The New Economics of Crime and Punishment

Our criminal justice system is a disaster. The incarceration rate in the US quadrupled between 1980 and 2000. It now costs more than $70 billion a year to keep 7 million people behind bars, on parole, or on probation.

In fact, we may be at a breaking point: Last year the Supreme Court upheld a lower court ruling that due to severe overcrowding in California prisons, the state was violating inmates’ Eighth Amendment rights—you know, the ones that mention “cruel and unusual punishment.” The court gave the state two years to fix things. Constructing new prisons to house more than 30,000 people is impossible in that time, which means California must reduce its inmate population by other means.

The best way to address this problem is not with ad hoc political negotiations or by rehashing the age-old debate about retribution versus rehabilitation. What we need is convictonomics: a coldly rational economics-based approach to crime and punishment.

Assigning a price to an assault or a burglary might sound preposterous. We like to pretend that justice exists on a philosophical and moral plane separate from dollars and cents. Yet money is integral to the criminal justice system. Think of the way fines are imposed and damage awards assessed for injuries and wrongful death. New research suggests that we can, and should, measure the dollar value of capital-V values—things like freedom and the fear of crime—and trade them against one another in a comprehensive cost-benefit analysis.

David Abrams, an economist who teaches at the University of Pennsylvania Law School, believes that researchers can now empirically examine matters of crime and punishment thanks to enormous digital archives of data about everything from bail to sentencing to recidivism. Researchers today also have tools like instrumental variables, which can be used to estimate cause and effect in situations where it’s not feasible to run a controlled experiment. “This is a more scientific way to look at criminal justice, one that fills in the logical steps in a way we couldn’t in the past,” Abrams says. “I want to take these technical ideas and make them useful for policymakers.
In a forthcoming paper for the Iowa Law Review, Abrams presents a cost-benefit analysis of possible modifications to incarceration policy that should be of interest to officials in California. He puts costs on things like deterrence, incapacitation, housing of inmates, and the value of freedom. Based on his analysis, the huge net benefit of a onetime release of prisoners outweighs the fact that some of those released will commit more crimes. Even if just 1,000 prisoners were released 60 days early, the benefits would outweigh the costs to the tune of $5 million.
Another concrete recommendation: reclassify some low-level offenses so they don’t lead to jail time in the first place. “That seems to be a way to spend much less money on incarceration without increasing crime all that much,” Abrams says. “With theft, for example, about $2,000 per prisoner per year would be saved if we nudged the bar a little higher as far as what does or doesn’t land you in jail.” (And lest you think this is a soft-on-crime scheme in disguise, Abrams concludes that for robbery, reducing incarceration is not beneficial. For that category of crime, we may want to lower the bar for prison time.)
Some people recoil at the idea of using cost-benefit analyses in matters of criminal justice. The left-leaning critique: You can’t put a price on freedom. People on the right, meanwhile, insist that a criminal’s loss of liberty shouldn’t be incorporated as a cost at all—some even see it as an added benefit.
But what the critics miss is how convictonomics could reduce strain on the entire system while also delivering a more consistent approach to justice itself. Because the true injustice is the status quo: a system driven by arbitrariness, plagued by inefficiency and exorbitant costs. Oh, and it’s unconstitutional too.
You see, justice should be blind. But she should also make sense.
David Wolman

0 yorum: