Does Juvenile Incarceration Act as a Deterrent?

Is putting a juvenile offender in jail a useful deterrent or one big step in the wrong direction? That’s the question asked in a new working paper (abstract; PDF) by Anna Aizer and Joseph J. Doyle. It appears that the deterrence argument doesn’t hold much water:

Over 130,000 juveniles are detained in the U.S. each year with 70,000 in detention on any given day, yet little is known whether such a penalty deters future crime or interrupts social and human capital formation in a way that increases the likelihood of later criminal behavior. This paper uses the incarceration tendency of randomly-assigned judges as an instrumental variable to estimate causal effects of juvenile incarceration on high school completion and adult recidivism. Estimates based on over 35,000 juvenile offenders over a ten-year period from a large urban county in the US suggest that juvenile incarceration results in large decreases in the likelihood of high school completion and large increases in the likelihood of adult incarceration. These results are in stark contrast to the small effects typically found for adult incarceration, but consistent with larger impacts of policies aimed at adolescents.


I dislike this sort of blanket statement based on averages.

I could agree that non-violent or first-time young offenders should not be incarcerated, but when it comes to repeat and violent offenders the need to protect the community outweighs the desire to prevent recidivism.


"the need to protect the community outweighs the desire to prevent recidivism."

I don't see how those two things can be separated.

RJ Roy

I agree. Preventing recidivism directly protects the community. If a person commits a crime, and then manages to successfully never commit another crime again, is the community not protected due to the actions of the rehabilitation system?

Even rehabilitating a person who commits a violent and/or heinious act protects the community, because by definition they won't commit the act again.

Gray Turner

I watch lots of juvenile delinquency cases while waiting on my own cases to be called. Therefore I see lots of juveniles be placed in detention. I think a study like this is very difficult because the question: "why did the juvenile go to detention" is convoluted and not easily answered by just looking at the data. So for example, a juvenile is caught shoplifting. He is not arrested but given an intake date. In my country - this isn't even a court date. The juvenile is given an opportunity to have non-court consequences such as community service and an essay. A good percentate of cases are disposed of in this way, often called diversion or pre-trial probation. These might not show up in statistics because there was never a "conviction" ( or as we call it delinquency adjudication) Each county might call it something different and handle it different. However, if between the time of the original incident and completion of their program, the get a new charge or drop out of the program - they get a new charge - then they have two sets of charges.

Lets take the same juvenile. He may have two shoplifting charges now. He has already been given an opportunity to avoid court but didn't finish. However, most of these don't go to detention. They enter a plea and are given a trial date. Many are put on pre-trial probation which would include rules such as not skipping school, a curfew, not doing drugs and the like. Again, if they break these rules, they can get detained. Once found guilty, or adjudicated, they are on true probation.

So why do they go to detention? Many times not because of the initial offense - but because they fail to follow probation. One of the easiest ways to get your probation revoked and subsequently detained - is to get suspended from school or stop going to school. Simlarly, drug use will often end up in detention.

Say we have a juvenile who has either already had diversion or failed diversion - and then was placed on probation. He curses his teacher out and is suspended for 3 days, or expelled if it was a repeated issue. His probation officer has him come in and drug tested and he is positive for THS so he is detained. So why did this kid go to detention ? For shoplifting ? Yes. For getting suspended from school / not attending school ? Yes For using drugs - ? Many of these delinquents have no intention of completing school and if they do go don't engage. Yes He might stay in detention a weekend - he might stay 90 days - or he might go off to an indefinite committment in the Division of Youth Services at this point. In essence --- he is going to detention for failing to meaningfully participate in school and disrupting the school process. Even while in detention or youth services, they go to class every day.

Many other delinquents do want to finish school. A juvenile on probation - who goes to class every day, who does not get suspended, is making decent progress in school, is much less likely to be sent to detention even if using drugs. A judge is very reluctant to incarcerate a student who is trying.

I have no doubt there is a strong correlation between juvenile incarceration and failure to complete high school. I'm not convinced it is a causal relationship though.


Enter your name...

