I am one of the lucky ones who never went back.

I got out over 20 years ago, and I was determined to do whatever it took to successfully complete my parole and put all that crap behind me.

I still have scars. I'm 47 now. Dentures are in my near future because of the fighting from my prison days.

I had nightmares about the fighting and killing (saw two guys get killed mere feet away from me) for 15 years after getting out.

As time goes by the memories are less and less, but they never go away.

It changes you permanently.

Okay, this are good numbers. However, I would also draw your attention to Table 7. Only about half of the prisoners released who were re-arrested were re-arrested for violent crimes. So, the question I ask is, are these released prisoners actually getting the support they need to succeed after release, are are they so weighed down by the stigma of prison that they can't reasonably obtain a non-criminal life?

Absolutely the latter. There simply isn't any support for the vast majority of people.

Yes, you can argue that some communities have support groups or employment opportunities specific for ex-prisoners, but this isn't available universally, isn't well funded, and most people won't turn to help even if it's offered. That last part is true of free people as well: think of how many people in need refuse to reach out.

The stigma is real and crime is only ever a few inches away from most people.

I recall there was a study a few years back that claimed real recidivism rates were actually much lower than commonly stated due to a selection bias effect.

Here are the slides: https://www.bgcheckinfo.org/sites/default/files/public/5thMt...

That PDF you gave makes such an excellent argument that I thought it would be helpful to quote the best part:

"When criminologists examine prison populations, they see a concentration of repeat offenders, most of whom will recidivate. Low-risk offenders enter and exit prison once, so in any survey of a prison stock, low-risk offenders are underrepresented as a proportion of the offender population.

An analogy is helpful: A Mall Exit Survey

Survey researchers sometimes use mall exit surveys to estimate shoppers’ purchasing habits.

Suppose that:

•25% of mall visitors go to the mall once per day.

•25% go once per week.

•25% go once per month.

•25% go once per year.

A naïve one week exit survey of visitors exiting the mall will find that:

•85% of mall visitors go to the mall once per day.

•12% go once per week.

•3% go once per month.

•Fewer than 1% go once per year.

The problem is that high-rate mall visitors churn, leaving an impression that most shoppers are frequent visitors."

Quoted from "Criminal Recidivism: Most Incarcerated Offenders do not Return to Prison", by William Rhodes, from the link in parent

Interesting.

But the slides still find evidence of recidivism using a less biased offender-based analysis. Moreover they find that the people are still more likely to return to prison the longer they have stayed in prison. The numbers are lower, but are definitely there. Ancestor’s claim about prisons creating criminals still stands.

Then there is the issue of low level drug offenders returning to prison for violent crimes in the next round. It would be interesting to see the numbers on that.

what's the counterfactual? How can you tell the difference between "criminals are undeterred by prison and will commit offenses no matter what" and "prison hardens criminals and does not help with their reinsertion"?

You can test that by looking at data from countries with less abusive prison systems. IIRC, they typically have much lower recidivism rates.

That 100% ignores cultural, education, diversity, and tons of other issues.

Yea, I bet recidivism is low in Norway, who also has a very small chapter of MS13 or whatever gang you like, coincidentally.

I disagree you can look to some other country and make a fair comparison. The only way to do it would be to look at rates before and after some policy change.

It sounds like you’re claiming that the US just has more criminals per capita.

There’s a reason we have so many violent gangs in the US. We don’t take care of people the way other countries do, and we take advantage of every marginalized population we can.

Except that you can't do that either because someone acting in bad faith can just say that there were other confounders, because there will always be other confounders, since we can't make a proper control group out of a society. Immigration/emigration, other policy changes on county/state/federal level, delayed effects of 30-year old policies (such as Roe vs. Wade or banning lead in gasoline), etc. There's no such thing as fair comparisons in the social sciences; we can only and have to act on imperfect information.

What has diversity got to do with anything? Australia is a country of immigrants with people for all over the world living with dramatically less crime per capital than the US. Are you claiming that certain races or cultures are somehow inherently more criminal? If yes please provide evidence.