Recent evidence from Fama and French (1992, 1996) and others shows that betas and returns are not related empirically. They interpret this as evidence against the validity of the capital asset pricing model and conclude that the beta is not a good measure of risk. This paper claims that usual...