The Systemic Problem with Private Prisons

Alex Tabarrok links to an interesting NY Times article by John Tierney on the benefits of spending money on prisons vs. police. One thing I learned from it is that New York did not follow the general U.S. trend towards more and more imprisonment. Here's a snippet:
Dr. Ludwig and Philip J. Cook, a Duke University economist, calculate that nationwide, money diverted from prison to policing would buy at least four times as much reduction in crime. They suggest shrinking the prison population by a quarter and using the savings to hire another 100,000 police officers.

Diverting that money to the police would be tricky politically, because corrections budgets are zealously defended in state capitals by prison administrators, unions and legislators.

But there is at least one prison administrator, Dr. Jacobson, the former correction commissioner in New York, who would send the money elsewhere.

“If you had a dollar to spend on reducing crime, and you looked at the science instead of the politics, you would never spend it on the prison system,” Dr. Jacobson said. “There is no better example of big government run amok.”
There's been a lot of argument about private prisons focusing on whether they will treat prisoners too harshly, or perhaps too well. But perhaps the real problem is that when you have private prison companies you have created a lobby group that has an interest in keeping behaviours illegal and sentences long. It almost looks like U.S. politicians are on a mission to prove that the Marxists were right after all.

I wonder how much lower U.S. imprisonment would be if there were no private prisons.


Eric Crampton said...

This is pretty clearly testable as different states will have different proportions of total prison populations being handled by the private and public system, and different likelihoods that any increase in prisoner numbers will be met by private or public capacity expansions.

But note too that prison guard unions even in a public system have strong incentive to lobby for more use of prisons. Question then is whether public prison unions draw more rents than do private prison operators. If governments save money by outsourcing, the answer isn't a priori clear: some prison union rents are eroded via x-inefficiency rather than money that could be used for lobbying.

LemmusLemmus said...

Even after a quick readup on x-inefficiency, I don't think I understand your last sentence.

Otherwise, yes, an interesting test case; I would only add the standard caution that the correlation is probably not going to be a perfect representation of the causal effect.

Stuart Buck said...

See also the California case: http://reason.com/reasontv/2012/10/30/crowded-prisons-unions-and-three-strikes

Les Cargill said...

I don't think the existence of a lobby groups is sufficient to show that private prisons increase incarceration rates. the counterfactual should be something like - under non-private prisons, how many beds are there and with pure-private, how many are there? Include some sort of overcrowding factor estimate. Something like the "three strikes" law is non artifact of lobbying on its face; it's just a just-so story.

If there is an increase in capacity with private prisons, then this is most likely due to private prisons being run as an expense while non-private are run more as "capital goods".

LemmusLemmus said...

Les: I certainly agree with your first sentence, but I don't think *capacity* is the dependent variable I'd be interested in.