Long-Term Pay for Performance: Cumulative Evidence over CEO Tenure

This study revisits the CEO pay-for-performance relation using two methodological innovations designed to address well-known measurement challenges. First, we aggregate pay and performance over the CEO’s entire tenure, allowing us to account for ex post settling up in pay. Second, we use realized compensation rather than ex ante estimates to better capture the actual economic rewards that CEOs receive. We find strong evidence of pay for performance in the full sample with a magnitude 2–10 times greater than that reported in previous studies. However, the pay-for-performance relation is asymmetric. CEOs who outperform their peer firms exhibit strong pay-for-performance sensitivity; those who underperform do not. This asymmetry arises because boards offset weak performance with incremental equity grants, undoing the mechanical correlation that results from declining stock prices. Finally, we show that underperforming CEOs are significantly more likely to be dismissed, suggesting that turnover, rather than lower pay, could be a disciplinary mechanism used by boards. Collectively, our findings demonstrate that tenure-aggregated realized pay better reveals how boards structure incentives and accountability over the full horizon of a CEO’s leadership. This paper was accepted by Ranjani Krishnan, accounting. Funding: We gratefully acknowledge financial support from Columbia Business School, Columbia University. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.04322 .