Missouri Tells Judges Cost of Sentences | Hacker News Reader

Missouri Tells Judges Cost of Sentences

nytimes.com

13 pointsraheemm36 comments

I thought "Wow, only $37,000 for a three-year sentence? I thought it cost more than that per year!" But apparently I was looking at California standards. Compare:

http://nicic.gov/features/statestats/?State=MO average cost per inmate $16K to

http://nicic.gov/features/statestats/?State=CA average cost per inmate $45K.

At the extremes, New York pays $55K per prisoner and Montana just $11K. This is insane.

It looks like California should be looking at solving its budget problems by outsourcing its prisons to Montana. This would save approximately four billion dollars a year.

I believe Schwarzenegger wants to do exactly that, but has been prevented by prison guard unions.

I don't care much for the jailers, but what about the families of inmates who might want to visit them? Will they be forced to travel out of state? Couldn't denying someone visitors be construed as cruel and unusual punishment?

Denying someone visitors would be cruel and unusual punishment. Making visitors drive for two days (or fly for two hours) in order to visit, though, isn't.

Think about it this way: in the days before motorised transport, were prisoners always kept within easy travelling distance of their relatives? I think not, and it wasn't cruel-and-unusual then.

For practical purposes, though, Montana to California is a bit too far. But Nevada spends only $20K, and there's not much of Calfornia which isn't within a reasonable daytrip of somewhere in Nevada.

Mexico probably spends even less, though contracting prisons out to Mexico might cause genuine constitutional issues...

Good point! Theoretically, you could have Montana come and run a prison in California. Of course, that would never be allowed, for the same reasons that it costs so much to house prisoners in California now...

You can probably subsidize the travel with part of the savings.

Won't happen.I think it would be horrible politically to offer vacations to families with members in jail, no matter how much was saved.

They won an injunction against him in 2007, but the decision was reversed a year later by the CA court of appeals, which ruled that prisoner transfer is in fact legal [1]. The state has transferred about 8000 prisoners to other states to relieve overcrowding (or rather, the unconstitutional conditions which result from it) and courts have approved plans to transfer more [2]. However, the issue is due for review by the state Supreme Court [3].

1. http://scholar.google.com/scholar_case?case=6294791264839700...

2. http://articles.latimes.com/2010/jan/13/local/la-me-prisons1...

3. http://www.rbg-law.com/prison-overcrowding-case-will-be-hear...

I recall there being some talk about taking over some closing prisons in Michigan in the last year or so.

They would save significantly less in Michigan so is that more of a capacity measure than cost savings?

At the extremes, New York pays $55K per prisoner and Montana just $11K. This is insane.

Population density in Montana is 6.5 people per square mile - it's the 4th largest state, but ranks only 48th in population, with under 1m people. New York has a population density of 408 people/sqm. I imagine the price of land and salary differences account for most of the extra cost.

I believe that prison for non-violent crimes is an absolutely antiquate concept and it doesn't belong to a modern society. There are far more effective ways to prevent crime and educate people. How much good could be done for society if we had millions of people doing community work instead of being imprisoned? But I'm afraid society is not ready yet for such a liberal stance, and there are powerful interests working against it as well (think private prisons).

I agree, although it would be important to define "violent" in some formal way. Is theft through fraud violence? What about inciting others to violence?

One possible hack you could try is defining three categories:

1. Definitely violent crimes, like stabbing someone and taking his wallet.

2. Definitely non-violent crimes, like selling illegal drugs.

3. Crimes we're not entirely sure about. If you imprison people whose crimes fall in this category, they should only go to prisons with other people in this category.

Of course nothing is a perfect solution here, but I think this might be a helpful way of thinking about it.

Perhaps it's best to say we shouldn't have prison sentences for victim-less crimes. That way, we can clearly identify drug possession or dealing as victim-less, and fraud clearly has victims.

Of course, at that point, you can easily argue that "victim-less crimes" shouldn't be crimes at all. Some sort of exception would have to be made for simple stuff like traffic/parking violations.

