« FIRST STEP Act sentence reduction used to address "FBI invented" conspiracy in which feds were "the real lead conspirator" | Main | Big new report examines developing "hybrid justice systems" for offenders between 18 and 25 »

July 31, 2023

"Conviction, Incarceration, and Recidivism: Understanding the Revolving Door"

The title of this post is the title of this new paper that looks to provide a notable (and lengthy) empirical account of contributions to recidivism.  The piece was recently posted to SSRN and is authored by John Eric Humphries, Aurelie Ouss, Kamelia Stavreva, Megan T. Stevenson and Winnie van Dijk.  Here is its abstract:

We study the effects of conviction and incarceration on recidivism using quasi-random judge assignment.  We extend the typical binary-treatment framework to a setting with multiple treatments, and outline a set of assumptions under which standard 2SLS regressions recover causal and margin-specific treatment effects.  Under these assumptions, 2SLS regressions applied to data on felony cases in Virginia imply that conviction leads to a large and long-lasting increase in recidivism relative to dismissal, consistent with a criminogenic effect of a criminal record.  In contrast, incarceration reduces recidivism, but only in the short run.  The assumptions we outline could be considered restrictive in the random judge framework, ruling out some reasonable models of judge decision-making.  Indeed, a key assumption is empirically rejected in our data.  Nevertheless, after deriving an expression for the resulting asymptotic bias, we argue that the failure of this assumption is unlikely to overturn our qualitative conclusions.  Finally, we propose and implement alternative identification strategies.  Consistent with our characterization of the bias, these analyses yield estimates qualitatively similar to those based on the 2SLS estimates. Taken together, our results suggest that conviction is an important and potentially overlooked driver of recidivism, while incarceration mainly has shorter-term incapacitation effects.

July 31, 2023 at 03:40 PM | Permalink


"Taken together, our results suggest that conviction is an important and potentially overlooked driver of recidivism, while incarceration mainly has shorter-term incapacitation effects."

It's not the conviction itself, but the defendant's attitude and behavior that brought about the conviction, that drive recidivism.

But yes, incarceration has incapacitating effects for as long as it lasts. What this means, obviously, is that the longer the incarceration, the more incapacitation you're going to get.

Posted by: Bill Otis | Aug 1, 2023 7:36:53 AM

Bill - this study suggests that innovative strategies that rely on alternatives to the traditional criminal justice process (restorative justice, alternative programs) are much better at reducing recidivism rates than the traditional conviction and jail processes. Brett Miler

Posted by: Brett Miler | Aug 1, 2023 8:44:22 AM

Brett Miller --

Since you seem to have missed the first part of my comment, I'll repeat it:

It's not the conviction itself, but the defendant's attitude and behavior that brought about the conviction, that drive recidivism.

Still, let's cut to the chase. The thing that will reduce recidivism rates to zero is for ex-convicts to decide to live a normal, law-abiding life just like the rest of us have to. It's not what the state does, it's what THEY do.

Posted by: Bill Otis | Aug 1, 2023 9:00:25 AM

Bill - why are you so insistent on personal responsibility? A more effective and holistic justice system can be more effective than a more punitive one. Brett Miler

Posted by: Brett Miler | Aug 1, 2023 12:01:39 PM

Brett Miller --

"Bill - why are you so insistent on personal responsibility?"

Because my parents' demanding personal responsibility from me put me on the road to Stanford Law, DOJ, and the White House, among other neat stuff.

People who don't believe in personal responsibility wound up as my defendants, basically because they thought stealing is OK. No yuuuuuuuchy personal responsibility, dontcha know.

To demand personal responsibility is not to lack compassion. To the exact contrary, it is among the most loving things an authority figure can do.

Posted by: Bill Otis | Aug 1, 2023 3:24:11 PM

An insistence on personal responsibility ignores the needed changes that society should make to be more compassionate to its most vulnerable and at-risk populations. Brett Miler

Posted by: Brett Miler | Aug 1, 2023 6:32:55 PM


I've been arguing these points with our friend Bill for some time now. In past discussions, I advocated for more of a focus upon rehabilitation (vs longer punishments), including the need for experimentation in respect to alternative approaches (e.g., restorative justice programs; 'therapeutic community' model, either in a prison or outside of a prison setting, etc.).

While in the past, Bill has admitted that he has little to no expertise in the field of rehabilitation (which serves to reduce recidivism), I am not aware of any new efforts on his part to study and educate himself in this field. I hope that I am wrong, but I doubt it. And if I am wrong, I have no doubt Bill will correct me.

