Corporate Governance and Executive Pay: Evidence from Takeover Legislation

We examine the e ects of anti-takeover legislation on CEO pay. Since these laws altered an important component of governance, the threat of takeover, they provide a natural testing ground for theories of executive compensation. Under skimming models, where entrenched CEOs pay themselves, we would expect mean pay to rise as less governed CEOs manage to skim more. Under contracting models, where a principal optimally sets pay, we would expect a fall (or no e ect) in mean pay since CEOs no longer need to be compensated for the risk of takeover. We might expect, however, a rise in use of pay for performance to o set the reduced incentives. Consistent with skimming, we nd that mean pay rose in rms a ected by the laws (relative to a control group). Moreover, the rise in pay was largest in rms that did not have a large shareholder present prior to the law. Turning to pay for performance, we nd evidence of a rise in pay for performance on accounting measures. This rise, however, seems to have been concentrated in rms with large shareholders. Together, these results suggest that rms without large shareholders match more closely the skimming model. The optimal contracting model, on the other hand, may have more relevance in the presence of a large shareholder who can serve as the principal these models posit. We are extremely grateful to David Yermack and Andrei Shleifer for making the CEO data available to us, and to Amber Batata, George Baker, Drew Fudenberg, Je Liebman, and especially Larry Katz, Caroline Minter-Hoxby and Andrei Shleifer for very helpful discussions. This paper was started while the authors were graduate students in the Economics Department at Harvard University. e-mail: mbertran@princeton.edu and mullain@mit.edu.