Stock Buybacks: How Issuance Cancels Most of the Headline

Headline buyback announcements routinely mask the parallel issuance that absorbs them. Net shareholder yield, buybacks minus issuance, is the only figure that measures what shareholders actually receive.

S&P 500 companies repurchased $1.020 trillion of their own shares over the twelve months to September 2025, a record according to S&P Dow Jones Indices (release of 18 December 2025), including $777 billion in the first three quarters of the year. Yet in Q3 2025, 160 index members had more shares used for earnings per share than a year earlier, against 312 with fewer. The gap between those two facts is the entire phenomenon this article unpacks. Parallel issuance (stock-based compensation, capital raises, bond conversions) neutralises a substantial portion of every buyback programme before it reaches the reported share count. The dynamics also shape how share supply interacts with price formation in the ETF ecosystem, the venue where most of this flow ultimately clears.

The blind spot is structural. The financial press reports the headline buyback figure. Quarterly investor decks lead with the dollar amount of programmes announced. The line item that captures the net effect, change in diluted share count, sits deeper in the 10-K and rarely makes the headlines. The total shareholder yield framework only delivers what it claims when both sides of the equation are read together.

Where the dilution comes from

The dominant source in US large caps is stock-based compensation, heaviest in technology and in fast-growing software publishers. The number rarely appears in earnings press releases but sits explicitly in the consolidated cash flow statement, line “stock-based compensation expense”.

The mechanism is straightforward. A company announces a $10 billion buyback programme. Over the same fiscal year, it issues $6 billion of stock options and restricted stock units to staff. Net buybacks come in at $4 billion, 60% below the headline. A shareholder reading only the announcement overstates the effective return by a factor of 2.5. On the same question: how buybacks return cash to shareholders.

Other issuance channels operate more sporadically: equity raises to fund acquisitions, conversions of convertible bonds, exercise of warrants. These appear in irregular bursts and can dominate the dilution line during M&A waves or post-IPO unlock periods.

Where the dilution concentrates

Net dilution varies sharply by sector. A scan of 2024-2025 annual reports surfaces three distinct profiles.

Technology records the highest issuance rates. The equity-compensation culture inherited from Silicon Valley generates structural dilution that even aggressive buyback programmes struggle to absorb. Several cloud infrastructure and cybersecurity names display a negative net balance over multi-year windows: they print more shares than they retire.

Mature sectors (utilities, consumer staples, traditional energy) use equity compensation sparingly. Their buybacks translate almost fully into a reduction of the float. In these segments, the gross figure is a reasonable proxy for the actual return.

Financials sit in between. Large US banks combine large buyback envelopes with regular issuance linked to deferred bonus plans. Their effective return shows in each group's change in share count, always below the sum of the programmes announced.

The single metric that captures the reality

The metric that compresses the entire problem into one number is net buyback yield: buybacks minus issuance, divided by average market capitalisation over the period. A 4% gross figure can shrink to 1.5% net. In tech-heavy names, it can turn negative.

Combined with the dividend yield, this gives the net shareholder yield, the only figure that compares distribution policies on a like-for-like basis across sectors with different equity-compensation cultures. The distribution-cycle framework breaks the components down further.

Sell-side consensus tends to lag the adjustment. Shareholder-yield projections relayed by financial media generally start from announced buyback envelopes without integrating expected issuance. The bias is structural and largest in sectors with high equity components, which happen to be the same sectors that dominate index weight in 2026. A related read: our buybacks-versus-dividends comparison.

What changes if the gap widens

Net dilution reshapes several balances. For companies, equity compensation preserves cash while attracting talent, but it transfers value from existing shareholders to employees. For the index aggregate, the net balance still favours earnings per share: according to S&P Dow Jones Indices, 17.1% of S&P 500 companies had cut their share count by at least 4% year on year in Q3 2025, lifting their EPS by as much. The average hides the other camp: in that quarter alone, 156 companies increased their count.

A cycle reversal would amplify the asymmetry. In contraction phases, companies typically suspend or scale back buyback programmes, but issuance tied to compensation plans already granted continues to vest. The net balance then turns negative precisely when the share price is falling, compounding the drawdown.

Reading buybacks gross, without netting issuance, leads to overestimating the technical support buybacks provide to the index. A net reading delivers a less flattering but more accurate view of what listed companies actually return to their owners, and what they retain for other claims on cash.

What this dynamic reveals: the announced buyback figure does not measure the generosity of a distribution policy. Only the net balance, buybacks minus issuance, measures what reaches existing shareholders. In sectors with heavy equity-based compensation, that gap can exceed half of the headline amount, year after year, without ever appearing in the press cycle. In depth: the criteria behind a dividend screen.

Last updated — 7 October 2026