Two streams of agency theory research have focused on different aspects of the contracting relationship between shareholders and CEOs. The first stream of agency theory research examines the role of multiple performance measures in a CEOs compensation contract in maximizing firm value in a single-period setting. The second stream of agency theory research focuses on the horizon problem and examines how a CEOs career concerns, his desire to remain employed or cease employment in the near future, affect his utility and his compensation contract. Both streams of research provide useful evidence on the shareholder-CEO contracting relationship but are incomplete because they do not simultaneously consider multiple performance measure, multi-period compensation contracting.The second chapter of this dissertation examines multiple performance measures in the presence of the horizon problem to determine whether a CEOs compensation structure, the ratio of earnings-based compensation to total compensation, is affected by the CEOs career concerns. In examining whether the CEOs compensation contract is affected by his career concerns, I recognize that a CEO who is nearing retirement can leave the firm for a variety of reasons. As such, I use a competing risks regression, a hazard model that allows for the comparison between observation groups by how the CEO left the firm, to incorporate the reason a CEO left into the analysis of the relation between a CEOs compensation structure and his tenure. The three groups of interest in the competing-risks regression are whether the CEO retired, resigned following poor performance, or left following a merger or acquisition. I find no evidence that a CEOs tenure for each of these groups is associated with his compensation structure. I also do not find evidence of significant differences between the compensation structures of CEOs who retired, resigned following poor performance, or left following a merger or acquisition. The lack of associations and differences between compensation structures does not offer insight into whether compensation committees adjust compensation contracts to offset a CEOs career concerns as he gets ready to leave the firm but do suggest that, if the compensation structure is being adjusted, it is done in a manner that assumes all CEOs will leave for the same reason. I do find evidence that a CEOs tenure is associated with firm performance for both the retirement and poor-performance groups. This suggests that the board of directors actively monitors the CEOs performance.I use the estimates of CEO tenure generated by the competing risks regression in a maximum-likelihood regression of the determinants of a CEOs compensation structure. I find that CEOs with more earnings-based compensation have longer tenures. This is consistent with the notion that CEOs who are approaching retirement are likely to become more conservative and unwind their equity positions. I also find that the relative noise ratio, the ratio of the noise in earnings to the noise in stock price, is positively associated with a CEOs compensation structure. This finding is contrary to prior theoretical and empirical research that examines the determinants of a CEOs compensation structure in a single-period setting, which predicts and finds evidence of a negative relation between a CEOs compensation structure and the relative noise ratio. In my research, I examine a multi-period setting. Certain features of a CEOs compensation structure, such as stock options, have been shown to encourage CEOs to increase volatility. Thus, in a multi-period setting, the relative noise ratio would have a positive relationwith a CEOs compensation structure.While a CEOs compensation structure is not related to his tenure, changes in his compensation structure are related to his tenure for both the poor performance and the merger and acquisition groups. For the poor performance group, stock-based compensation increased with tenure. In addition, the changes in compensation structure for both the poor performance and retirement groups are similar. Finally, changes in performance are negatively associated with tenure for CEOs in the poor performance group. The combination of these results suggest that compensation committees increase stock-based compensation as CEOs approach retirement but not in a manner that suggests the compensation committee distinguishes between how a CEO will leave the firm. Instead, it appears that compensation committees allow firm performance to determine how a CEO, who is of retirement age, leaves the firm.For the mergers and acquisition groups, increases in stock-based compensation are associated with shorter tenures. This is consistent with the notion that significant increases in stock-based compensation can motivate CEOs to find ways to unwind their equity positions as their career concerns change. Becoming a target of a merger or acquisition may afford the CEO such an opportunity. The changes in compensation structure for the mergers and acquisition group are significantly different than the changes in compensation structure of both the retirement and poor performance groups. While the changes in compensation structure for the mergers and acquisition group are significantly different from the retirement and poor performance groups, univariate statistics indicate that the changes in compensation structure for the mergers and acquisition group result in compensation structures that are similar to the retirement groups for the period immediately preceding when the CEO left the firm. Thus, the changes in compensation structure for the mergers and acquisition group also appears to support the notion that, while compensation committees appear to adjust a CEOs compensation structure as he approaches retirement, they do not do so in a manner that is indicative that they anticipate how a CEO will leave the firm.A different method of examining the shareholder-CEO contracting relationship is to examine the sensitivity of a CEOs compensation to changes in shareholders wealth. In the third chapter of this dissertation, I examine the shareholder-CEO contracting relationship in regard to the incentive ratio, which is the ratio of a measure of the sensitivity of changes in the CEOs earnings-based compensation contract to changes in shareholder wealth to a measure of the sensitivity of the CEOs stock-based compensation to changes in shareholder wealth. While the proxy for a CEOs incentive ratio is positively correlated with my proxy of the CEOs compensation structure, the correlation is not statistically significant, suggesting that these two measures capture different features of the shareholder-CEO contracting relationship.While the empirical evidence in Chapter 2 does not indicate any association between a CEOs compensation structure and his tenure, the empirical evidence in Chapter 3 suggests a significant association between the CEOs incentive ratio and his tenure. For both the retirement and merger and acquisition groups, larger stock-based incentives are associated with shorter tenures. This is consistent with the notion that, as a CEO approaches retirement, he may become more conservative and wish to unwind his risky equity positions by leaving the firm. Interestingly, larger earnings-based incentives are associated with shorter tenures for the poor performance group. This is consistent with the notion that a compensation committee will force a resignation sooner for CEOs who have significant earnings-based incentives but are not realizing the expected performance targets. When the estimates of tenure from the competing risk regression are included in a maximum-likelihood regression that examines the determinants of the CEOs incentive ratio, I find a negative association between the CEOs tenure and his incentive ratio. This is consistent with the notion that stock-based incentives encourage the CEO to leave so that he can unwind his risky equity positions.For the analysis of whether there is an association between changes in a CEOs incentive ratio and his tenure, I find that increases in stock-based incentives are associated with longer tenures for CEOs who leave following a merger or acquisition. This is inconsistent with the notion that increases in a CEOs stock-based incentives will motivate the CEO to shorten his tenure by engaging in a merger or acquisition. However, these results are consistent with the notion that increases in a CEOs stock-based incentives will motivate the CEO to engage in a merger or acquisition when it is favorable for the CEO to do so.Overall, the results from both Chapter 2 and Chapter 3 suggest that compensation committees adjust compensation as a CEO moves through his tenure with a firm but not in a manner that suggests that the compensation committee anticipates how a CEO will leave the firm. However, the evidence also suggests that these adjustments, when coupled with firm performance, do affect how a CEO leaves the firm. These results may be of interest to shareholders, regulators, and researchers and aid our understanding of how a CEOs personal characteristics may affect the contracting relationship as he moves through his tenure with a firm.
