Bryan Caplan  

Why Not? Weight Classes for Prisons

From the Vault: My 1983 Memo o... Odious Debt...
Why do we have separate men's and women's prisons?  You don't have to envision the alternative for long to have your answer: Co-ed prisons would be a living hell of rape and brutality. Or perhaps I should say: Even more of a living hell of rape and brutality, because that already describes single-sex prisons today.

I'm not raising this issue to ruin Mothers' Day.  It's a lead-in to a serious proposal for prison reform.  Once you know how to make prisons even worse, you know how to make them better.  Namely: Reduce the variance of strength and aggression within single-sex prisons by separating prisoners into something like "weight classes." 

In boxing, heavyweights don't fight featherweights.  It's not a fair fight.  But in prison, heavyweights serve their time side-by-side with featherweights.  A simple remedy for rape and brutality, then, is split up prisoners by size and strength.  You could assign the various classes of prisoners to different wings.  Or if that's too logistically difficult, you could assign each prison a weight class, then reallocate existing prisoners. 

Admittedly, we already have prisons for different kinds of offenses - minimum security, maximum security, and everything in between.  Like separating men and women, this probably makes the prison experience a little less hellish.  But there's still a long way to go, and my proposal seems like an obvious and cheap improvement over the status quo.

Note: The "weight class system" targets prisoner-on-prisoner abuse.  It has no direct effect on the comparable problem of authority-on-prisoner abuse.  Even there, though, my reform would do some good.  It's probably easier for a guard to get away with raping a prisoner in an environment where prisoners are raping each other on a regular basis. 

On the surface, it seems like weight classes for prisons should appeal to people from many different ideological standpoints.  Unless you tacitly favor prison rape (perhaps as an extra deterrent in an overly forgiving legal system?), it's hard to see the downside.  Still, I don't expect my proposal to sweep the world.  Does anything other than status quo bias stand in its way?

Comments and Sharing

COMMENTS (25 to date)
David Youngberg writes:

Two things I can think of: one is that you'd just change what prisoners use to dominate another prisoner. Gang members within a prison will teach each other martial arts (or just better fighting skills) and the best student becomes the rapist. Trading bulk for fighting talent.

Second option, and less likely but potentially more problematic, is that it increases the danger of riots at the heavy-weight prisons. Yeah, you would compensate guards more and the guards would be better armed and in better shape, while at the feather-weights the opposite is true: you could call it a wash (I have no idea what the distribution of weight is amongst prisoners) but I imagine that there's some solid economies of scale in the danger/effectiveness of riots when it comes to inmate strength. If that's true, then there's an additional incentive for the mid-weights to "bulk up" and sent to a different prison, since the cost of rioting relative to the benefit is lower (as a way to blow off steam, more important since raping is more expensive). This could create a cost spiral, more bulky prisoners, more rioting, more prisoners, etc. This story, though, doesn't affect the rape rates so that might be enough to still call it an improvement.

OneEyedMan writes:

I doubt bulking up will be a problem. In weight class sports everyone tries as hard as possible to be in the lowest weight class so that they can dominate it.

This seems like a great idea, though the transition costs of moving prisoners would be significant.

One concern is that we often want to put prisoners near their families as a way to motivate their rehabilitation. I don't know if this is a dominant factor, but if we sorted prisoners this way then fewer could be near their families and we would have the associated lower rehabilitation rates.

One thing we could do is build a new prison for a lower weight class and make it an option. That would pull out those who saw themselves as likely targets of prison rape. Maybe that doesn't fix the problem because it just raises the amount of rape on those who remain, possibly leading to an equilibrium where everyone sorts on this criteria, but rehabilitation is less likely because most people are far from their families.

Maybe there are enough prisons that this doesn't matter.

Personally, I think it would be a good start just to stop putting non-violent offenders in with the violent ones. Despite the existence of minimum security jails white-collar crimals and minor drug offenses are still placed in the general population with murderers etc.

