The evolution of the gender earnings gap amongst Canadian university graduates

The paper first offers some descriptive statistics which track the earnings patterns of male and female graduates in the years following graduation, and then reports the findings of a standard regression-based decomposition analysis which breaks the overall male-female difference in mean earnings into the part due to differences in the mean values of the explanatory variables and the part due to t [...] The cross-cohort trends clearly favour female graduates: male graduates’ earnings generally declined over time, with the mean earnings of the third group of graduates laying 6.3 percent below those of the first cohort as of the first interview, and 8.1 percent lower as of the second interview (these occurring at roughly comparable points in the economic cycle), whereas the mean earnings of the fem [...] The gender differences in mean earnings were then decomposed into the part due to differences in the mean values of the explanatory variables (field of study, job experience, province of residence, individual characteristics, etc.), and the part due to the differences in the associated coefficient estimates. [...] The effects of the differences in the explanatory variables (the first component of the decomposition) were evaluated using the male coefficient estimates, while the effects of the different coefficient estimates were evaluated using the female mean levels. [...] The overall beta and explanatory effects Following convention, results are presented in terms of the effects of the differences in the coefficient estimates (“Beta”) and the effects of the explanatory variables (“X”).