Steven Levitt responds to some critics of his “legal abortion lowers crime” thesis, drawing a comments reply from Steve Sailer, who’s been among the most vocal opponents of the hypothesis. I have nothing useful to contribute on the econometric dispute about the data, but I do want to take a stab at one of the questions Sailer raises: Given that there’s an intuitive plausibility to the thesis—abortion will tend to select out children whose mothers aren’t prepared to raise them well, which is to say, children more likely (on average) to end up committing crimes—why might it be wrong if it is wrong?
One partial answer is suggested by Promises I Can Keep, a study of poor single mothers in Philadelphia. What the authors point out is that women’s disposition to take efforts to either prevent or terminate unplanned pregnancies is going to be a function of the perceived opportunity costs of (young) motherhood. That is to say, women for whom a pregnancy at, say, 15, is seen as derailing plans for college or a remunerative career have a much stronger incentive to postpone childbearing by whatever means—even if, objectively, they might be relatively well situated to care for a child—than women without such prospects. So even if in some sense refraining from abortion tracks “wantedness” in the aggregate, it may not track very well the other conditions that make for effective childrearing.