Measuring the Body Count

The studies all suck.

After decades of studies demonstrating that the death penalty deters murders, that it increases murders, that it has no effect on the number of murders, we're pretty much left back where we started. The anecdotes are, for once, as meaningful as the data.

Here's what the anecdotes show:

• Some people say that they would have killed but didn't want to risk the death penalty.
• Some people say that they did kill in order to get the death penalty.
• Most people never thought about it. Either they acted out of sudden passion or, if they did think about it, figured they wouldn't get caught.

No shit. You don't have to be a scholar to figure that out. You just have to pay a bit of attention and then think about it for a few minutes.

Anyway, there's a new study about to be released in the American Journal of Criminology. Carefully done and peer reviewed, it indicates, according to a report in the Houston Chronicle, that executions in Texas, where they kill enough people to have a noticeable effect, have some slight deterrent effect. When you run the numbers, though, that slight effect adds up to maybe 10 or 12 a year. That's a fair number of dead bodies.

There are a few things to say about this.

First, there's no particular reason to believe it. I'm sure the authors were painstaking. The authors of bunches of studies have been painstaking. They generally stand behind their conclusions. But their conclusions disagree so much that about all you can do is throw your hands up in the air in frustration.

Besides this study has a couple of other curious points. I'm particularly taken by this one: Executions deter killings even if they get no publicity. That is, murders will go down as a result of an execution even if nobody knows about the execution. Maybe it's something in the air. Or maybe it's evidence that the study is BS. My money's on the BS theory.

Second, while 10 or 12 people a year is a lot, we'd save a whole lot more people if we got a few more drunk drivers off the road. Or if we hired a few more cops. Or if we actually made some efforts to rehabilitate. Or if we offered people an alternative to just hanging out. (I've said before that Bill Clinton's much maligned idea of midnight basketball was actually one of the most sensible proposals that ever got shot down.)

This part of the math is clear. Whatever you think about whether it's a good idea, executions fail the cost-benefit analsyis test. Scott Henson explains:

Perhaps there is some vanishingly small, short-term deterrent effect to the death penalty. Like the existence of a God that created man in His own image, I doubt it but won't rule it out. But in its current form, the death penalty is a political boondoggle and distraction involving a miniscule number of cases - a costly sideshow carnival act of the first order that distracts from more important discussions. Any legitimate cost-benefit analysis would find death-penalty deterrence doesn't measure up compared to underfunded but less-expensive programs that would save more lives and do more to reduce crime and deaths.

I'm all for the academic enterprise. I believe in it deeply. But I've yet to see a study of deterrence that made any sense. There may be plausible arguments in favor of the death penalty. (I don't buy into them, but they may exist.) That it deters enough crime to be worth keeping isn't one of them.