The gendered pay gap in Canada, unpaid housework and earnings in the mid-1990s

Abstract

This study extends research into the gap between women's and men's average earnings in Canada. Among other predictors, the effects on pay for employment from workers' unpaid time doing housework, caring for children, and assisting seniors are investigated. Statistics Canada data from the 1994 and the 1995 cycles of the General Social Survey are used. The main analysis is for adults, 25 to 69 years old in 1994, who are employed full-time. The 1994 and 1995 surveys also provide subsamples of full-time workers whose spouses are employed. Other 1994 GSS subsamples are young adults 18 to 24 years of age who work full-time, and part-time workers 25 to 69 years of age, 80 percent of whom are women. Gender is consistently a strong influence on hourly pay rates across models with various controls and across subsamples, except in models of pay for part-time workers. In "family responsibility models" for full-time workers, women's disadvantage in average wages is altered little with control for marital status, housework, child care, and recent interruption to paid work (or career interruption).

For all except the part-time workers, women's education moderates the disadvantage in pay for women compared to men. Except among young adults, housework time accounts for a small part of the gendered difference in full-time pay and is consistently a significant disadvantage for women's pay, but not for men's pay. When housework is included, unpaid child care time has little additional impact on pay. Housework is a significant negative influence on pay even in a "life cycle model" with the added effects on pay of social origins (parental occupations and educations), a unionized workplace, and self-employment.

For workers in dual-earner couples, an interruption since beginning regular employment is negatively associated with pay for women, but not for men, in "family responsibility models," and is similar in importance to housework as an influence on women's pay. In the analyses for dual-earner couples, spousal education differences do not alter the effects of gender on pay. Women's disagreement with breadwinning as the sole role for husbands and fathers is associated with increasing pay among women. In these analyses, gender attitudes are slightly more important than housework and an interrupted career in accounting for variance in pay.

The overall set of results provide more support for explanations of the gendered pay gap based on gender relations theory than explanations derived from human capital theory.