Our expectation is that beta in the fi rst period predicts beta in the next period, but that alpha in the fi rst period has no power to predict alpha in the next period. (In other words, the regression coeffi cient on fi rst-period beta will be statistically signifi cant in explaining second-period beta, but the coeffi cient on alpha will not be.) Why does this prediction make sense? Is it borne out by the data? LO.1