The Effect of Manager Gender on Hedge Fund Risk and Performance

PURPOSE OF THE STUDY: This master's thesis examines among the hedge funds managed by a sole manager, whether differences in perfomance and risk between ones managed by female (female funds) and ones managed by male managers (male funds) exists. For simplicity, in this study such differences are refered as the ”gender effects” of hedge fund managers. I also explore whether the gender effect on perfomance can be explained by the gender effect on risk. DATA AND METHODOLOGY: The primary data is extracted from the Academic Lipper TASS database. The sample use in this study includes 5697 live hedge funds that began operation in the period January 1994 to December 2013. The excluded funds are the ones that was borned outside that period of time or the ones that have ceased functioning. Manually collected data regarding managers’ gender is used to complement the primary data. Other data including Fung-Hsieh factors and three month T-bill rate and are retrieved on David A. Hsieh’s data library website and DataStream database. Hedge funds managed by a sole female manager are matched with similar hedge funds managed by a sole male manager using Propensity Score Matching method. The variables used for matching are the size of the funds, management fee, incentive fee, leverage usage, and managers’ capital investment. The gender effects on perfomance and risk are calculated as the average differences between the paired funds in Fung-Hsieh 8-factor risk adjusted return and net volatility. FINDINGS OF THE STUDY: There is a weak evidence that female funds overperform equivalent male funds by 0.17 percent monthly return over the period January 1994 to December 2013. However, during the two financial crises 1997 and 2008, female funds net excess return of about 0.16 and 0.31 monthly respectively compared to male funds, significant at 1 percent level. Volatility is not found to be significantly different for male and female funds. The gender effect on returns is not adequately exlplained by the gender effect on volatility, accounting for about 4 percent of the variance. I propose several explainations for the gender effect on perfomance based on the existing literature.