Using Census Bureau databases that track individual workers and firms over time, a team of economists found that at age 25, college-educated women make about 89 percent of what college-educated men do. But that drops to about 57 percent at age 45.
The researchers wanted to find out how much of this 43 percent gap was due to choices women made, as opposed to differing skills or discrimination. While the data doesn’t address motherhood directly, the gap clearly widens during women’s childbearing years.
First, they looked at differences in earnings as people either stayed in jobs or switched. They found that most of the earnings difference happened when employees stay put — when men and women both stay at the same firm, men have faster earnings growth.
"If all the guys said ‘We also want to see our kids, we also want temporal flexibility,’ then this difference wouldn’t exist.” Claudia Goldin
But the pay gap also widens when men and women switch jobs: Men get a bigger pay bump when they join a new company, and are somewhat more likely to make the move. Even though women changed jobs almost as frequently as men, the pay bump was especially small for women who were married.
That’s in part because a married couple might make choices based on the “primary career,” which is usually the man’s, the researchers say. So a joint relocation might benefit the primary career and do little for (or even hurt) the secondary one.
Married women tend to be “more stuck in place” because of family obligations, says Claudia Goldin, an economics professor at Harvard and one of the authors of the study. “You’re attached to another person, maybe several people. So is your optimum their optimum? Maybe not.”
See What Your State Is Doing to Close the Gender Wage Gap
In a second study, the researchers looked at how the choice of different occupations came into play. Women are more likely to work in lower-paying jobs like retail or services, and in lower-paying occupations within fields like business.
Still, the researchers said that different occupations accounted for only about a third of the widening gender gap.
The wage gap was less pronounced for people without college degrees. But even there, men earn about 30 percent more from their mid-20s onward.
“They’re sort of flat. There’s very minimal growth with age. It’s not because high school dropouts or high school-educated women do so well, it’s more about the men not doing very well,” says Sari Kerr, an economist at Wellesley College and an author of both papers.
Goldin says that the difference for college-educated workers in large part comes down to women choosing flexibility over higher pay, something she calls “choice under constraint.” These women might want to stay in their job but not come in Friday mornings, for example, but because that is not an option for them, they’ll take a lower paying job with more flexibility.
Her previous analysis suggests that removing the penalty for flexibility would reduce and perhaps eliminate the gender pay gap.
“Women are disproportionately losing out during these periods of time when families are being formed,” she says. “If there are jobs let’s say in finance, where employers put tremendous demands on their workers, and only the guys are the ones who go into those jobs and women go into other sectors or small firms that have more temporal flexibility with lower salaries, it really takes two sides here to make that market. So if all the guys said ‘We also want to see our kids, we also want temporal flexibility,’ then this difference wouldn’t exist.”