Here’s a story that women in their 50s can talk about to those in their 30s. There are not many topics that have received as much attention in the social sciences as the situation of women in the labor market. Despite that, there are still a lot of things we do not know.
We know a lot about pay differences between men and women, and we know a fair bit about progress or limits to it in getting women into corporate leadership positions. But we know very little overall about the actual work that men and women do.
Along with my colleagues Shoshana Schwartz and Yang Yang, we looked into this with a nationally representative sample where we could compare the job experiences of men and women over time. We initially focused on white collar jobs and the college graduates in them for men and women who graduated from college in 1993. We saw what their first job was in 1994 and then the job they had 10 years later. These are men and women who are now about 52. Then we looked at the same information for women who graduated in 2008 and their first job in 2009 then again 10 years later in 2018. These are men and women who are now about 38. What did we find?
We compared the job requirements for the positions women held and compared them to the jobs that equivalent men held (same college majors, same grades, having children or not, and so forth). Perhaps not too surprising, the jobs women got in 1994 required less skill, broadly defined, than those the men held. That matters because jobs that require more skill pay more, are more interesting, and carry more prestige and authority. But what happened over the next 10 years? The gap got worse. Other evidence suggests that bias lessens when we have more contact with people, but that did not happen even when women stayed with the same employer and the same people over those 10 years.
How about for the graduates in 2008—did things improve? Actually, the initial gap with men was worse for women than it was for the 1994 graduates. It was a lousy job market for everyone, but why it was worse for women graduates is not clear. But over the next 10 years, the gap in job requirements for women and men narrowed considerably and was much smaller than it had been 10 years earlier. So, there has been some real progress more recently.
An additional issue we have heard a lot about in recent years is that women tend to be in—and possibly pushed into—jobs that make more demands of interpersonal skills including those we might associate with learning and caregiving. We did find that in both periods. But here is the interesting change over time: Those jobs that require more of these social skills actually pay more in the recent period. Even more surprising is that the bump in pay from being in a job that required more social skills was greater for women than it was for men.
Behind every silver lining there is a cloud, in this case a continued opportunity for plaintiff lawyers. We know a fair bit from other research about the pay gap between men and women. While it narrowed quite a bit in earlier decades, it has not moved much since then. There are other factors that affect pay besides the skill requirements of jobs, of course. But it is hard to claim that the pay gap between men and women now is due to differences in what jobs require of men and women. Something that is not easily explained is holding it up, and that could be bias.
Peter Cappelli
George W. Taylor Professor of Management
Director – Center for Human Resources for the The Wharton School
