Phi Beta Cons

The Problem with ‘Control Everything’ Economics

A new study about the gender pay gap is making the rounds, and a Slate article (unintentionally) sums up the problem with this kind of research:

To understand workplace fairness, economists analyze how much of the overall gender pay gap — women still earn 78 cents on the $1 of what men earn — can be explained by the characteristics of workers and the jobs they hold and how much cannot be explained by anything except the person’s gender. In other words, they compare workers of the same educational attainment holding the same kinds of jobs — male college-educated electrical engineers and female college-educated electrical engineers. The American Association of University Women tackled the pay gap question this way and found that for college-educated women, the gap emerges as soon as they graduate. Their research shows that a woman earns 5 percent less the first year out of school than a man who goes to the same college, gets the same grades, has the same major, takes the same kind of job with similar workplace flexibility perks, and has the same personal characteristics, such as marital status, race, and number of children. Ten years later, even if she keeps working on par with the men around her — that is, continues to have the same level of on-the-job experience — the AAUW found that she’ll earn 12 percent less.

The problem is the assumption that economists can identify and measure “the characteristics of the workers” so well that the only remaining difference between these two groups is gender. In fact, there is no reason whatsoever to assume that a “male college-educated electrical engineer” is the same as a “female college-educated electrical engineer,” even after accounting for the school they attended, the grades they got, and a few measurable “personal characteristics.”

Thanks to some mix of culture and biology, men and women are different in lots of ways (assertiveness, emphasis on family in the family-career balance, etc.). Even with a boatload of “controls,” there’s no way to tell whether men and women are getting different results because they’re behaving differently, or because our evil patriarchal society is oppressing women.

Exit mobile version