Data Quality • 399
managers, but we do not need to invoke skill to explain the vast majority
of managed fund performance. While Fama and French do leave the door
open for skill, it need not be any more than a minor player in explaining
fund results.
A 2002 study by Mark Carhart of Goldman Sachs, Jennifer Carpenter
and Anthony Lynch of New York University, and David Musto of the
University of Pennsylvania demonstrated empirical evidence of survivor-
ship bias.
Evidence suggests that funds disappear following poor multi-year perfor-
mance… [W]e demonstrate both analytically and empirically that this
survival rule typically causes the bias in estimates of average annual per-
formance to increase in the sample length, at a declining rate.
23
e authors ...