"The responsiveness of CEO compensation to firm value -- that is, the percentage change in compensation from the prior year divided by the percentage change in firm value -- more than tripled from 1980 to 1994, rising from 1.2 to 3.9."
The exploding use of stock options to compensate executives has increased CEO pay significantly. But the justification that CEOs and corporate boards most often give for generous stock options -- that they effectively link pay to performance -- is often scoffed at. And in years of academic research, study after study has shown little relationship between CEO compensation and corporate performance.
In Are CEOs Really Paid Like Bureaucrats? (NBER Working Paper No. 6213), however, NBER Faculty Research Fellows Brian Hall and Jeffrey Liebman point out that times have changed. Using data from large U.S. corporations from 1980 to 1994, they show that CEO pay has become much more sensitive to corporate performance than it once was. And they credit stock options for this change.
Hall and Liebman show that, when one accounts for revaluations of stock and stock options, CEO pay often changes by millions of dollars for only modest changes in firm performance. Moreover, they estimate that the responsiveness of CEO compensation to firm value -- that is, the percentage change in compensation from the prior year divided by the percentage change in firm value -- more than tripled from 1980 to 1994, rising from 1.2 to 3.9. The elasticity of salary-plus-bonus (with stock and options excluded) to firm value was much lower, although it had increased from 0.13 in the early 1980s to 0.24 in the late 1980s and early 1990s. Hall and Liebman argue that the "pay-to-performance sensitivities from changes in salary and bonus are swamped by sensitivity generated by changes in the value of stock and stock options."
From 1982 to 1994, CEO pay rose sharply -- a median real increase of 120 percent - outpacing that of average workers (who had a real increase of about 7 percent) and even of other groups of highly paid Americans. Only professional athletes did better. But the large increase in median CEO pay masks the risk of large, year-to-year variations in pay. For example, in 1994, a mediocre year for the stock market, almost a quarter of the CEOs in the survey actually lost money as the value of their company stock holdings declined.
Hall and Liebman do not claim that current pay-to-performance sensitivity is sufficiently high. Also, they point out that stock options often reward (or punish) a CEO for market-wide or industry-wide gains (or losses), rather than for the performance of the CEO's corporation relative to other corporations. But, they argue, "Our findings do contradict the claim that CEO contracts are wildly inefficient because there is no correlation between performance and pay..."