Levitt’s comment is reasonably fair-minded, although open to challenge in several places. I’m going to focus on one.

Levitt writes: “7) There is one thing I would take Bennett to task for: first saying that he doesn’t believe our abortion-crime hypothesis but then revealing that he does believe it with his comments about black babies. You can’t have it both ways” (emphasis Levitt’s).

I think if Levitt reconsiders this question he will see that he is making an error here. These are two different theories and they’re not open to the same objections.

Levitt argues that states that legalized abortion early saw their crime rates drop early. That sort of argument is open to counter-arguments of the same type: Well, what about the fact that the same states saw their crimes rates rise earlier beforehand? But the reason Levitt’s argument is plausible to people who don’t follow all of these arguments about crime rates and correlations is that they make a simple mistake.

People fall into confusion all the time when they ignore the fact–which Levitt, to his credit, mentions in his book–that the legalization of abortion increased the conception rate by 30 percent (while causing the birth rate to drop by 6 percent). Many times I have heard pro-lifers make comments about how there would be 40 million more taxpayers if not for Roe. That sort of comment is what spurred Bennett’s discussion. There are a lot of flaws with this type of reasoning–Bennett was, in my view, absolutely right to note that it is a distraction from the decisive moral issues–but what’s important here is to remember that not every abortion represents a child who would have been conceived absent legalization. The flip side also holds true: As paradoxical as it sounds, there are children who would never have been born without legal abortion.

So the effect of legal abortion is to change the mix of people in the next generation somewhat and to reduce its numbers somewhat. We have no reason to be certain that the net effect is to reduce the number of children in crime-prone groups. For example, it could in theory be that the effect of abortional choice is to reduce the number of kids born to middle-class, married black couples while raising the number of kids born to poor black single mothers–in which case it could raise the crime rate. Whether it’s true depends on the validity of Levitt’s empirical claims; its truth can’t simply be deduced. (I am at work on a book that deals with this question, among others.) Bennett’s hypothetical, in which all black babies are somehow aborted, isn’t open to the same objections.