I recall a story that scientists are often unable to explain the "tricks" performed by magicians. Scientists tend to be smart, but also rather linear thinkers. They are not used to their test tubes trying to deceive them. Something similar occurs in economics. I was reminded of this when reading a post on why women often pay more for things like haircuts and dry cleaning.
One thing that kind of bothered me about the post is that it ignored the highly competitive nature of these industries. In highly competitive industries, prices are very closely tied to costs. So if women pay more for services then presumably it costs businesses more to serve them.
I recall a debate on this in class, where I speculated that perhaps it costs more for dry cleaners to clean women's clothing. One student had actually worked at a dry cleaner and told me that women paid more even if they wore men's shirts. The student had a big smile on his face---how could I explain that!
I'm also a somewhat linear thinker, easily fooled by magicians, so normally I would not have been able to "explain that." But in this case my personal life offered a hypothesis. When I pick up dry cleaning I just stick it in the closet, without looking at it. My wife looks at it carefully, and often notices things that need to be redone by the cleaner. If the two genders differ in terms of how picky they are about dry cleaning quality (which is an unproven hypothesis, but at least plausible) then dry cleaners would have a cost justification for charging women more, even for the same shirt.
The economy operates in very subtle ways, and often when I read academic studies of issues like discrimination, the techniques seem incredibly naive to me. They might put in all the attributes of male and female labor productivity they can think of, and then simply assume than any unexplained residual must be due to "discrimination." And they do this in cases where there is no obvious reason to assume discrimination. It would be like a scientist assuming that magicians created a white rabbit out of thin air, at the snap of their fingers, because they can't think of any other explanation of how it got into the black hat!
The point of this post is not to claim discrimination does not occur---indeed I believe blacks are discriminated against in the labor market---but rather to point out that we cannot rely on academic studies of discrimination. They are no more reliable than scientific studies of magicians. I focus more on economic theory, combined with real world market failures that might be expected to lead to discrimination.
These subtle side effects explain why many government regulations that look good at first glance, are actually counterproductive.