UC Berkeley

This dissertation explores the extent to which managerial overconfidence affects corporate decisions. This analysis includes three essays, which address a wide range of corporate decisions including financing, investment, acquisition, innovation, liquidity management and advertising decisions.

The first essay introduces a fine-tuned test of the relationship between managerial overconfidence and corporate decisions by taking the chief financial officer (CFO) overconfidence effect into account. Ex-ante, I identify financial policies and non-financial policies such as investment, innovation and acquisition as the primary managerial duties of CFOs and chief executive officers (CEOs) respectively. I construct overconfidence measures for both CEOs and CFOs and test the impact of CEO and CFO overconfidence, both on financial decisions and on nonfinancial decisions. Based on a sample of 1,173 S&P 1500 firms, I find that financial policies are primarily affected by CFO overconfidence while only CEO overconfidence affects nonfinancial decisions. My findings demonstrate that managerial biases affect corporate decisions and managerial duties shape the ways in which top managers influence corporate policies.

The second essay investigates how overconfident CEOs allocate resources toward innovation activities. It argues that overconfident CEOs tend to have greater innovation input. To finance innovation, they save more cash out of the cash flow and spend more on innovation when the cash flow is high. Results from an empirical analysis of 1,015 S&P 1500 firms support this argument. Moreover, based on a series of financial constraint measurements, the effect of CEO overconfidence on liquidity management is found to be more pronounced in financially constrained firms and in highly innovative firms, but not in firms without financial constraints. With regards to innovation performance, overconfident CEOs tend to have more patents, but the overall quality of their patents is not significantly better than that of rational CEOs.

The third essay introduces a simple model of firm advertising behavior in monopolistic competition industries and applies it to the situation of managerial overconfidence. The model shows that the optimal advertising to sales ratio is determined by both firm advertising competency and consumer preference. Overconfident CEOs are more willing to use advertising as a means to convey the quality of their firms and products. Such overestimation of the effects of advertising by overconfident CEOs will result in overspending on advertising. When financially constrained, an overconfident CEO's tendency to overspend will be curbed to some extent, but his amount of advertising will increase with cash flows. An empirical analysis of 654 S&P 1500 firms supports these predictions. The distorted effect of managerial overconfidence is more prominent when firms are financially constrained and when the overconfidence measure is continuous.