But otherwise the weight classes seems to be an improvment over the current system so I'd be for it.

NiccoloA writes:

Prisons have always been inhumane places and looking back to the theoretical invention of them by Jeremy Bentham, the reaction to the common practices of justice seem to have backfired, at least in my opinion.

At least in a system of penal labour, you had some retraining mechanism that occupied inmates.

What the US has today is just a breeding ground for violence and terror. People leave prison far worse than they went into it.

agnostic writes:

It would never fly because of "disparate impact." Blacks and whites don't differ in height, but they do in muscularity, strength, whatever you want to call it. You can hear it now:

"Oh, so you're saying put all the skinny white boys with each other where they'll be safe, meanwhile all the big black dudes get crammed in together to duke it out over territory like a pitbull fight."

It would improve the status of whites -- fewer big black guys around to rape them -- while lowering the status of blacks -- the average jailmate would be a lot stronger and menacing. Kiss this policy good-bye.

sytse writes:

It is likely that prison management already tries to reduce violence by sorting the prisoners smartly, for example by gang membership. Any top-down directions (like weight-classes) might reduce their opportunity to do so.

Lode Cossaer writes:

Maybe that path dependency is an argument? That the gains of the change are offset by the costs?

PriceTheoryEconomist writes:

Another, more empirical link would have been to Katz, Levitt, and Shustorovich's "Prisons Conditions, Capital Punishment, and Deterrence", which found, using a state-level panel data set, that when the death rate of prisoners in prisons go up, crime goes down.

It is entirely reasonable to believe that prison rape rates are as robust a deterrent as prison death rates.

Robin Hanson writes:

Great suggestion Bryan! This sort of thing works better the further you can house prisoners from home, allowing bigger scale economies from being selective about where each one goes.

David R. Henderson writes:

One thing I noticed missing in your argument, which is unusual for you, is any evidence for your premise. That is, is it the case that the heavy prey on the light?

panopticon writes:

[Comment removed for supplying false email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Tom writes:

Interesting idea, but it mistakes the reality of prison rape for a vulgar stereotype. The majority of prison rape does not fall into the “bigger guy grabbing smaller guy in the shower” scenario that we’ve seen a hundred times in movies. In reality, doing that carries too many risks for the perpetrator, even if they are twice the size of their target (being ratted on, being caught, getting stabbed in revenge, etc.) Less overt methods of coercion and pressure are more often used, and the primary targets tend to be transgender inmates or those with mental health issues, IOW, people who are vulnerable in ways other than based on physical size.

Here’s a link to a report that was done for the California Department of Corrections and Rehabilitation on the problem recently:

And for a more informal insider account, look here:

kentucky writes:

I think this is a great idea. The conditions in US prisons are really shameful, in my opinion US prisons and the War on Drugs make up the two most damning examples of the US government's abuse of its citizens.

anon writes:

[Comment removed for supplying false email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Zdeno writes:

Surely we can do better than just segregating prisoners by weight.

Our goal is to reduce prison violence, especially for those prisoners who would rather serve their time than claw their way up the prison dominance hierarchy. Here's an easy way to do that:

1) Segregate prisoners by how dangerous they are. At first, this would be based on the type of offense they committed.

2) Prisoners move up or down in the hierarchy depending on their behaviour. Keep your hands to yourself, obey the rules, and you'll move quickly into a prison full of similarly well-behaved felons. Infractions, especially those involving violence, are punished by demotions through the ranks of prison wings filled with progressively more brutal prisoners.

3) Profit! Specifically, the prisoners who have the capacity for self-control can live in peace, while the few that don't collect in the inner circle of hell that is the lower rung of such a system, in which they are free to beat and rape each other at will. A year of good behaviour gets you a second chance, and there are no third chances.

Bottom line: All prisoners have an incentive to live and let live. As a result, most do. The few unreformables that don't, good riddance.