To those whose only tool is a hammer, everything begins to appear as a nail. "Bill's hammer" is "personal responsibility" and every defendant, the nail.

What else is interesting is that Bill's sweeping generalization and belief that all lawbreakers who recidivate do not take "personal responsibility" seems to be based largely upon his own personal upbringing and education. Again, his basic assumptions are rooted not in extensive research, education, training, and/or experience, but rather in his childhood training, as good as it was.

Prosecuting criminal cases, I would argue, does not sufficiently educate a prosecutor as to the root causes of recidivism. In my experience, the AUSA's I knew who prosecuted criminal cases did everything they could to keep from interacting with defendants. Instead, they relied upon the law enforcement agents as to their take on a defendant (a good number of federal agents were no better and often worse than local police. And many federal agents came from local police agencies).

In other words, my experience on the streets, in rehabilitation centers, as well as my experience as a criminal defense investigator for almost 30 years, and college degrees in this field, has provided me a far greater understanding of these issues than most AUSAs, including Bill Otis.

Posted by: SG | Aug 2, 2023 5:26:35 AM

Only in my dreams could I have imagined something more happy-making than to be condemned for taking the side of personal responsibility.

Thank you SG and Brett!

Posted by: Bill Otis | Aug 2, 2023 10:23:30 AM

You're welcome Bill

Posted by: Brett Miler | Aug 2, 2023 11:41:33 AM

How many people who spout this nonsense have been the victim of an actual violent crime? I can tell you from personal experience that the people committing them think that they have the right to do it. How do you deal with that level of malevolence? These are bad people. They have to be fundamentally changed.

Posted by: federalist | Aug 2, 2023 11:46:28 AM


You attest, as your diagnosis, that these folks need to be "fundamentally changed"...and your method of accomplishing this is ...what? Longer prison sentences? Harsher conditions? More stringent parole/probation conditions? Attending once a week 'sessions' with a counselor?

Posted by: SG | Aug 3, 2023 10:43:57 PM


Same question that I posed to Federalist, I pose to you. Your formula for reducing recidivism is what???

Posted by: SG | Aug 3, 2023 10:45:10 PM

SG --

"Same question that I posed to Federalist, I pose to you. Your formula for reducing recidivism is what???"

The same formula that so massively reduced crime (much of it recidivist crime) for an entire generation, 1991 - 2014 (when crime rates fell by nearly half): More police, more proactive policing, more incarceration, more law-driven sentencing.

It worked and worked wonderfully.

We tried your answer -- more social programs and being reluctant to hold criminals personally responsible -- during the Sixties and Seventies and it was a disastrous failure.

The choice is easy. Success or failure. The country has had a lot of experience with both. Which do you prefer?

Posted by: Bill Otis | Aug 6, 2023 2:42:37 PM

Bill - I don't necessarily think that trying to deal with criminal behavior without so much prison will result in "failure" just as I don't necessarily think that so much prison use is a "success". A truly successful crime policy in my view is one in which we can both reduce unnecessary prison use and enhance safety at the same time and prosecutors who attempt to do this should be applauded not scorned. Brett Miler

Posted by: Brett Miler | Aug 6, 2023 8:58:28 PM


First, the "social programs" for which I may advocate has never been HONESTLY undertaken by the "powers that be" (i.e., those who operate the criminal-justice system,etc.) These folks have regularly subverted and undremined programs intended to reduce recidivism, both in prison and after prison.

The subversion takes the form of unnecessary and questionable rules, regulations and conditions, coupled with heightened penalties for any perceived transgression.
Example: 'To qualify for Section 8 public housing a felony (of an offender seeking to rent) needs to be at least 5 years old, and it cannot be a disqualifying felony. Disqualifying felonies are typically related to violent crimes, drug trafficking, and sex crimes that would require the felon to register as a sex offender'. And should any ex-felon violate even a minor rule, their Section 8 housing 'privelege' would then be in jeopardy, and they and their family would then be subject to eviction. What is the sense in this?

Lack of housing and social destabilization is a major factor leading to HIGHER RECIDIVISM RATES. Housing and other factors such as 'family stability and support', as well as meaningful employment, education, network of friends and support, are all linked together.

