Bryan Caplan  

Incentives Matter: Prison Rape Edition

PRINT
Random Financial Commentary... The Best Question I Heard Yest...
If I had to serve time, I'd prefer to be sent to a for-profit prison factory.  The main reason: I think my employer would make an effort to protect me from severe abuse.  After all, a victimized indentured servant is an unproductive indentured servant!

The most notorious form of prison abuse, of course, is rape.  How severe is this risk?  According to the recently released National Prison Rape Elimination Commission Report, it's pretty severe, almost 5% per year:
BJS conducted the first wave of surveys in 2007 in a random sample of 146 State and Federal prisons and 282 local jails. A total of 63,817 incarcerated individuals completed surveys, providing the most comprehensive snapshot of sexual abuse in prisons and jails to date. Four-and-a-half percent of prisoners surveyed reported experiencing sexual abuse one or more times during the 12 months preceding the survey or over their term of incarceration if they had been confined in that facility for less than 12 months. Extrapolated to the national prison population, an estimated 60,500 State and Federal prisoners were sexually abused during that 12-month period.
You might think that rape is an inevitable feature of communities of convicts.  But rates vary widely - too widely, I think, for inmate composition to explain:
Although sexual abuse of prisoners is widespread, rates vary across facilities. For example, 10 facilities had comparatively high rates, between 9.3 and 15.7 percent, whereas in six of the facilities no one reported abuse during that time period. [emphasis mine]
As economists and non-economists alike would predict, smaller, weaker prisoners are victimized more:
Youth, small stature, and lack of experience in correctional facilities appear to increase the risk of sexual abuse by other prisoners. So does having a mental disability or serious mental illness. Research on sexual abuse in correctional facilities consistently documents the vulnerability of men and women with non-heterosexual orientations and transgender individuals.
Girls are disproportionately represented among sexual abuse victims. According to data collected by BJS in 2005-2006, 36 percent of all victims in substantiated incidents of sexual violence were female, even though girls represented only 15 percent of confined youth in 2006.
The most striking result, though, is one that only fans of Gary Becker and/or Lord Acton would have expected:
More prisoners reported abuse by staff than abuse by other prisoners: 2.9 percent of respondents compared with about 2 percent.
The lesson: You might think that no one would be more inclined to sexual abuse than criminals in unisex confinement.  But if you give authority to relatively normal people who can leave the prison anytime they like, they're worse.


Comments and Sharing





COMMENTS (12 to date)
JH writes:

"So does having a mental disability or serious mental illness."

Wow. I find it all pretty disturbing, but this is just...wow.

ZC writes:

I agree with your premise of for profit prison factories and the potential for better treatment. Although given the large number of incarcerated people in the US, with no end in sight, treatment may little improve as the 'company' may view convicts as expendable -- with conditions akin to a gulag.

With regards to sexual abuse in prison, I take issue with the use to self-reported survey results to document the issue. How accurate can these possibly be? I would guess not very. Maybe those with mental disability and mental illness (I would argue that all criminals have some form of mental illness) just claim to be abused more frequently (no doubt they may have other hallucinations or feelings of persecution). And, how many prisoners claim abuse, hoping the ACLU or some attorney will come along to plead their case and garner them money or freedom.

Since this is about incentives, it seems the prisoners would have no incentive not to lie about such treatment, and plenty of potential incentive to lie.

Blackadder writes:

I guess it depends on what kind of for-profit prison we're talking about. I wouldn't want to be in a private prison which got its money from the state (i.e. the state pays the prison a set amount per prisoner to house and feed them, etc.) In that case the incentives are all towards letting conditions in the prison deteriorate, a la what happens to apartments under rent control. Bryan seems to be envisioning that the prison could hire him out to some firm and make his profits from his paycheck, which may or may not work, but isn't how private prisons are set up now.

If you were going to have private prisons, I would want prisons paid a lump sum per prisoner (based on the sentence) with the understanding that if the prisoner gets out and re-offends the original prison that housed him has to pay all subsequent incarceration fees. I bet you could cut the recidivism rate by at least 90%.

Grant Gould writes:

Blackadder:

A tempting line of reasoning, but consider that the prison might conclude that crippled/dead prisoners are much less likely to reoffend...

Brian Artiaco writes:

I'm not sure why there is an assumption that prison violence of any kind would be reduced in a privately operated prison system.