As for the political feasibility of such a solution, I think agnostic is perceptive. I recall plenty of liberal snark regarding the fates of Skilling, Lay et al. post-Enron.



Pensatulla writes:

One problem with making prisons more like monasteries is that prison inmates aren't monks.

The other problem of making prisons meritocracies is that if it's a succesful merit-based system then the convicts may prefer that system to the outside world.

The better solution is to use hard labor with a purpose- such as cleaning up brown fields or garbage dumps. This gives inmates something to do with their time, there is the bonding effect of shared suffering that cuts across racial lines, and they would be learning job skills.

Lars Petrus writes:

I expect the gender separation is there more to prevent consensual sex than rape.

Also, avoiding prisoner procreation is reason enough.

David writes:


I'm not sure if you are being facetious or simply not very familiar with many state's correctional systems. But, the system you described is essentially how California and many other states explicitly function when allocating inmates across facilities. So to answer your questions. First, it is politically feasible. Second, it doesn't seem to reduce violence and misconduct significantly (given that the original post and most comments seem to at least implicitly agree that the status quo is not satisfactory.

eccdogg writes:


Do you think that difference between blacks and whites really exist in size/muscularity?

Seems to me the really big difference is between fast twitch and slow twitch muscles which leads to a difference in quickness/explosiveness not outright strength.

A couple of anecdotal examples.

1) The ESPN Worlds Strongest Man competition. It is almost all Norwegians and other people of northern european decent.

2)Mixed martial arts competitions do not seem to have any racial favorite.

3)Football offensive lines pretty much match the percentages of whites and blacks in the US.

4)Collegiate wrestling also pretty much matches the racial make up of the US.

5)Other strength sports, shot put, weight lifting, discus don't seem to be dominated by people of african decent.

phil writes:

Actually your fast twitch fibers provide the most strength and explosiveness, both vital for your strong man comps. Who dominates at a particular sport has much more to do with who is participating in it in the most numbers. The idea of only big black guys preying on little white guys is not reality.

Rico writes:

What a simplistic and ignorant view. prison hierarchy is dominated by gang affilliation, not weight class.

bill writes:

Separating prisoners by weight is one of the silliest ideas I have ever heard. Is this really the best we have?

Aside from ignoring the numerous real-world problems of implementing this system, this ignores the main issue of us simply having too many prisoners.

Rather than house prisoners better, we need to get them out of prison. Many states now spend more on corrections than they do on higher education. We spend more for prisoners than students. You want to solve the prison problem? get people out of prison, on parole, probation, wearing anklets, etc.

What's more, this really weird idea comes from an educated, cerebral, policy-wonk, white-collar guy. I'm sorry, but who else would think of something as dumb as using boxing regulation to solve a social problem?

Dogbert Mugabe writes:

1)This doesn't address what may be the largest factor: race.
2)Weight is easily gamed.
3)A better idea would be to hand out cheap webcams and secrete them everywhere. They're cheap and the video would be watched by outside voyeurs 24/7. It should be the right of prisoners everywhere to be constantly monitored.

Hootie writes:

I'm genuinely surprised by how ignorant a lot of these comments are. "Prisoners will bulk up, and then we'll have super prisoners!"

I'm not sure how many have actually been in prison here, but that already goes on. Prisoners exercise in order to deter rape. Hell, even Bernie Madoff was getting in shape before he went to prison to avoid rape.

Prison rape is one of the most pressing issues of our time and under a nation of laws, is the indirect threat levied against everyone who doesn't obey the law and is threatened with jail time. That's something disturbing.

Anything that can be done to change the system should be done, because it's broken. Badly broken. As-bad-as-it-can-be broken.

Bryan, thanks for looking at this pragmatically. It'd be worth an experiment.

FatMan writes:

What about fat out-of-shape prisoners whose weight is all fat and no muscle. If they were left with the other heavy-weights they would be heavily victimized.

Comments for this entry have been closed
Return to top