crimeLike a lot of people interested in prison reform, I am skeptical of the idea that draconian punishments are good crime deterrents.

The basic logic behind the belief in crime deterrence is totally sensible.  If people are rational, they’re probably weighing the costs and benefits of their actions before making any decision.  If you’re considering whether to go to college, you’d consider the tuition bills and your lost wages while you’re sitting in classrooms instead of working and the effort it’d take to pass all your classes, then you’d balance that against increased earning potential and personal satisfaction and whatever else you might hope to gain (incredible beer pong prowess?). 

If you’re considering whether to stab somebody and steal his wallet, you’d consider the risk of being caught, the likely punishment you’ll face, the moral qualms you’ll feel later, and balance that against how much cash you think the dude is carrying.

For that latter calculation, increasing the length of prison terms, and making prisons more miserable places to be in, should make potential criminals less likely to stab & rob people.

All very sensible.  Unfortunately, it doesn’t seem to be true.  There’s some empirical evidence suggesting that it’s false, like similar crime rates in adjacent states with differing severity of punishment, and similar crime rates before and after bills affecting the severity of punishment were passed.  The data suggest that actual criminals simply are not making that sort of risk/benefit calculation.

(For someone else’s take on this, check out p. 11 of this reference from the Federal Judicial Center… from 1994.  And yet only now, over a decade after we realized that mandatory minimums weren’t helping, is there tentative talk from politicians about fixing them.)

paddleAnd there are reasons why we might’ve anticipated that the relationship (the “more punishment = less crime” idea) would not hold.  For one thing, behavioral economists have documented countless ways in which humans are not very rational creatures.  We pay too much for gym memberships.  We eat poorly, even though we know we don’t really want to eat that whole bag of jellybeans tonight (I too spend many evenings roiling in bed, clutching my belly, moaning whhhyyyyy ).  Males routinely underestimate risks and overestimate rewards, especially if there’s money involved, and especially if there’s sex involved.

That’s for the populace as a whole.  None of us are very rational.  It’s probably reasonable to assume that the sort of person whose circumstances are so dire as to make petty crime seem like the only option is even less likely to make accurate, level-headed assessments of risks and rewards.

Furthermore, a lot of criminals act upon the passions of the moment.  Not everyone –hardly anyone, I’d say — who commits murder is like Raskolnikov, plotting out the perfect crime in advance.  If you feel mad, pick up a gun, and shoot someone, there’s hardly time to think about how many years you might spend in prison, or whether or not you’ll receive the death penalty, or even how much remorse you’ll feel ten minutes later.

I attended a class at Bloomington Woodworks recently, and the instructor gave an interesting answer to the question, “If the joiner can do all that in a minute, why were we leveling all those boards by hand?”

He told us, “The machines make your work go faster.  But, if you make a mistake?  They make your mistakes go faster too.  While you’re working by hand, you can stop and check your progress and if something looks wrong, you can fix it.  If you made that mistake on a machine, it’d already be too late.”

firearm-409000_640Violent, impassioned people often kill people with guns.  Yes, they could kill people with knives or hammers or their bare hands, too.  But guns kill people faster.  That’s why the risk of successful suicide skyrockets in homes with guns — the time from a bad thought to being dead is so short.  Similarly, if you get angry and pick up a gun, there’s little time to think.  If you grab somebody by the neck, there’s at least a few extra seconds for the little voice in your head to ask, “Um, dude, what are you doing??”

So I’m skeptical that the absurdly long mandatory minimums in the U.S. actually accomplish anything.  Huge numbers of people in prison should not be there still.  I don’t think anyone whose behavior caused no harm to others should’ve ever been incarcerated.  But, beyond that, a lot of harmful people shouldn’t be in prison still.  Their sentences are often too long, too.

A big problem is our country’s dismal efforts toward rehabilitation.  We put convicted criminals into stressful, violent prisons, let them languish for many years, spend little or nothing on their education or job training… and then don’t want to let them out because they’re still “scary.”

Instead, criminals should be locked away for shorter periods of time, treated better while they’re in prison, and given the training they’ll need to successfully re-enter the outside world.  But few politicians would vote for that.  The problem is, people would argue that curtailing punishment makes crime more attractive.  They’d base their reasoning on that same inaccurate theory that criminals are making cold calculations of the risks and benefits of each illegal act in advance.

As I mentioned, for most crime, that theory doesn’t seem to be true.

And that’s why I was so pleased that Gretchen Morgenson’s recent “Fair Game” column brought my attention to a study documenting a type of crime for which that theory does hold.  Apparently there are some criminals out there who appear to be making nuanced risk / benefit calculations before their misdeeds.

Kedia et al., in their study “Evidence on Contagion in Earnings Management” (which I haven’t had a chance to read, sadly.  It’s still in press, so even though I can read all the fancy academic journals through the local university library, I can’t access it yet), showed a significant inverse correlation between regulatory action and earnings manipulations by others within an industry.  In other words, more punishment = less crime.  Exactly what politicians had been claiming was true for poor people selling drugs or stealing socks or vandalizing alleyways.

handcuffs-257995_640The relationship does not hold in general, but Kedia et al. provided evidence that high-level financial crime can be held in check by the threat of punishment.

In a way, it makes a lot of sense that this would be the case.  Humans aren’t very rational, but I think it’s safe to say that accountants are more rational than the rest of us.  Financial crimes also take a long time and only slowly build enough momentum as to seem irreversible.

If you get mad, pull out a gun, and shoot your spouse, it doesn’t matter how you feel that evening.  Your spouse is dead.  You screwed up.  The end.  But, if you’ve cooked the books?  You could go back the next day and correct the spreadsheet.  You could fix it next week.  You could probably fix it next year.

Many financial crimes require constant renewed commitment to criminal behavior.  Another good example is the — thankfully illegal, unfortunately still common — tendency for lenders to deny mortgages to black borrowers, or charge them much higher interest rates than they charge whites.  If you’re a banker and you turn away a qualified black borrower, well, maybe you can’t call that person up later that day to say, “I’m sorry, I was a racist jerk, I’ve reconsidered.”  But you could very easily approve the loan of the next qualified black borrower.

Criminal lending practices afforded the perpetrators hundreds, sometimes thousands of opportunities to stop breaking the law and instead do the right thing.

So, yes, it seems sensible that for these types of crime, the threat of punishment really would alter how much crime occurs.

Is it time for the sad coda to this essay now? 

For types of crimes that are not deterred by heavy punishment, the U.S. ruthlessly pursues draconian sentences.  But for the sort of financial crime that would be deterred by the threat of serious punishment?  In the U.S., perpetrators are typically let off scott-free.  They’re treated more gently than kindergarteners: they’re often not required to admit they did anything wrong or offer an apology.

(“Associated Bank denies any allegation that it engaged in discriminatory lending on a prohibited basis…”)