Date of Award
Doctor of Philosophy (PhD)
The first essay is entitled: "CEO Overconfidence, Corporate Governance Practices and Firm Innovation". In this study, I examine if overconfident CEOs overinvest or underinvest in innovative projects. I also investigate if overconfident CEOs pursue innovative projects to benefit personal interests or the interest of shareholders. By focusing on the effect of corporate governance in monitoring the behavior of overconfident CEOs, my results show that there is a negative relation between CEO overconfidence and firm innovation among firms with poor governance. In these cases, the finding is consistent with the implication that overconfident CEO are entrenched and invest inadequately in innovative projects when the firm has poor corporate governance. My results also show that for well-governanced firms, the overconfident CEOs do not overinvest because good risky projects are accepted and results in an increase in firm value. The other important result of my study is that well-governanced firms could influence the behavior of overconfident CEOs so that firm performance (measured by Tobin's q, ROA, and sales growth) is enhanced. On the other hand, it is also found that overconfident CEOs of poorly-governanced firms might pursue innovative projects so aggressively that projects with low expected payoff are accepted and results in poorer firm performance. These results add to the literature on managerial overconfidence by showing that the behavior of overconfident CEOs can be influenced and corporate governance could guide overconfident managers to invest selectively in good risky projects. In addition, my results offer insights into the puzzle why irrational managers are hired by showing that overconfident managers are hired because they could enhance firm performance if they were monitored by good governance practices.
The second essay is entitled "Book-Tax Income Differences: a New Measure of Earnings Management'. In this study I develop a comprehensive measure that captures taxable income related earnings management in addition to the conventional book income related accruals management. To confirm the adequacy of the new measure of earnings management developed in this study, I revisit several earnings management related issues previously investigated in the literature to see if the new measure performs better than conventional measures. The issues I re-examine include:
1. Executive stock option exercises - Insiders have incentives to use private information about forthcoming earnings to time their stock option exercises. Conflicting results have been documented by Bartov (2004), Efendi (2007) and Armstrong et al. (2009). My new measure that captures both income tax related and book income related earnings management may provide better insights. My results also show that executives time their stock option exercises regardless of the source of earnings management.
2. Firm credit ratings - Earnings management has led to downward revisions of firm credit ratings because credit analysts are able to see through the information content of earnings management (Ayers et al. 2009). The book-tax earnings management measure I develop can provide additional information to credit analysts. My results show that credit analysts react favorably to tax savings associated with tax planning activity when other information is obscured by earnings management.
3. Firm Value and earnings management - Outside investors are in general unable to fully comprehend the information underlying earnings management. The new book-tax earnings management measure I develop may help investors respond more correctly as the measure incorporates more information. My results show that once investors decipher the information of earnings management, they correct their errors in the following time periods.
Vu, Quang V..
"Two Essays on Behavioral Finance"
(2012). Doctor of Philosophy (PhD), Dissertation, , Old Dominion University, DOI: 10.25777/h4d0-7s83