Three essays on the agency problem: CEO compensation, director pay, and proxy disclosure choices

This dissertation examines some of the implications of the separation of organizational ownership and control. The first essay extends the CEO compensation literature in two areas. First, it tests multiple determinants of CEO pay, including financial performance, CEO risk-bearing, CEO human capital, and board structure. Second, it examines the pay of hospital CEOs, which permits a test of the importance of an organization's status as being either investor-owned or nonprofit. The empirical findings suggest that each of the multiple determinants of CEO pay is important, and that there are few differences in the predictors of CEO pay between for-profit and not-for-profit hospitals. The second essay examines the factors associated with director compensation at large, publicly traded U.S. firms. Three approaches, based on different assumptions about the nature of the board, are tested. First, directors can be paid to act in shareholder interests, suggesting that pay is related to corporate performance and the need for board monitoring. A second approach suggests that directors are pawns of managers and that board pay is related to managerial cooptation. A third method is to pay directors solely for the time they spend attending to board matters. The empirical results support to differing extents, all three approaches. The third essay examines the reactions of corporate managers to the proxy statement disclosure rule requiring a five year performance graph. Managers could choose to include an additional graph presenting a more favorable impression of the company's performance. Following prior governance research, I hypothesize that an additional graph is more likely if the company underperformed its peers, paid its CEO a higher salary, and lacked a large shareholder. These hypotheses are tested on a sample of large firms. The results show that impression management is important to managers.