Affirm 'Skyrocketed' on a $1.6B Profit. $1.45B of It Was a Tax Entry.

Affirm Holdings (NASDAQ: AFRM) reported its fiscal fourth-quarter results on Thursday, and the market's first reaction was a share-price rocket. Revenue rose 33% to $1.166 billion, gross merchandise volume (GMV) climbed 36% to $14.1 billion, and GAAP net income came in at $1.6 billion — the company's "most profitable quarter ever." Both revenue and GMV beat Wall Street's estimates. Opening Friday, the stock gapped up about 11% to $86, ran to $90.44, and then faded all the way back to roughly $78 by the close — about where it started the week, on a day when major payments peers fell anywhere from 4% to 13%.

The reason the pop didn't stick is the same reason the "record" profit needs a second look.

Roughly 90% of the headline profit was an accounting entry.

Of that $1.6 billion in net income, about $1.45 billion was a one-time, non-cash income tax benefit from releasing a deferred tax valuation allowance. The real operating earnings power of the quarter was pre-tax income of $169 million, up 136% from a year ago. Affirm's own letter calls it the most profitable quarter ever excluding the tax allowance release.

Here is the mechanism, because the same entry is going to confuse next year's numbers too. Over its loss-making years, Affirm built up large deferred tax assets — mostly net operating loss carryforwards it might never get to use. GAAP forces a "valuation allowance" to offset those assets until they look realizable. Now that Affirm is consistently profitable, the accounting concludes the assets are real, so the offsetting allowance is released all at once, producing a one-time, non-cash benefit; no cash changes hands. The hangover is a baseline reset: going forward, the GAAP income statement carries a real tax rate in the mid-to-high 20s percent instead of the near-zero rate of past years. Next year's on-paper profit gets harder to grow at the exact moment the incentive to read it the wrong way peaks.

That is also why the "cheap" number on the quote screen is a mirage. The stock shows a trailing P/E of about 13, and its denominator is that $1.6 billion — mostly tax accounting. Divide the roughly $26 billion market cap by the $169 million of quarterly pre-tax earnings power instead, and the stock is not trading at 13x anything real.

Underneath the entry, the operating engine genuinely improved.

What actually got better, on numbers that do not depend on an accounting release: Revenue less transaction costs (RLTC) — the metric that subtracts the cost of funding loans, the honest way to measure a BNPL's take — grew 39% to $589 million, or 4.2% of GMV. GAAP operating margin came in at 12.6%, up 6 percentage points from 6.6% a year ago, with the adjusted measure around 30%. Credit improved rather than deteriorated: the 30-day delinquency rate on monthly installment loans fell to 2.5% from a 2.7%–2.8% band in prior quarters. Active consumers grew 21% to 27.8 million, Affirm Card active users jumped 125% to 5.2 million, and full-year free cash flow was $993 million versus $602 million the year before. Whatever the tax entry did to the headline, this is a business compounding GMV at 36% while revenue as a percentage of GMV held near 8.3% and losses on loans kept coming down.

Guidance says growth slows — mechanically, not structurally.

Management guided fiscal 2027 to GMV above $64 billion (roughly 28% growth), revenue at about 8.5% of GMV, and GAAP operating margin above 14.5% (with the adjusted measure above 30.5%). The near-term step-down is what to notice: first-quarter GMV growth is guided to about 29%, down from 36%, for two named reasons — a large promotional event that shifted out of the quarter and the lapping of an enterprise merchant that moved its pay-later volume to its own wallet. Those are calendar and comparability frictions, not evidence that demand is breaking, and the outlook deliberately assumes short-term rates rise, so management built the headwind into the numbers rather than pretending it away. On the same day, Affirm took Shop Pay Installments live in Australia, the next step in its exclusive global partnership with Shopify that already spans the U.S., Canada, and the U.K. — what headlines called a big Shopify deal. But Affirm is explicit that international is not a material contributor to fiscal 2027. The international option is a fiscal 2028+ story, so it should not be what funds today's price.

The price is where the math stops being kind.

The stock had already climbed roughly 85% off a 52-week low near $42 and was up 53% over the prior 120 days before this week's pop. Friday's spike stopped about $3 short of the $93.20 high, and the fade back to $78 closed at the exact level where the week began — a session that behaved like a market happy to cheer the headline but unwilling to pay up at new highs. On the guide, a $26 billion market capitalization is about 5x fiscal 2027 revenue, and an enterprise value near $34 billion — which includes the debt that funds the loan book — is roughly 13x fiscal 2027 RLTC. For a business compounding GMV at 28%–36% with margins climbing and credit improving, that is a full growth-stock price, not a distressed one. The entire re-rating now rests on one variable: whether GAAP operating margin really travels from 12.6% toward the guided 14.5%-plus by the end of fiscal 2027 while RLTC stays near 4.2% of GMV and delinquency holds around 2.5%.

That is the number to track. The market celebrated a $1.6 billion profit that was mostly an accounting entry, on top of a run that had already priced in the underlying improvement. The operating momentum is real — but at roughly $78, this is not a value bet on a cheap P/E. It is a growth bet on margin delivery. If the margin path lands, the multiple gets cheaper all by itself. If it stalls, the "cheap" P/E that this week's headline printing press is generating turns out to be exactly what the accounting entry was hiding.