Or perhaps one more big social policy push can bring full gender equality. Mothers in northern European countries work at rates about ten percentage points higher than American mothers do. However, they largely work at “feminine” part-time jobs because Scandinavian policies make it easy to work part-time but costly to stay home. American mothers mostly work full-time or not at all because part-time jobs here, sociologist Alex Janus suggests, have difficult hours, low pay, and negligible benefits. The Scandanavian model seems, despite great egalitarian efforts, to reinforce female-first care-giving.
But marriage markets aren’t as simple as pairing eligible men and women together at random. Even in places where the average earnings are a little higher among unmarried women, a few simple sorting preferences produce (straight) couples that overwhelmingly lean male-dominant. The most important preferences are for race/ethnicity and education: most couples match up along these lines. Within racial/ethnic and education groups, men earn more at every level. On top of that, the male partner is usually a few years older. With those parameters set, men will earn more in most couples. If you add an additional preference for higher-earning men within a couple—which is still what most people appear to want, whether they say it explicitly or not—then the male-dominant skew in the resulting marriages is even stronger.
Take Atlanta. Among childless full-time workers there, unmarried young women earn more than unmarried young men, but Census data confirm that just-married men’s incomes are higher than just-married women’s. Three-quarters of Atlanta couples marry on the same side of the college/non-college divide, and in cases where both spouses have college degrees, women earn more in just one-third of couples. Overall, only 38 percent of newlywed Atlanta couples have a higher-earning wife. As these couples advance along their careers, that percentage is unlikely to rise.
The growing prevalence of breadwinning wives is an important phenomenon, but it is not the end of gender inequality as we know it, and that prevalence is not irreversibly increasing.