Qunfeng Liao

Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Accounting



First Advisor

Bin Srinidhi


Prior literature has shown that managers have incentives to opportunistically and selectively withhold bad news from investors because of career concerns, compensation contracts, litigation risks, earnings targets, and empire building. In their 2006 paper, Jin and Myers develop the “Bad News Hoarding” theory which suggests that when managers conceal bad news for extended periods of time, negative information is likely to get stockpiled within the firm. When managers’ incentives for hiding bad news collapse or when the accumulation of bad news reaches a critical threshold level, all of the hitherto undisclosed negative firm-specific shocks become public at once, resulting in an abrupt decline in stock prices. Earnings management (EM) has been identified as the primary means employed by managers to conceal bad news. Earlier studies have shown separately that overvalued firms and firms characterized by high EM are associated with a greater risk of future stock price crash risk. In this thesis, I investigate the joint effect of extreme overvaluation and high EM on future stock price crash risk. It is shown that there is a robust positive relationship between extreme overvaluation accompanied by high EM and one-year ahead stock price crashes for a sample of U.S. public firms during the years 1995-2011. This result is consistent with Jensen’s (2004, 2005) argument that when a firm becomes extremely overvalued it sets up organizational forces and incentives that are likely to impair the value of the firm. However, I also find that extremely overvalued firms that are not accompanied by high EM as well as firms with high EM that are not extremely overvalued do not exhibit greater crash risk. The results are robust to alternative proxies of crash risk and EM and hold after controlling for endogeneity. The effects are more pronounced in the post-SOX period and for firms that engage in real earnings management (REM), are small size, or have low analyst coverage. In addition, I find that accrual earnings management (AEM) is positively associated with future stock price crash risk in the early stages of overvaluation whereas REM is positively associated with future stock price crash risk in the late stages of overvaluation. Finally, I find that extreme overvaluation with high EM is negatively associated with future stock price jumps. I interpret these results as suggesting that the incentives to conceal bad news through EM do not necessarily arise in all cases of overvaluation and that both extreme overvaluation and high EM should co-exist for the crash risk to increase. In this way, my results fine tune Jensen’s conjecture regarding overvalued firms.


Accounting | Business


Degree granted by The University of Texas at Arlington

Included in

Accounting Commons