Why Gender Discrimination is Overrated

Last week, I published a video on the main lessons I learned from reading Claudia Goldin's Career and Family. I mostly focused on the explanations for the gender wage gap. Goldin's claim is that the gap is driven primarily by career choices, then by occupation. Think of occupation as "doctor vs teacher" and career as "pediatrician vs surgeon."

What about discrimination? She says it explains little of the gap.

Woof. I got some pushback for mentioning that one.

I can take responsibility for that pushback. I could have done a better job explaining it. Instead, I approached it like creating a connect-the-dots, putting all the points out there and letting everyone else connect them. But sometimes dots are placed too far apart and the connection isn't obvious. So I'll first touch on some evidence related to discrimination, why even if discrimination existed everywhere it would still be small, and the true areas of discrimination we should be worried about.

Evidence on Discrimination

One comment accused Goldin of taking sides with the toxic part of the men's rights movement. Goldin is about as far away from that view as I can imagine. But the truth is that discrimination against women has significantly decreased the past few decades.

In 2016, some sociologists decided to test for whether employers discriminate against interviewing women. They sent over 21,000 applications to job openings in the US, UK, Norway, Spain, Netherlands, and Germany. But these were not normal applications. The resumes belonged to fake people. The researchers fabricated an entire work and education history, then randomly changed the applicant to either a woman or a man. The design lets the researchers control for all factors an employer considers before offering an interview to an applicant, so any difference in interview rates between male applicants and female applicants is attributed to gender discrimination.

What did they find?

They found no evidence for discrimination against women. In fact, as you can see in the graph, there was a slight bias towards women.

“Gender Discrimination in Hiring: Evidence from a Cross-National Harmonized Field Experiment”

That's for interviews, but what about pay? The general finding today is that a woman who works continuously through her career tends to move up the ladder at the same rate as a man. We'll get back to that later.

I can personally attest to the efforts that employers make to reduce pay discrimination. As I was finishing my PhD, I got a soft job offer. I say soft because they wanted to hire me and they gave me the pay information, but there was some uncertainty about whether the job still existed. They had two openings, and they wanted me and a woman, but we had to take on the risk that it might not go through. So I asked if there was room to increase the salary offer to compensate me for the risk. They replied that they could offer a higher salary, and they would also increase the salary offered to the woman so there was no pay discrimination. I don't know why they included that detail, but I was excited because that meant both of us were getting a better deal. And it made me see that employers are more sensitive about how pay discrimination arises.

I'm not saying discrimination in pay has vanished. But it has significantly declined. And it's likely to continue declining as more women are offered the same opportunities as men. For example, Yale Law School's most recent class was 51% women.

The real problem

For the sake of argument, let's stay there's still significant pay disparities between men and women attributed 100% to gender discrimination. To be specific, let's say that in any job, a woman makes 10% less than an equally qualified man.

What difference do career choices make? Well about 90% of elementary school teachers are women, and the average salary for an elementary school teacher is $60,000. When we look at physicians, only 36% are women and the average salary is $223,000.

Just from those numbers, it's clear that occupation choices are going to swamp discrimination. But what does Goldin mean by careers? In the video, I give the example of a surgeon vs a pediatrician. Even among those who become doctors, women tend to choose careers that are more like a pediatrician than a surgeon. The average salary for a pediatrician is $213,000 and the average salary for a surgeon is $414,000. Again, it's clear that the difference in careers is going to dominate discrimination.

Discrimination is real. But outright discrimination from employers is clearly a much smaller factor in the gender pay gap than the choice of occupations and careers.

The real discrimination problem

But why are women choosing different occupations and careers? As I explain in the video, it's because women are expected to take a larger role in the family. Women are expected to be on call for their kids. Men do not have those expectations.

That's why today we see a huge drop in earnings for women when a child comes. This paper shows that in Denmark, women take a significant earnings penalty when a child joins the family while men don't. An interesting feature of the paper is that they look at biological and adopted children and find no difference in effect, meaning this isn't because of the physical toll of having children. And the penalty lasts for at least a decade. This is a clear and stark effect of women taking a larger role in caring for children.

“Does Biology Drive Child Penalties? Evidence from Biological and Adoptive Families”

In a sense, this is discrimination. Not the discrimination we normally discuss, where a woman is passed over for an equally or lower qualified man. This is discrimination in expectations for men and women. And that difference in expectations alters everything about a woman's career path.

This is one of the challenges getting women into economics. Women are more likely to choose majors that lead to careers with greater flexibility in expectation that they will be taking care of kids. That's why you might notice much more women in your accounting classes. One way to get more women into economics is to show it leads to careers that let you have a family.

But more importantly, I think it's totally reasonable to ask men to share more in family responsibilities. I was lucky to have a dad like this. He's a work from home hipster, doing it before it was cool. I remember forgetting an assignment in high school, and it was my dad who showed up on campus to deliver it to me. He came to the football games and the track meets. He had the flexibility to spend time with his kids. Now that I have five kids, I'm trying to emulate his example.

If we want to attack the discrimination behind the gender pay gap, the biggest dent we could make is convincing men to be with their children.