Mutual Fund Survivorship

This article provides a comprehensive study of survivorshipissues using the mutual fund data of Carhart (1997). We demonstratetheoretically that when survival depends on multiperiod performance,the survivorship bias in average performance typically increaseswith the sample length. This is empirically relevant becauseevidence suggests a multiyear survival rule for U.S. mutualfunds. In the data we find the annual bias increases from 0.07%for 1-year samples to 1% for samples longer than 15 years. Wefind that survivor conditioning weakens evidence of performancepersistence. Finally, we explain how survivor conditioning affectsthe relation between performance and fund characteristics.