Given the potential for harm and the desire for good data, I wonder why we don't have individual judges volunteering for proper randomized trials. I see it working like this:

The judge disposes of all 'obvious' cases the old-fashioned way: violent sociopaths are given serious sentences with no hope of alternative. But for the others, he or she determines the sentence as usual, and if (and only if) it's an incarceration sentence, the kid gets a random chance of having that sentence converted to a reasonably equivalent non-incarceration sentence.

Then you compare the outcomes for the ones who were sentenced to incarceration but were randomly given an alternate non-incarceration sentence against the ones who were sentenced to incarceration and served the original sentence.

This approach should be ethical (incarceration is currently believed to produce some harms, so giving some kids an opportunity to escape a risk of harm is ethical), receive very few objections, and should give you the ability to make straightforward comparisons between equal groups due to the random design.



I'm curious which "large urban county" they studied. I trained as a child psychiatrist in a very progressive state where the juvenile justice system is very much focused on trying to address the root causes of why the juveniles ended up in detention in the first place.

Frequently, teenagers who end up committing offenses come from broken homes where they were subject to severe abuse and/or neglect. Once in the juvenile justice system (in that state at least), they have much more access to individual therapy, drug rehab, and educational support than they would have otherwise, and there's evidence to suggest that approaches based on therapeutic interventions, rather than punishment/deterrence, can be effective in preventing recidivism.

So I do think what is done in a particular juvenile justice system matters, especially in a study like this that just looks at one geographic area.

- Psycritic


Alan Gunn

I was once told of a judge who locked up juvenile first offenders for a day, tops. His reasoning was that if you left a kid in detention for a longer time he'd figure "I can take this," and there goes deterrence. A day of unpleasantness, though, would make him reluctant to do things that would ge3t him put back there. Probably not the best answer for murderers, but seems sensible enough to be worth a try for lesser offenses.


This doesn't surprise me. Does it surprise anyone really?


It is a deterrent if the juvenile has been incarcerated once because they realize what incarceration entails, especially that it means they are not able to see their families quite as frequently. However, some juveniles are more productive around the board, academically and socially, when they are in the structured environment of juvenile detention centers.

Brooke Barnes

This topic fascinates me in many ways. I ponder the many affects juvenile delinquents have on the economy and society, both in the macro sense and the micro sense.

I recently wrote a paper on whether therapeutic juvenile jurisprudence was the solution to the high recidivism rates among juveniles who are placed in juvenile detention centers or adult jails/prisons.

Juvenile therapeutic jurisprudence involves many variables such as the school's involvement, parental involvement (voluntary or involuntary), the community, chemical and emotional treatment, and, of course, the individual. Every aspect of a youth's life is impacted and involved. This theory involves looking at the youth as a whole, not individually.

I analyzed many different case studies which included youth's placed in juvenile detention, youth's placed in adult jail/prison, adult's placed in similar therapeutic jurisprudence programs, and juvenile's placed in therapeutic juvenile programs.

Cost is another influential factor- juveniles that frequently revisit the criminal justice system throughout their youth and well into their adult lives is costly. Therapeutic adult programs have shown to initially be expensive but in the long run drastically cheaper. There is not enough data to know for certain whether juvenile therapeutic system is more cost effective than juvenile detention or adult jail/prison. Although the small amount of data available on the cost juvenile therapeutic systems and the recidivism rate is positive in that it shows a correlation on cost effective measures as with adult therapeutic programs.

My conclusion- juvenile detention centers are not working and neither does the scare tactic of placing youth's in adult jail/prisons seem to work. So, why not try something different? There is very little known on juvenile therapeutic jurisprudence but quite a lot is known on the success of adult therapeutic programs. So.... why not try patching the holes in a ship than continue bailing water in a sinking ship?


P Scott

Perhaps they should lock up the parents or fine them instead of punishing the kid. Post the parent's names and/or pictures on local media. If one doesn't know where their kid is after 9 pm, who they are with and what they are doing, there's a problem. We told our kids they needn't call us for bail if they landed in jail. Seemed to be a good deterrent at the time.