Hendricks, Patel and Zeckhauser (1995) find that the response of current to past returns for mutual funds in the presence of survivorship is non-linear. In our rejoinder to their paper, we verify their results through simulation, provide some intuition for why the result is true, and evaluate the power of their proposed test based upon the "J-shape" pattern. There's is a useful contribution to the growing literature about the issue of survival biases in empirical finance. It may help to explain puzzling results reported in the mutual fund literature, and may provide a guide for future experimental design. Our investigation of the HPZ results led us to a more complete understanding of how differential volatility affects survival-conditioned returns. Our simulations of the test statistic proposed by HPZ suggest that power of the test is dependent upon the absolute level of the threshold, as well as upon the magnitude of the cross-sectional differences in variance. While it would be useful to have a reliable test of the conjecture that survivorship is not driving an observed empirical result, we are only beginning to understand the kind of empirical regularities that survival may induce.

Download Working Paper?