As we go through the numbers about the gender pay gap, something I’ve promised to do here, we need to make a distinction between two different types of discrimination.

The first is what is usually referred to as “taste discrimination”. This certainly was very common but we really rather hope that it’s extinct now. What we mean by “taste discrimination” is direct discrimination: a company doesn’t hire blacks because they don’t like blacks. Or women get paid less simply because the boss insists that women should be paid less (this lasted a lot longer under the cover of “men get paid more because they have families to support”). Not only is this form of discrimination now illegal (almost all of the time, at least) it’s also to an economist irrational -if you refuse to hire a good black or female worker just because they are black or female then your competitor down the road gets those good employees cheaper and drives you out of business (the corollary of this is that what actually improves the wages and status of workers is many employers competing for their labor).

The second type of discrimination is “statistical discrimination”. It’s discussed in Tim Harford’s new book and Bryan Caplan has a good discussion of it here. While the original discussion was about race it does extend to our questions about the gender pay gap.

For example, many (most?) women want to have children. This puts those who do not, or who value their career above being at home with their children, at a disadvantage. For employers will rationally discriminate against women (that is, statistically) because any random woman applying for a job is likely to want to take time out to have and raise her children. Thus the employer will offer women as a class worse working conditions: perhaps lower pay (although that’s very difficult nowadays), or less training, or fewer promotions. Because that employer knows that she’s likely to lose that employee for some years of her working life.

That, in turn, means that women will on average have lower wages than men and thus make it more likely, if a family has a choice, that it is indeed the woman who raises the children rather than the man being a house husband… because he’s likely to be getting higher wages than she is. The two thus reinforce each other.

But our woman who doesn’t want to drop out of the rat race also gets clobbered by the same problems: she’s likely to be overlooked for promotions, paid less, trained less well. Quite how we get out of this isn’t really certain at this point. Caplan has a very inventive idea though:

Example: Some young women are 100% focused on their careers, and don’t want kids. Most young women, however, do want kids, and intend to strike a balance between work and family. That balance often involves receiving expensive job training from a firm, then quiting before the firm can recoup its expenses.

Under current law, an employer isn’t even allowed to ask about a female applicant’s child-bearing plans. If you wanted to blow up the glass ceiling, though, you should not only allow employers to ask; you should allow them to offer deals like “We’ll hire you, but your health insurance doesn’t cover pregnancy.” The career woman would be happy to sign, reassuring the employer.

How will that help women? It won’t! On average, it’s a wash: It will help career-minded women, and hurt the rest. And if you want to judge female workers on the basis of individual productivity, that is exactly what should happen.

We could go further: we would be able to remove some of the motivation for the gender pay gap by removing some of the general rights that surround pregnancy and child rearing. For example, there is a connection between long maternity leaves and women’s careers: we can see that in countries with long such times out of the workforce, women have greater difficulty in breaking the glass ceiling.

Quite whether we want to do that is another matter, but that’s the sort of question we’re going to try and work through over the next month or two.