Real Wages over the Business Cycle: Estimating the Impact of Heterogeneity with Micro Data

Abstract

New evidence on the correlation between the cycle and the real wage is provided by using panel data to adjust for the aggregation cum selectivity bias that arises when those who move in and out of the work force over the cycle have systematically different unobserved permanent and transitory wage components than those who do not. The results show that selectivity bias is present, for those with high transitory wages are more likely to lose employment, and that a true, compositio n-constant real wage index moves countercyclically, but less quantitatively than previous estimates using micro data have suggested. Copyright 1988 by University of Chicago Press.