I use data from the 1968-1988 National Longitudinal Survey of Young Wo
men to investigate the lower wages of mothers. In pooled cross-section
al models, difference models, and fixed-effects models, the negative e
ffect of children on women's wages is not entirely explained by differ
ences in labor market experience. I consider two alternative explanati
ons for the residual penalties associated with having children: unobse
rved pay-relevant differences between mothers and non-mothers, which f
ixed-effects models show do not account for the child penalty; and par
t-time employment, which does account for some of the child penalty Ho
wever even after controlling for part-time employment, a negative effe
ct of children on women's pay remains.