Date of Degree


Document Type


Degree Name





Janet Gornick

Committee Members

Mary Clare Lennon

Paul Attewell

Subject Categories

Inequality and Stratification


Gender inequality, family, social policy, economic dependence


The gender revolution of the mid-20th century resulted in extensive changes in social, economic, and policy realms. With their unprecedented entrance into the paid labor market, women moved toward greater economic independence. However, it remains unclear whether these changes in women’s economic well-being resulted in positive change for women across the class spectrum. The large-scale entrance of women into paid labor was not accompanied by men’s equal take up in unpaid domestic work. Thus, women found themselves struggling to balance work and family life and as a result, gender inequality in employment outcomes persists. In this dissertation I explore the claim that in the absence of government provided support to reconcile work and family, less privileged women (i.e. women of lower social class) face greater challenges in balancing work and family life than more privileged women. If women’s responsibility for unpaid domestic work is causing women to trail behind men in the paid labor market, then women’s possibility to outsource unpaid domestic labor will affect their participation in the paid labor market. There is a greater ability among women of higher social class to outsource domestic labor or use market substitutes. Therefore, in countries that have not implemented generous work-family reconciliation policies I expect less privileged women to be more economically dependent on their male partners than more privileged women.

To assess these expectations, I revisit traditional analysis of gender inequality and develop a measure of economic dependence to assess women’s economic well-being across the class spectrum in 22 Western countries. While most evaluations of women’s agency have assumed that workforce participation is itself a measure of independence, this dissertation goes beyond this assumption and assesses women’s economic dependence by measuring women’s share of the couple’s joint income. I use a comprehensive earnings measure, augmented earnings, which consists of labor market earnings, as well as public and private transfers, and capital income. The analysis is framed in terms of national level work-family reconciliation policies, thus revealing how different policies influence not only women’s economic independence, but also the distribution of that independence across social classes.

I first provide an extensive cross-national descriptive analysis of women’s employment outcomes (e.g. employment rates, work intensity, and female to male earnings ratios), and women’s share of the couple’s joint income. Both employment outcomes and economic dependence are analyzed by family type and class status (class is measured with household income and education).

The findings from these descriptive chapters reveal three major gender gaps in modern Western society: 1) in employment rates, 2) hours among the employed, 3) and earnings per hour. I find that even when accounting for transfers, in all countries, women contribute on average less than 50% to their household income and in some instances as low as 20%. However, the degree of dependency does vary by class and country. Overall countries with more generous work-family reconciliation policies exhibit more class equality in women’s economic independence.

Following the descriptive analysis, I employ a random effect within-between model (REWB) to explore whether the effect of class (measured with education) on women’s economic dependence is moderated by work-family reconciliation policy. Specifically, I analyze the cross-level interaction effect of class and work-family reconciliation policy on women’s augmented earnings relative to the median man in their country. I run the regression models for all women, not just those in employment. Thus, my dependent variable, relative earnings, is a composite index meant to capture the three major gender gaps: in employment, earnings, and work intensity. Findings reveal that being less educated significantly decreases women’s relative earnings, but the association between education and relative earnings is moderated by work-family reconciliation policy.

In countries with an “ideal” maternity leave, exclusive paternity leave, more childcare support and more extensive working time regulations, the negative effect of low education on relative earnings is lessened. First, in countries that have these policies in place, women of all levels of education do better than in countries with less generous work-family reconciliation policies. Second, there is also less class inequality among women in countries with these policies. That is, the gap in relative earnings between women of different levels of education is smaller than in countries without these policies. Finally, I find that the moderating effect of work-family reconciliation policies can operate through more than one pathway, i.e., via employment rates, hours worked, or earnings per hour.

This work is embargoed and will be available for download on Monday, May 31, 2021

Graduate Center users:
To read this work, log in to your GC ILL account and place a thesis request.

Non-GC Users:
See the GC’s lending policies to learn more.