Traffic and parking crimes have victims, they're just not easy to identify. When you make crazy maneuvers on a highway without signalling and trigger a traffic jam that persists for two hours, you have quite a few victims, all affected in minor ways, but it adds up. Parking violations do not always have victims, but clearly you shouldn't be off the hook just because you got lucky that nobody handicapped showed up to find you had taken the spot reserved for people like them.

We don't need any special exceptions in order to keep those violations in the same category as fraud.

I wonder if the algorithm includes the taxes that people won't be paying due to sentences. That could get really complicated. For example, having a felony conviction on your record probably hugely reduces lifetime incomes. For certain people, the lost taxes might be much larger than the "direct" costs.

If we can predict that, shouldn't we also be estimating the cost of the crimes they won't be committing?

A burglar maybe steals $52,000 worth of stuff every year, and it winds up being worthwhile locking him up. But a child rapist has no major impact on economic productivity, so... yeah, I'm starting to see the problem with this mercenary approach to sentencing.

I'll respond out of order:

But a child rapist has no major impact on economic productivity, so... yeah, I'm starting to see the problem with this mercenary approach to sentencing.

This is almost trivial to account for, though, as long as you have a person looking at the numbers and making a decision - even if a judge sees that there's a large price tag if he sentences a child rapist, he's still going to do it, because the monetary value equivalent to "letting a child rapist back out into the world" is going to be ridiculously high, probably on the order of many millions of dollars. Same thing for other very serious crimes.

Where it gets more interesting is cases like "college kid caught carrying eight pounds of marijuana", where there's no clear and present threat to anyone if the kid was given some sort of reduced sentence.

If we can predict that, shouldn't we also be estimating the cost of the crimes they won't be committing?

Sure, why not? In theory we lock people up because it helps society, right? So why wouldn't we want to consider all the possible higher order effects of different sentences on society as a whole?

The problem is mainly that figuring out these effects can be very difficult, and I suspect that some of the second and third order effects end up being comparable in magnitude to the first order ones (direct cost of sentencing).

Plus, in many cases we're really locking people up to satisfy a thirst for vengeance, not necessarily to help society in any way whatsoever. It would be nice if we had a better measure of what it actually costs us, though.

It's not going to be used that crudely. The child rapist is going to get a hefty sentence regardless of the cost because people want to punish serious crimes more than they want to save a couple of bucks.

Where it does come in handy is for relatively minor crimes[1], serving as a spur to consider whether the cost of incarceration is really offset by the benefits to society.

[1] Things like drug offenses, etc.

If we are committed to the mercenary approach, we would have to define how bad a murder is "in dollars".

If you think we should be focused on justice instead of a cold cost-benefit analysis, may I suggest defining how valuable justice is, in dollars? (I'm joking, I think...)

You mean putting a dollar number on a human life? This is done everyday by actuaries primarily for insurance purposes.

Whatever dollar value may be put on a human life for insurance purposes it's not an actual value. For instance, I can't pay that amount and then kill you.

I meant that I was joking about putting a dollar number on "justice". Although, I'm not sure how unreasonable that is.

I doubt it. It's rare politicians consider costs past the next budget year.

This is actually a bit more interesting than merely providing judges with the costs. FTFA:

The concept is simple: fill in an offender’s conviction code, criminal history and other background, and the program spits out a range of recommended sentences, new statistical information about the likelihood that Missouri criminals with similar profiles (and the sentences they received) might commit more crimes, and the various options’ price tags.

The fact that the system actually gives info about likelihood of committing future crimes is pretty neat, if it's done well. This is one of the first practical moves I've seen towards data-driven sentencing, and if it spreads and grows, it could be the start of a major tidal shift in our entire philosophy of criminal justice.

Are you suggesting that people should be judged on the probability that they might commit more crimes in the future, that is, for crimes they have not committed?

Are you suggesting that a criminal with a strong history of reoffending - eg, a serial rapist - should not have that taken into account during sentencing? A large part of the reason why we jail people is to reduce their risk to society.

