Capital One (NYSE: COF) reported Q2 2026 adjusted EPS of $5.81, smashing the analyst consensus of $4.69, as net income swung to $3.0 billion from a loss of $4.3 billion a year ago. The headline beat was powered almost entirely by a $1.1 billion drop in its provision for credit losses — roughly 96% of the entire increase in pretax income.
Capital One's numbers looked spectacular on the surface. Revenue up 4%, EPS crushing estimates by 24%, stock nudging higher after hours. A founder reads that and thinks: great business.
A CFO reads it differently. Here is the first question a CFO would ask: where did the profit actually come from?
The answer is uncomfortable. Provision for credit losses fell $1.1 billion quarter-over-quarter — and that single line item accounted for roughly 96% of the entire rise in pretax income. Pre-provision earnings, the number that strips out this accounting adjustment, were essentially flat.
Provision for credit losses is the amount a bank sets aside anticipating loans that may go bad. When credit conditions improve, they release some of that reserve back into income — it flows straight through the P&L and inflates profit. It is real money, but it is not operating momentum. It is the bank saying: 'we were too cautious before.'
A CFO would separate the story into two parts. One: is the underlying business growing? (Net interest margin ticked up to 8.01% — that is genuinely good.) Two: is this quarter's profit repeatable? A reserve release is a one-time tailwind; you cannot release the same reserves twice.
The lesson: always ask whether earnings are driven by operating performance or by below-the-line adjustments. A P&L can look healthy while the engine underneath is idling. That distinction is what ratio analysis is for.
📚 Learn the concept: Ratio Analysis
Source: https://finance.yahoo.com/markets/stocks/articles/capital-one-q2-2026-earnings-212653230.html
▶ Play the 90-second CFO game All daily posts