Tutorial

Account

Forums

Thursday, October 06, 2005

Bennett and Freakonomics

William Bennett claims to know that by aborting all black babies in America, the crime rate would go down.  I thought it would be wise to analyze this in a value-free manner.

There are plenty of people who will pass moral and ethical judgments on what he said and why.  I don’t find a lot of use in this kind of judgment, so I’ll try to stick to logic.

It seems that Bennett’s claim comes in part from the government statistics on the race of convicted criminals.  The portion of the American citizenry that is black (12.3%) is much smaller than the portion of convicted criminals that are black (24.6%).  However, the portion of the citizenry that are males over 20 (34.4%) is also much smaller than the portion of convicted criminals that are males over 20 (86.0% male times 94.9% over 20).  In fact, the disparity in proportions in the population and in convictions for blacks (whose conviction proportion is twice as high) is much smaller than that for males over 20 (whose conviction proportion is 2.5 times as high).

Using the same logic Bennett used, one could say that killing off all the males over 20 would be even more effective at reducing crime than killing black unborn children.  But neither conclusion is yet reasonable.  This analysis that I’m doing is the kind of thinking that was done by the author of Freakonomics.  In Mr. Bennett’s radio show, both the caller and Mr. Bennett agreed that the conclusion that author came to - that abortion has reduced the crime rate - is incorrect.

The argument that I imagine Freakonomics uses is that the same conditions that motivate an abortion also motivate criminality.  Specifically, unwanted children are more likely to become criminals, and poverty motivates poor people to break laws and avoid having children.  In all cases of pregnancy in which these conditions will either produce a criminal or an abortion, it cannot be avoided that the abortion lowers the crime rate.  Let’s call these cases “criminal or dead” cases.  Among the rest of the cases, are there pregnancies in which not aborting might lower the crime rate?  Of course there are, but if there were enough to counteract number of “criminal or dead” cases, then the crime rate among babies that are unwanted and/or come from poverty would have to be lower than in the general population.  Freakonomics merely points out that the the same conditions produce both criminalty and abortion.  So when the abortion is chosen, some part of the criminalty is eliminated.

The cause of skin color has nothing to do with the reasons people become criminals.  Mr. Bennett has confused the result and the cause.  The direction of causality is one of the major stumbling blocks for most politicians.  To test his claim on the direction of causality, consider turning his phrase around:  “If the crime rate goes up, then there would certainly be more aborted black babies.”  Who wants to raise a child in a world full of criminals?  And of course, there would also be more aborted white babies.  Likewise, for my example, “If the crime rate goes up, there’ll be a lot more killing of males over 20.”