millennial and Gen Z customers, a demographic historically not associated with the Amex brand, better known for its older, affluent business traveler.

Despite this quarter and a raised full year 2026 revenue growth guidance of 10%, the stock fell more than 5% in premarket trading. The market chose to focus on the slight revenue miss rather than the profit beat, and especially on management's decision to reinvest part of the profitability upside into growth (marketing, loyalty benefits) instead of letting it flow entirely to earnings per share. This is a genuine strategic choice, and a genuine communication risk: reinvesting today to fuel tomorrow's growth is often the right long term call, but the market almost always punishes it in the short term, because the payoff is invisible in the quarter that just closed.

Why my quality filter validates only 6 out of 10 criteria

My standard model recognizes real strengths at American Express: a 14.1% net margin (the company is profitable), a 31.9% free cash flow margin (nearly a third of every revenue dollar turns into cash left over after expenses AND investments), a 122% cash conversion rate (accounting profit actually turns into MORE cash than reported, a sign of accounting quality, here partly because credit loss provisions, an accounting charge that is not an immediate cash outflow, weigh down reported profit without reducing cash by the same amount), and a share count declining 2.6% a year thanks to buybacks, a sign of shareholder discipline.

But two points weigh heavily on the score. First, growth: revenue grows 7.5% a year on average over 5 years, a decent rate but below the 10% a year bar my model requires, and free cash flow per share grows only 3.7% a year over the same period, a weak trajectory. The real problem is not decline but marked VOLATILITY: annual free cash flow fell from $13.1 billion in 2021 to a peak of $19.2 billion in 2022, then dropped to $12.1 billion in 2024 before recovering to $16 billion in 2025. A sawtooth trajectory, not a steady machine, largely driven by swings in credit loss provisions tied to the economic cycle and by marketing investment that varies year to year. Second, my model flags margin compression (costs growing faster than revenue) over the period: this is precisely the cost of the repositioning described above, acquiring younger and international customers is expensive in marketing and loyalty benefits before it pays off in recurring annual fees.

The hidden mechanic: why American Express is not a simple payment network like Visa or Mastercard

My model also flags debt that would take 12 years of free cash flow to repay, a figure that would look alarming for an industrial company. But American Express is not Visa or Mastercard, and the distinction matters to understand this number. Visa and Mastercard operate an open loop network: they only route the transaction between the bank that issued the card and the merchant, never lending money directly to the cardholder, which keeps their balance sheet very light. American Express operates a closed loop network: it is simultaneously the network AND the issuer that lends directly to nearly all of its cardholders, meaning it carries, on its own balance sheet, the money its customers owe it over time.

It is this direct lending activity that explains why total debt rose from $40.9 billion in 2021 to $57.8 billion in 2025, up 41% over four years: more debt here does not signal growing default risk, it reflects a card member loan book that grows alongside billed volumes rising 9% a year. It is the same mechanic as a bank funding its loans with customer deposits, or an insurer investing float (premiums collected before claims are paid): a liability that directly funds a revenue generating activity, not a sign of fragility in itself. Above all, it means a generic threshold of 'debt repayable within 3 years' (calibrated for a typical industrial company) will almost always fail for a closed loop card network like American Express, exactly as it systematically fails for a bank or an insurer.

The price: a 16.3 times P/FCF that looks reasonable, a model that says otherwise

The P/FCF (the share price divided by free cash flow generated per share over the trailing twelve months) comes out at 16.3 times, a figure that, taken alone, looks almost cheap against the 25 times threshold my model uses as a general benchmark and against the far higher multiples recently seen on tech names. The issue is not this headline multiple: it is that my reasonable buy price model, which projects the ACTUAL free cash flow per share trajectory of the past five years (that weak, choppy 3.7% a year) to compute what the stock should be worth today to deliver a fair future return, arrives at a target of only $111.96, against a $326.17 share price. That is a 65.7% premium over what the recent cash generation history would strictly justify.