I guess the problem is how far the statistical analysis should be taken.

Should someone who has a history of violence that commits his first murder get a longer sentence? I imagine the consensus would be yes. But what about someone who has a history of non violent crime? Antisocial behaviour?

Should it be taken into account if you're a loner? What about your name - it's possible that's statistically significant (even if you're already starting with people who have already murdered)?

What about race?

So it pretty quickly turns into something I disagree with, probably long before the statistical significance wears off. After a few minutes of reflection, I can't figure out exactly where I draw the line.

Abso-f@#$ing-lutely.

If there's zero probability that someone's going to commit a crime again, then any sentence that they receive is purely punitive, and it's there for only two reasons, 1) to satisfy our emotional need to punish someone that did something wrong, and 2) to signal to other potential offenders that if they do the same thing, they won't get away with it.

But if someone is very likely to commit more crimes, locking them up adds 3) they will not be able to commit these crimes while they're incarcerated.

The relative values of 1), 2), and 3) shift sharply depending on the seriousness of the crime and the likelihood of future crimes, and the value that we as a society gain (or lose) by locking someone up depends critically on how they add up.

If two men commit exactly the same murder (physically speaking), but one of them does it because he caught his wife in bed with another man and the other does it because she didn't cook the steak rare enough that night, I would be perfectly fine with the first man getting a lighter sentence: it took a very extreme situation for him to resort to murder (though it's still not justified by any means, don't get me wrong, and he should still be punished based on 1) and 2)), whereas the second guy is pretty clearly a loose cannon, and is likely to pose a far greater threat to everyone if he's allowed back on the streets.

They already are, by the ultimate opaque system, semi-random collections of human brains. Recall all the senses of the word "judge", after all; judges certainly take that into account as part of their human decisions.

This is more transparent and more subject to validation and verification. Are you sure you think this is actually a step backward?

I did not say so, but I do think this is a road to hell, paved with good intentions. 'Minority Report' is on that road.

Judges may sometimes or often be biased, but, institutionalizing that is IMO a step too far.

Having said that, I think there is gray area; for example, I do not believe in the British "if it's murder, it must be life in prison", and I do think that sentences like "follow this anger management course or go to prison for three months" can be based on statistics of repeat offending.

Judges' sentencing criteria are not opaque, because they're set out in the opinion and subject to appeal. Sentencing guidelines have been a mixed bag at the federal level as far as quality of results goes, and their original mandatory nature were ruled unconstitutional by the Supreme court in 2005, largely due to sustained opposition by Justice Scalia.

I think what he's just saying is this:

Punishment is typically expensive; rehabilitation is typically cheap and more effective in reducing future crimes after the person is out in society again.

Punishment, therefore, should be reserved for the more heinous crimes where rehabilitation isn't likely to work, and rehabilitation should be used for less-heinous crimes where rehabilitation is likely to succeed.

Yes—not on a person-by-person basis, but on a crime-by-crime basis. This is perfectly in accordance with judging on the intensity of or the harm caused by a crime.

No, that someone CONVICTED should be sentenced based on the data.

I can see it coming...

Judge: The program says that the defendant should be put in jail for 3 years.

Defense Lawyer: My expert witness says that the program is buggy.

Prosecutor: Can you prove that?

Defense Lawyer: I would, if the program were open-source...

Hey, if there is a practical incentive for judges to incarcerate less people in this psychotic, off-the-deep-end "justice" system where almost nobody who is incarcerated long-term should be, I'm all for it. Whatever it takes.

On the other hand, I am intensely wary of formulations like this: "The imprisonment of X will cost taxpayers $YYY,YYY over his lifetime." That sounds awfully similar to posters for the Nazi Aktion T4 euthanasia program, e.g. "This mental invalid costs the Reich 60,000 marks over his lifetime. Do the merciful thing... it's your money." There has got to be a more delicate way to put that.

Reply on news.ycombinator.com