In this paper, we provide a comprehensive picture, circa the late 1980s, of
occupational gender segregation in Canada and its consequences for wages.
Our analysis reveals sensitivity of the estimated penalty to 'female work'
to both specification and estimation strategy. Our preferred estimates indi
cate that the wage penalties for women in female jobs in Canada are general
ly smaller than penalties in the United States. Of particular note, while t
here is some heterogeneity across worker groups, on average the link betwee
n female wages and gender composition is small and generally not statistica
lly significant.