Buybacks against stock-based compensation: what shareholders actually kept

Across the 375 largest US-listed companies, stock-based compensation went from 0.71% of revenue in 2015 to 1.33% in 2025 — 1.9 times the share of revenue it was a decade ago. Over the same decade buybacks have not grown at anything like that rate, which is why the ratio in the fourth column has been falling: each dollar of repurchase now has more issuance to cancel before it starts returning anything.

That ratio is the honest way to read a buyback announcement. A company repurchasing three dollars of stock for every dollar it grants is genuinely shrinking; one repurchasing a dollar for a dollar has run a payroll through the equity account and reported it as a return of capital. The median company's share count is in the table beside it, which is the outcome those two flows actually produce.

The sector split is the part most people guess wrong. The stock-heavy sectors are not the ones diluting: they pay the most of their compensation in stock and also run the largest buybacks, so their share counts still fall. In FY2025 the median company issued more than it retired in 3 sectors — Real Estate, Utilities, Energy — which is where dilution is quietly winning.

Two caveats that cut against the whole page. Buyback figures here are gross repurchases from the cash-flow statement, not net of issuance — which is exactly why they are shown next to the compensation line rather than alone. And the timing is worth looking at before treating repurchases as a value strategy: across this universe they peaked in 2018 at 6.32% of revenue and were lowest in 2020 at 4.39%. Compare those two years against what the market did in each of them — the page on annual returns is one click away — and judge for yourself whether the aggregate was buying low.