Descriptive audit evaluates CEO pay against company performance in S&P 500, highlighting compensation disparities.
The S&P 500 edition of the CEO Pay-vs-Delivery Scorecard. It extends the published S&P-100 v0.3 to the full index using the same SEC EDGAR Pay-versus-Performance pipeline (Item 402(v) of Reg S-K), pulled straight from each DEF 14A: 2,741 company-year rows across 501 S&P-500 tickers, 494 companies in the assembled view, 492 TSR-comparable. It sets realized pay (Compensation Actually Paid) beside granted pay (the Summary Compensation Table figure the press quotes) and peer-relative shareholder return, issues no overpaid verdict, and lets the reader conclude. The move to 500 companies turns a curated-sample result into a population statistic: 229 of 492 CEOs (46.5 percent) beat their own board-chosen peer index over five years, barely different from the 46 of 96 (47.9 percent) at S&P-100 scale, so the coin-flip is not an artefact of a small high-pay sample. The median latest-year granted figure roughly halves, from 28.4 million USD (large-cap) to 17.9 million USD (market), and the broader universe surfaces a tier of nine-figure single-year grants the S&P 100 never held: Welltower's 821.1 million USD (verified to the filing, about 99 percent equity, index-conditioned), CrowdStrike's 247.6 million USD, and 2021-vintage founder grants at Robinhood (796 million USD) and The Trade Desk (835 million USD). A delivery floor becomes measurable at scale: 86 of 494 companies (17.4 percent) had a five-year total shareholder return below 100 (a 100 USD stake shrank) and 34 posted a negative latest-year Compensation Actually Paid. The hand-read board-targets layer (the 29 curated S&P-100 names) does not scale to 500 by design and is unchanged; v2's contribution is the automated spine plus the population statistics. Coverage is honest about its edges: a 7-company manual tail is documented by cause (three image-based proxies with no machine-readable table, one DEF 14C information-statement filer, one company with no discoverable DEF 14A, one with no PEO compensation tags, and one 2025 merger entity with no annual proxy yet), and the one filer-error erratum from v0 (a mis-signed Capital One CAP fact) carries forward as a keyed correction. Caveats stand: Compensation Actually Paid is an accounting fair-value re-mark, not cash; total shareholder return is not caused by the CEO; the peer group is the company's own choice. Descriptive analysis, not investment advice. Disclosure: the author holds no direct position in any company named; a workplace defined-contribution pension may hold some of these names indirectly through pooled funds the author does not direct.
No takes yet. Share an insight, caveat, or question.
N Milton (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: