Lawrence Sherman: Researchers want to know what works to solve problems, and when they see in an experiment or some other research that a policy that is different from the one we’re doing now does a whole lot better — it has a big effect — that’s what they get excited about. When we do systematic reviews, we report effect size. We don’t report a cost effect size or a cost-benefit ratio. I have not seen a systematic review that compares policies on the basis of cost benefit, just on the magnitude of the raw effect size. So I think what we’ve got to do, and I hope NIJ can take the lead on this, is to put much more pressure on researchers to include cost data in their studies and research grants, to have cost estimates and effect sizes adjusted by the cost involved, because once we do that, we begin to get much more realistic about how much crime prevention we can afford, how much we want to spend on it. Governor Schwarzenegger wants a constitutional amendment to limit the prison budget to the budget of the University of California. That’s the kind of context of the cost of criminal justice which had been rising faster than anything else except health, at least in the state budgets, and that’s almost all in prisons. But we have to put in the context of 100,000 to 300,000 teachers being laid off — which the secretary of education predicts for this coming year — is that because we’re putting too much into criminal justice? It’s hard to compare the cost-effectiveness of primary education to the cost-effectiveness of prison, but that’s the kind of thing we’re going to have to do to have a meaningful democratic deliberation about how we want to invest our scarce tax dollars. I argue that using our current research on prisons and police would tell us that we’ll get a lot more benefit out of prisons if we limit them to the most dangerous people, the people who are, with new techniques, now predictable to be the Willie Hortons of the future. And if we restrict prison to the incapacitation of those people for a long period of time, we can be very confident that that will be the result for those people. Right now the evidence suggests that there’s a lot of people being put in prison to protect us from crimes that they’re never going to commit and that we have spent money on doing that — putting people in prison who aren’t that dangerous — that we could have been spending on police who could prevent more crimes out on the street if they were in the hot spots, if they were doing the problem solving that the research shows is effective. So at 35,000 per prisoner who’s locked up who doesn’t need to be there, that’s close to the salary of a police officer who could be out there deterring hundreds of crimes a year. There’s not that many offenders who are going to commit hundreds of crimes a year — there’s a few — but even they aren’t the ones who are going to be committing mass murder or horrible crimes against children, and with the capacity to combine the prediction of very serious crime with the prediction of huge benefits from policing, I think we have the data right now that says, let’s reconfigure this portfolio, reduce your investment in prisons, increase your investment in policing, and then manage the investments much more aggressively in the direction of the strategies that are effective.