On pp. 82-84, Wheelan inserts a brief discussion on what he calls "rational discrimination." He uses a hypothetical law firm (go figure) to highlight a hiring decision between two candidates: one young man and one young woman, both fresh out of Harvard Law (in other words, both qualified).
He posits that "the rational choice is to hire the man." This is because
"[d]emographics suggest that both candidates are likely to start families in the near future. Yet only the female candidate will take paid maternity leave. More important, she may not return to work after having the child, which leaves the firm with the cost of finding, hiring, and training another lawyer."In short, the woman has a higher probability of taking the parental benefits and running. Now, before you curse Wheelan, he does take care to emphasize that "playing the statistical averages may offend our sensibilities and violate federal law." I will note that it almost certainly violates state and possibly local law as well.
Playing the odds in this case is a substitute for actual information on an individual. The man could be looking to stay at home and raise the kids, or the woman could be looking forward to being a working mother, or maybe neither of them has any interest whatsoever in having kids. There's simply no way to tell (and no do not ask them about kids in the interview!)... or is there?
Wheelan offers this solution: offer a "generous but refundable maternity package. Keep it if you come back to work, return it if you don't." The point? Interviewees that plan to take the benefits and run will generally self-select out of jobs with employers that offer refundable maternity packages. The plan won't work if you can't take the money. Further, employers can then give better maternity benefits because they incur fewer costs from employees leaving.
This illustrates a classic information disparity problem. The employer lacks information on the individual. By offering generous refundable benefits the employer will get better candidates who plan to stay (for the higher maternity benefits) and the candidates who planned on leaving will likely not choose to work there (opting for an employer who will let them take the benefits whether they come back or not).
One final note: Wheelan also explains that all of this will be unnecessary "when men assume a larger share of child-care responsibilities." In other words, the "rational discrimination" (or "profiling") becomes irrational when the profiles shift to equality.