'Long Prison Terms Are Wasteful Government Spending'

Criminologist Mark Kleiman on replacing severity with swiftness and certainty


UCLA criminologist Mark Kleiman says he's "angry about having much too much crime and an intolerable number of people behind bars." Kleiman believes America's astronomical incarceration rate isn't making us safer. In his recent book When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton University Press), he argues that when it comes to punishment, there is a tradeoff between severity and swiftness. For too long the U.S. has erred heavily on the side of severity, he says, but concentrating enforcement and providing immediate consequences for lawbreakers can reduce crime while putting fewer people in prison.

reason.tv's Zach Weissmueller spoke with Kleiman late last year. To see a video version of the interview, go to reason.tv/video/show/professor-mark-kleiman-on-too.

reason: What motivated you to write this book?

Mark Kleiman: I wrote When Brute Force Fails because I'm both excited and angry. I'm angry about having much too much crime and an intolerable number of people behind bars, and excited because out in the field people are doing things that could change that.

There are more people behind bars in the United States than in any other country in the world. We have more prisoners than China does. We have 5 percent of the world's population; we have 25 percent of the world's prisoners. If the criminal justice system were a parent, we'd call it abusive and neglectful. It punishes too much and not often enough. We have a criminal justice system that does not know what every competent parent knows: that you change people's behavior by giving them clear rules and by enforcing those rules consistently and quickly and fairly.

reason: What are the main reforms you're suggesting?

Kleiman: The worst thing about our criminal justice system is its randomized draconianism. We're very severe in the way we punish people, but we do so very irregularly and very erratically. The basic reform is to substitute swiftness and certainty for severity. 

The average probation violation leads to no punishment at all, but an occasional probation violation will lead to six months in prison. That's the best possible way to fill up your prisons and not change anyone's behavior. The typical probation department does drug testing and tells people that they're not supposed to use. If the test comes back positive, the probation officer says, "Don't do that again." The next time it happens, the probation officer says, "Don't do that again." The third time, the probation officer says, "You know, if you keep doing this, you're going to get in trouble." The fourth time, he says, "This is your last warning." And about the eighth or ninth or 12th time, they're seeing the judge, and the probationer might be off to prison for six months. Lunacy. He had no way of knowing that the last "last warning" was really the last warning.

(Interview continues below video.)

reason: In the book, you say, "Concentrating enforcement attention works better than dispersing it." Explain what you mean.

Kleiman: If you've got a large number of violations and can't punish all of them, there are two approaches: You can punish more or less at random, in which case everybody learns that mostly he's going to get away with it. Or you can pick some subset of offenders, of offenses, of locations, of times—pick some part of the universe—and say, "OK, here's the rule within that part of the universe." Concentrated enforcement means deciding what you're not going to tolerate and who you're not going to tolerate doing it, directly communicating that threat to the people whose behavior you want to change, and then carrying it out.

reason: You advocate a system that favors swiftness over severity. Why?

Kleiman: If offenders were perfectly rational, crime control would be easy: ratchet up the severity of the punishment to the point where even a small probability of being caught means it's not worth it. That's what we've been trying for the last 30 years, and it basically hasn't worked. Everybody is more sensitive to immediate consequences than to future consequences. Everybody is more sensitive to certain than to uncertain consequences. Offenders are probably more like that than the rest of us: They're more reckless; they're more impulsive. Therefore it's even more important to move the consequences close in time to the events and to have the link be highly probable.

reason: You also say there's a tradeoff between swiftness and severity. 

Kleiman: We've known for a long time that swiftness and certainty are more important than severity. What's not adequately understood is that severity is the enemy of swiftness and certainty. A severe punishment can't be swift because there's a lot of due process involved, and it can't be certain, because you're chewing up a lot of resources. That 25-year mandatory sentence under the California "three strikes" law—that's 25 people who can't be locked up for a year. It's a little strange that the people who are loudest about opposing wasteful government spending haven't noticed that long prison terms are wasteful government spending.

reason: You describe a tension between safety and vengeance, results and catharsis. What do you say to people who argue that vengeance is an important part of justice?

Kleiman: I agree that vengeance is an important part of justice. The tone-deafness of official criminology and the academy to the need for vengeance, it seems to me, has contributed to the problem. If you acknowledge the need for vengeance, then you can say "but it ought to be proportional."

The resistance to using DNA testing to find out whether somebody is guilty comes from prosecutors, the cops, and the victims—not all of them, but often enough. The psychological mechanism is clear. From the victim's point of view, what matters is that somebody was punished for that crime. But it ought to matter a little bit whether it was the right person.

reason: What about critics who point out that crime rates have dropped, especially since the 1990s? Does this mean incarceration works?

Kleiman: Of course our high rate of incarceration to some extent must work, because people who are in prison aren't committing crimes on the outside. Now, if we counted the crime rate inside prisons, the crime drop would not have been as dramatic. I was a strong advocate of building more prisons, back when we had fewer than half a million prison cells. The first additional half million was well worth doing; the next million and a half, not so much. We made do with a fifth as many prisoners as we have now. Everyone else in the world does that. We ought to figure out how to do that.

As [Justice Fellowship President] Pat Nolan says, right now we're imprisoning a lot of people we're mad at. We only ought to imprison people we're afraid of. There are three groups of people who ought to be in prison. There are people who do such appalling stuff that we want to make an example of them—say, Bernie Madoff. There are people who are violent criminals and whose rate of crime is high enough that it's worth $40,000 a year not to have them in our hair. And then there are people who won't behave on the outside. You put an ankle bracelet on him, he takes the ankle bracelet off; he's picked himself a prison cell. Everybody else we can adequately punish and control in the community.

reason: Why do you think the justice system has evolved (or devolved) into an exercise in brute force?

Kleiman: The key thing about the American system is that we have police chiefs who are appointed by elected mayors and prosecutors who are themselves elected officials. They are very sensitive to what the voters want. And what the voters have wanted ever since the crime movement in the 1960s is revenge on the criminals. In some ways, I think voters were right to say, "Hey, crime's a big problem. We should do something about it." Unfortunately, they were badly misled by their politicians into thinking that random severity was a good solution.