For most of the past decade, stock buyback investing* had a quiet superpower: it made earnings per share* look better than they really were. That superpower is fading — and investors who don’t notice are about to be caught off guard.
*Stock buyback: when a company uses its own cash to repurchase shares from the open market, reducing the number of shares outstanding.
*EPS (earnings per share): a company’s net profit divided by shares outstanding. Fewer shares = higher EPS, even with flat profits.
Here’s the mechanic. When a company buys back its own shares, the share count shrinks. The same profit now spreads across fewer shares, so EPS rises automatically. Over the past decade, this inflated earnings growth across the S&P 500 without requiring actual revenue gains. It was a mathematical trick disguised as performance.
That era is closing. JPMorgan estimates roughly $1.5 trillion in new shares will be issued by U.S. companies over the next two years — primarily to fund AI infrastructure. The five largest tech companies have shifted their capital spending from about 39% of operating cash flow (2010–2024 average) to over 90% today. To finance that, they’re issuing debt first, then equity. Nvidia tapped bond markets for $25 billion. Meta, $55 billion.
What follows is the reverse of the buyback era. Share counts rise. The same profits now spread across more shares, compressing EPS. A company can beat earnings estimates and still see its stock fall because dilution* is quietly working against the headline number.
*Dilution: when a company issues new shares, each existing share represents a smaller slice of the same pie.
The new filter for investors shifts from “did earnings beat?” to something harder: is the company’s ARR* actually growing? ARR — annual recurring revenue, the predictable income that renews without requiring new sales — is harder to fake and more durable than one-time earnings beats.
My take: The buyback era rewarded any company that could generate cash. The re-equitization era rewards companies that can prove their revenue will still be there next year. That’s a different filter — and it changes which stocks deserve a premium.
Disclaimer: This is not investment advice.