This is a genuine tension to work through, not an obvious call either way. The market is clearly not pricing American Express on its trailing five year free cash flow history: it is betting on the acceleration now underway (premium card fees +15%, international +12%, a younger customer base engaging more) which, if it holds and eventually pushes acquisition costs down relative to revenue, would justify a far stronger cash trajectory than the one the company posted between 2021 and 2024. The bet is reasonable on paper, but for now it is confirmed by only ONE quarter of encouraging signals, not yet by the multi-year trend my model relies on. Paying today for an acceleration not yet fully proven in the numbers is exactly the definition of a bet on the future rather than a read on the past.

How I read it

American Express is a genuinely quality business, with high margins, a solid return on capital, and consistent buyback discipline, engaged in an expensive but credible repositioning that already shows tangible traction: the double digit growth in premium card fees and international activity is not a slogan, it is numbers reported this very quarter. My score of 6 out of 10 reflects a five year window that mostly captures the phase where this bet was costly (compressed margins, volatile cash), not necessarily the trajectory of the coming quarters if the bet keeps paying off.

What would change my mind: if the coming quarters confirm that premium fee and international growth keep outpacing acquisition cost growth (the margin compression reversing), my model's cash trajectory would mechanically improve, and the current price would become easier to justify. If instead growth disappoints again as it did this quarter on sales, and acquisition costs do not come back down, the current 65.7% premium becomes much harder to defend. I am watching premium card fee growth and the marketing to revenue ratio in particular, the two clearest tells of this bet. You can find the full breakdown on the American Express analysis page, understand why rising debt is not automatically a problem in my article on debt in my method, dig into the return on capital calculation in my article on Cash ROCE, and my full methodology.

FAQ

Why did American Express stock fall despite beating profit estimates?

Revenue ($19.64 billion) came in slightly below consensus, and the market focused on management reinvesting part of the profitability upside into growth (marketing, loyalty benefits) rather than letting it flow entirely to earnings per share, a strategic choice often punished in the short term.

Why is American Express' quality score only 6 out of 10?

My filter judges a 5 year trajectory: sales growth (7.5%/year) and especially free cash flow per share growth (3.7%/year, on a very volatile cash history) stay below my thresholds, and costs grew faster than revenue over the period, the cost of repositioning toward younger and international customers.

Why does American Express carry so much debt relative to its cash flow?

Unlike Visa or Mastercard, which only route the transaction (open loop), American Express lends directly to its cardholders (closed loop) and therefore carries this loan book on its own balance sheet. The debt funds this lending activity, which grows with volumes, not a fragility signal comparable to an industrial company.

Is American Express stock expensive after these results?

Its 16.3 times P/FCF looks reasonable in absolute terms, but my buy price model, based on the actual and volatile 5 year cash trajectory, targets only $111.96 against a $326.17 share price, a 65.7% premium that bets on the recent acceleration in premium fees and international activity continuing.

Should I buy American Express stock after these results?

The company is a genuinely quality business with tangible signs of acceleration (premium fees +15%, international +12%), but this acceleration is confirmed by only one quarter against a model relying on a weaker 5 year history, hence a current premium that is hard to settle with certainty. This is not personalized investment advice: do your own research.

AXP: see the full analysis on Lubin Investment

About the author

Written by Lubin Danilo, founder of Lubin Investment. A self-taught individual investor, I find fundamental analysis fascinating, and it has delivered excellent results. For three years now, my performance has beaten the S&P 500. But analyzing every stock took too much time: sites with incomplete data, calculation methods and criteria never aligned with mine. And spotting the best stocks was just as time-consuming, even with my own well-defined checklist. So I put my software development background to work to build this software, base my investment strategy on its results, and share it with people who share the same passion as me. It judges a company's quality and its price separately, using criteria drawn from the financial literature (Warren Buffett, Michael Mauboussin, Aswath Damodaran).