Bill's solution, as I understand it, is that RECIDIVISM will be reduced by: more police, more prisons, and longer sentences, (this, of course, to be followed by even stricter parole and probation conditions,etc.). Is this your final answer?? Want to maybe call a friend first (federalist or Tarls)?

Bill..come on...you're a lot smarter than this... slamming your foot harder onto the gas pedal when you've already driven into a brick wall is hardly a viable solution to RECIDIVISM.

As to my suggestions to reduce the rates of recidivism, I would offer the following: across-the-board shorter sentences (perphaps two to three years MAX for non-violent drug and non-contact sex offenses, and for most cases even less time); followed by mandatory drug rehabilitation for drug addicts (at least two years minimum in a qualified residential rehab that is NOT operated, overseen, or regulated by the govt.); far less onerous conditions of probation/supervised release (e.g., no dehumanizing ankle bracelets); increase efforts in job training and job placement assistance; family counseling; parenting counseling/classes; housing assistance (get rid of those unnecessary disqualifying Section 8 housing regulations); monetary 'stimulus' to attend college, educational pursuits leading to meaningful careers (GED completion already takes place in most prisons). There are more suggestions, but that's it for now.

And for any of those who go through such regimens and then RECIDIVATE (and there will be a healthy number at first who do recidivate with multiple dirty tests, or worse), they may then be returned to incarceration for a short periord of time, at the discretion of the court. The amount of time would be dependent upon the nature and severity of the offender's behaviors.

So Bill...a heavier and larger hammer, swung with even more force, and with more frequency, is not a particularly effective tool in reducing recidivism...as well-intended as it may be.

Posted by: SG | Aug 6, 2023 11:09:28 PM

SG --

Your response is most revealing for what it omits -- namely, any mention of the almost 50 years worth of evidence that your policies fail and mine work. Just ignoring 50 years of evidence does not make a particularly compelling argument.

Still, I want to make sure I'm not misunderstanding you. You are aware, aren't you, that in the era of more police, more aggressive policing, more incarceration and more law-driven sentences, crime rates fell by close to 50%? And you think that's a good thing, no?

Posted by: Bill Otis | Aug 7, 2023 8:11:45 AM

SG --

One more thing. You want a big government, nanny-state regimen to take care of people as if they were little babies. I want adults -- you, me and everyone else -- to be responsible for their own life and behavior and not constantly be beggaring to Give Me More Stuff.

You want stuff? Fine, me too. Go out and earn it with an honest job.

Posted by: Bill Otis | Aug 7, 2023 8:15:49 AM

Bill - if a big government, nanny state regimen can keep some people from committing crimes, then the victims of those said crimes would probably consider it worth the cost. Brett Miler

Posted by: Brett Miler | Aug 7, 2023 4:53:46 PM


The issue has been "how to reduce recidivism". I see nothing from what you have proffered that would address this issue. Instead, you propose nothing new, and in fact, you do not mention recidivism at all.

Did you not read the original question I had posed?? I had asked you: "Your formula for reducing recidivism is what?" You seem to be avoiding answering this succinctly, clearly and directly, as I have done.

Were all wondering....why?

Posted by: SG | Aug 7, 2023 5:10:56 PM

SG --

As you can't help knowing, to talk about crime is to talk about recidivist crime. It's the same customers, again and again and again. You know this at least as well as I do. So the crime statistics I cite (and you don't question) speak volumes to recidivism.

You're just a big government, welfare state, hard Left liberal who thinks people who obey the law should pay the freight for those who refuse.

I decline. I'm responsible for my behavior and they're responsible for theirs. If instead of being responsible they want to steal, that's their choice. I was happy to have had a career giving them the consequences they earned.

Posted by: Bill Otis | Aug 7, 2023 9:09:33 PM


"To talk about crime, is to talk about recidivist crime".

Bill, what in the name of GOD are you trying to say? With this rather odd statement, I'm beginning to wonder about your sobriety.

But seriously, I believe that this is just another of your many attempts to avoid addressing the issue directly, which apparently is an area of which you know very, very little about (if you recall, you had clearly admitted to this in prior postings). So, I will not be chasing you down a rabbitt hole as much as you may have hoped I would.

Bill, if you are capable and willing, please address the issue directly: What would you do to reduce recidivism? What approaches would you take with offenders both in prison and after prison?

In my posting above, I reviewed a number of factors that all experts in this area have agreed directly impinge upon recidivism - all which you have summarily dismissed and ignored.

Posted by: SG | Aug 8, 2023 5:58:50 AM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB