Three Essays on CEO Compensation

Abstract

This thesis presents three essays on CEO compensation. The first essay studies the disagreement between shareholders' Say-on-Pay voting results and ISS (Institutional Shareholder Service) recommendations on CEO compensation packages. Shareholders are expected to follow ISS "Against" recommendations if CEOs' compensation packages are inappropriate for the firm. However, around 8% of shareholders overlook ISS "Against" recommendations despite these recommendations being deemed informative (Albuquerque et al., 2020). Based on my findings, the cause of the disagreement is institutional shareholders' investment horizon. When CEO compensation packages include a high proportion of option grants and salary payments, firms with higher short-term institutional shareholder ownership are more likely to disagree with ISS. These findings reveal that the disagreements are caused by short-term institutional shareholders strategically making decisions that align with their interests. I observe no effect of long-term institutional shareholder ownership on the disagreement. These results remain constant after addressing possible endogeneity concerns through firm fixed effects, two-stage regressions with instrumental variables, falsification tests and entropy balancing. These findings provide the first explanation for the cause of the disagreement and additional evidence of short-term institutional shareholder voting behaviour. The second essay explores how the CEO-employee pay ratio (a proxy of within-firm pay disparity) affects firms' labour investment efficiency. The results in this essay reveal a negative association between the pay ratio and labour investment efficiency, indicating that firms with a larger CEO-employee pay ratio tend to have lower labour investment efficiency. Further tests demonstrate that this association is more profound when firms over-invest in labour. This essay adds to the literature by offering additional evidence to support the Rent Extraction Theory, which posits that CEOs tend to recruit excess employees to "build their empire" to extract rent, leading to over-investment in labour (Stein, 2003, Williamson, 1963). Again, I employ firm fixed effects, falsification tests and entropy balancing to address potential endogeneity concerns. The results remain robust to these tests. This essay identifies the CEO-employee pay ratio as an essential determinant for labour investment efficiency, which may help firms’ stakeholders make informed decisions. The third essay examines how the CEO-employee pay ratio influences firms’ stock price crash risk. The stock price crash risk is a major concern for shareholder welfare. In this essay, I document a significant and positive association between the CEO-employee pay ratio and the stock price crash risk, revealing that firms with a more significant pay ratio have a higher stock price crash risk. This result remains robust after addressing plausible endogeneity concerns using tests including firm fixed effects, two-stage regressions with instrument variables, change of specifications and entropy balancing. To further boost robustness, I include additional control variables and an alternative measure of the pay ratio in my baseline estimations, and again my results remain constant. In cross-sectional analysis, I find that firms granting more stocks to CEOs have a lower chance of stock crashes when the pay ratio is larger. Based on these findings, shareholders can make better investment decisions, especially concerning the risk of a stock price crash.