The recently released White House report on Women in America (pdf) covers a great deal of interesting statistical data on the status of women in the United States. It also includes data indicating that in 2009 women on average earned about 75 cents for every dollar men earned. Some critics of markets claim this figure shows that markets discriminate against women, and they call for more regulation.
I will argue that the 25-cent difference does not necessarily show that markets discriminate. Rather, markets tend to pay people based on their anticipated productivity, and the differences between men’s and women’s pay largely derives from different choices that lead to differences in productivity. However, I will also argue that the differences in their productivity might result from sexism in other parts of society that cause men and women to be prepared differently for the job market, leading to a wage differential.
That 75-cent figure usually comes from comparing the average hourly salary of all women currently employed (sometimes restricted to full-time employment) to that of all men employed. What that figure does not represent is a pay difference between women and men in the same jobs and/or with the same level of experience, and so on.
When we economists look for evidence of wage discrimination, we first try to explain the differential by whatever other factors might be relevant. Whatever cannot be explained by those factors is then contingently assumed to be due to discrimination (or limitations of the data). In the case of the gender wage gap, the two major categories of “other factors” that account for most of the difference are human capital and preferences.
Human capital refers to the skills, knowledge, and experience people have that make them productive. The amount and type of such human capital is central to determining the wage one is likely to be paid. Compared to women, men tend to have more of the sort human capital that earns higher wages. For example, until recently men were more likely to be college educated, and they still tend to have more-on-the-job training and job experience than women.
Men who go to college are more likely to have majors that generate higher pay (such as computer science and engineering), while women tend toward psychology and education, which do not pay as well. Women are more likely than men to interrupt their careers to care for children. Women (and men) who do so tend to fall behind their cohort in job experience and in keeping current in their profession. Their wages thereby fall behind their cohort’s and are lower than they would have been had they not cared for the kids. All choices that affect human capital also affect wages, so discrimination in the marketplace is not required to explain pay differentials.
Moreover, because women are more likely to have primary child-care responsibilities, they tend to have different work preferences from men. For example, women tend to prefer jobs with flexible hours and fewer travel demands. Such jobs tend to pay less than ones with less flexible hours and more travel. Women more than men also tend to prefer part-time work, which pays less, and are less likely to work overtime than men.
Gap Nearly Disappears
When economists hold all these factors constant, they find that the gender wage gap largely disappears. Imagine two employees who are equal regarding human capital and preferences and who differ only by gender – what, if any, pay gap is present? In a study of people 27 to 33 who had never been married and who had no kids, the gap was only 2 cents. The strong implication is that if women want to earn as much as men, they need to adjust their human capital and preferences accordingly.
If we see the gap as a problem and the source isn’t the labor market, what might we do to correct the situation? Perhaps our expectations for girls and boys are different and thus we prepare them differently for the world of work. Rather than passing more laws, we could try harder to encourage boys and girls to pursue the same kinds of courses, majors, and jobs. We could encourage more men to see child raising as their responsibility as well, so that couples won’t blindly assume the woman will be the primary caregiver. If more women majored in computer science and more men majored in education — and if more men thought it was okay to work flexible hours so that they could care for the kids — we would make progress in reducing the gender wage gap. Markets will pay people for the value they produce, whatever the source of that value. Narrow the differences between men and women, and markets will narrow the wage gap.