Kothari, Shanken, and Sloan (1995) claim that beta s from annual retur
ns produce a stronger positive relation between beta and average retur
n than ps from monthly returns. They also contend that the relation be
tween average return and book-to-market equity (BE/ME) is seriously ex
aggerated by survivor bias. We argue that survivor bias does not expla
in the relation between BE/ME and average return. We also show that an
nual and monthly beta s produce the same inferences about the beta pre
mium. Our main point on the beta premium is, however, more basic. It c
annot save the Capital asset pricing model (CAPM), given the evidence
that beta alone cannot explain expected return.