In fact, a privately owned prison stands to benefit financially from prison violence. More crimes commited results in longer sentences. Longer sentences means more prisoners. More prisoners equals profit.

There is a reason why most mandatory minimum sentence laws are lobbied for by prison worker's unions. They are, after all, in the business of incarceration, not in the business of reform.

Also, while I don't know any data on rates of violence in private vs public prisons, I do know there have been a number of cases of prison riots and deaths, that occurred in privately operated prisons due to overpopulation and understaffing.

Blackadder writes:

A tempting line of reasoning, but consider that the prison might conclude that crippled/dead prisoners are much less likely to reoffend...

It's kind of hard to hide if all your inmates end up crippled or dead.

John Thacker writes:
I'm not sure why there is an assumption that prison violence of any kind would be reduced in a privately operated prison system. ... There is a reason why most mandatory minimum sentence laws are lobbied for by prison worker's unions.

Well, one crass assumption would be that both Democrats and Republicans suck up to the prison workers' unions in states like California, but Democrats wouldn't suck up to private prisons, and instead would regulate them fiercely.

Monte writes:

There doesn't seem to be a huge disparity between these numbers and national rape statistics for civilians. And I agree with ZC that we should question the reliability of self-reported incidents (although I would counter that rape is just as likely to be under-reported by prisoners who wish to avoid being stigmatized as those who hope to gain money or freedom by filing false claims of sexual abuse).

From what I've read, sexual misconduct at private correctional facilities happens with as much regularity as it does within federal or state-run facilities (albeit incidents of this nature are primarily isolated to female staff/inmates or juveniles). According to Critical Resistance (a political interest group that seeks to abolish the Prison Industrial Complex (PIC)), the PIC "depends upon the oppressive systems of racism, classism, sexism, and homophobia. It includes human rights violations, the death penalty, industry and labor issues, policing, courts, media, community powerlessness, the imprisonment of political prisoners, and the elimination of dissent", so I wouldn't be too anxious to climb into a Corrections Corporation of America or Wakenhut jumpsuit.

NYCO writes:

I think that people need to do a better job checking their sources. While the is always a possibility of sexual abuse in the corrections environment, statistically, you have a better chances getting raped in an ivy school then on the blocks. The idea that a for profit organization would "do a better job protecting them...." just goes showcases the ignorance of the study. With liability cases at a all time high... there isn't a insurance company out there to back it. As a Corrections Officer for NYS, I take great offense to your have ass study. Please try to work a block before you talk about one. The most notorious form of prison abuse is not rape.... its stabbings.

Levi writes:
While the is always a possibility of sexual abuse in the corrections environment, statistically, you have a better chances getting raped in an ivy school then on the blocks.

Brown University, 8,020 students, 3 forcible sex offenses

Columbia University, 24,820, 9 forcible sex offenses

Cornell University, 19,800 students, 3 forcible sex offenses

Dartmouth College, 4,147 students, 6 forcible sex offenses

Harvard University, 19,140 students, ~17 forcible sex offenses

Princeton University, 4,918 students. ~12 forcible sex offenses

University of Pennsylvania, 19,816 students, 12 forcible sex offenses

Yale University, 11 students, 9 forcible sex offenses

I don't feel like calculating out the chances, but a browse over the data leads me to naively conclude that you're wrong.

Rivka writes:

Actually, given staffing levels and other choices at private prisons, violence tends to be higher than in public prisons (though neither group is safe). See Sharon Dolovich, State Punishment and Private Prisons, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=890326, especially pp. 502-08. Even if prisons make prisoners labor, the guards--who, as you note, are major perpetrators of sexual abuse--have no particular incentive to make sure that large numbers of widgets are produced, since that's not determinative of their pay. (And that's setting aside the extent to which rape actually prevents people from working, another attenuating factor in your hypothesis. If letting some powerful people abuse less powerful people makes the prison easier to run in the sense of keeping some people cowed and others satisfied, plenty of guards will tolerate it even if they don't themselves rape prisoners.) If you'd prefer to be in some hypothetical profit-seeking prison, that's one thing, though I still think you're wrong, but the prisons we have don't work that way.

Patrick McCann writes:

This study seems absurd, because taking a random sample of jails is a lot different than a random sample of prisoners. That is just poor data quality; try reweighting the data for prison size and you might get somewhere, because I have a feeling this is biased towards small women's only centers, where you would expect staff abuse to be higher than for the typical prison inmate.

Comments for this entry have been closed
Return to top