The headline of the new American Association of University Women study on the differences between men and women’s earnings after college could have been, “Study Confirms Men and Women’s Choices Drive Differences in Pay.” Alas, you have to read well into the report summary and most of the news stories to get to this key finding.
Instead, the AAUW leads with this: “Women New to the Workforce Are Paid Only 82 Cents to Men’s Dollar.” From this heading, it’s clear that they want the media and readers to assume that the study found that two recent graduates—who sat next to each other during college, completing the same coursework, earning the same GPA and major, who then moved to the same city and were hired by the same firm for the same position—typically receive radically different compensation packages based on their sex, with the female grad receiving just 82 percent of her male colleague’s salary.
Yet that’s not what the study found at all. Yes, a wage gap persists even among new grads. But when the choice or major, hours worked, and career choices are taken into account, the wage gap shrinks to 6.6 percent. That 6.6 percent is “unexplained.” Discrimination could play a role in that gap, but so could other factors. For example, as the AAUW discusses, differences in men and women’s willingness to negotiate their salaries could play a role.
The AAUW doesn’t explore other potential explanations further—just concluding that discrimination likely is responsible for some portion of this gap—but other factors likely come into play, such as young women’s preferences for type of employer, specific specialties, willingness to relocate, and the shifts worked. In his book, Why Men Earn More, Dr. Warren Farrell explored these issues and found that the choices made by men and women on such particulars often drive men’s earnings up and women’s earnings down.
At a minimum, the AAUW and the politicians who use the “pernicious wage gap” to call for more government intervention into compensation practices should begin with the 6.6 percent “unexplained gap” statistic, instead of misleading the public by suggesting that an apples-to-apples comparison yields an 18 or 23 percent unexplained gap.