Failure Risk and the Cross-Section of Hedge Fund Returns

Jung-Min Kim
2009 Social Science Research Network  
On average, hedge funds fail slowly rather than through sudden crashes. I model a fund's probability of failure using a dynamic logit regression and find that fund failures are predicted by past performance and fund flows measured with a lag of seven months. Hedge funds fail as poor performance over a period of time leads to fund withdrawals by investors. A fund's failure risk predicts negatively the fund's future returns. Sorting hedge funds into quintiles by the predicted failure probability
more » ... ailure probability based on information lagged by seven months, I find that the return spread between the two extreme quintiles is 7.6~8.9% per year after adjusting for nine commonly used hedge fund risk factors and a return smoothing effect over the period of July 1996 to September 2007. The negative failure risk effect on future fund returns is sharply higher for funds with weak share restrictions and is not subsumed by the findings of the prior literature. † I would like to thank Kewei Hou, Andrew Karolyi, Scott Yonker, seminar participants at the Ohio State University, and especially my dissertation advisor, René Stulz, for their helpful comments and suggestions. The Dice Center for Financial Economics provided financial support. All errors are my own. ‡ Please direct correspondence to Jung-Min Kim,
doi:10.2139/ssrn.1324341 fatcat:nsfz5qhunnglpfsioiysmuv3pi