In the 1980s, both wages and labor supply of poorly educated men fell substantially relative to those of educated men. Some observers have interpreted this positive association between changes in wages and labor supply as reflecting movement along stable labor supply curves. The author casts doubt on this interpretation by showing that the wage elasticity necessary to account, by itself, for the observed labor supply decline would greatly exceed elasticity levels typically found in prior studies. Analysis of Census data shows little relationship between changes in relative wages at the state level and changes in male labor supply. Also, panel data analysis shows no strong correlation between long-run changes in individual hours and wages. The small implied labor supply elasticities suggest that very little of the labor supply changes of men during the 1980s can be related to changes in relative wages.
