This isn’t a study. It’s a press release. There are some citations down the bottom. Which one(s) do you believe support your premise?

“Findings show a decrease in self-reported offending over time by the most serious adolescent offenders, the relative inefficacy of longer juvenile incarcerations in decreasing recidivism, the effectiveness of community-based supervision as a component of aftercare for incarcerated youth, and the effectiveness of substance abuse treatment in reducing both substance use and offending by serious adolescent offenders.”

“Compared with noncustodial sanctions, incarceration appears to have a null or mildly criminogenic effect on future criminal behavior. This conclusion is not sufficiently firm to guide policy generally, though it casts doubt on claims that imprisonment has strong specific deterrent effects.”

Why do you think a 40 year old study on human behaviour is suddenly invalid? Have humans evolved so spectacularly over 40 years that we no longer respond to negative and positive stimuli the same way? That’s a while claim requiring some wild evidence.

People haven’t changed, of course, but society has changed a lot since the early 80s. Any data from such a study is limited to the 80s and earlier. This doesn’t entirely invalidate any findings (whatever they are, since we can’t read the actual article), but it means we should weigh more recent studies more highly.

It doesn’t explicitly show that locking ram-raiders up for longer reduces recidivism or propensity to offend. I provided it to refute the your claim that “the evidence shows harsher prison sentences does little to reduce crime rates.” It does do that.

Firstly, I will acknowledge that I should have been more careful with my statement. Thank you for keeping me intellectually honest. I should have said “studies do not support the finding that increasing sentences leads to reduced crime rates”, which is a slightly different statement.

That being said, the study does not support this. It shows that criminals who are imprisoned longer are less likely to re-offend, it does not show that increasing prison sentences reduces crime. These are not the same things. For example, longer prison sentences tend to be for violent crimes, but it is a known effect that violent crime offenders have lower recidivism following incarceration than non-violent offenders.

My argument is precisely that it is more complicated than one single thing. I am not arguing that severity is the only proven effective deterrence method. In fact, certainty of conviction is even more effective! You argued that severity does little to deter crime. I am providing evidence that it does. You can read the full study here if interested. From the conclusion:

The article does not support your assertions. In fact, your statement that ‘certainty of conviction is even more effective than fear of punishment’ is precisely what the article questions. Its conclusion is, to paraphrase, that criminals do not weigh risk of being caught and severity of punishment if they are caught differently, but rather they are all equally weighted, and that policy should not weigh one factor above the other. Crucially, it says nothing about the actual importance of these factors to criminals. Equal weighting can still be 0. In the end, all it says is that criminals are complicated, and that many things factor into their propensity to commit crime. I agree with this, but it does not support the idea that harsher punishments will reduce crime rates.

They found a correlation with severity and burglary only in this study. That does not mean a correlation for other crimes is absent. Remember: an absence of evidence is not evidence of absence. Even if the correlation were only present for burglary, I would be very happy with fewer burglaries.

Absence of evidence is exactly what I’m arguing. There is an absence of good evidence that increasing punishments will lower crime. There is concrete evidence, however, that post-incarceration interventions lower recidivism. Also, you cannot take part of the data and ignore the rest, which is exactly why they conclude that there is not enough of an effect to draw a conclusion based on their findings. In other words, their findings do not show a statistically significant effect that increasing punishments lowers crime rate.

I refer you to page 178. The rate of incarceration only began to meaningfully increase in the 80s. It is well established that criminal policies rarely immediately affect social change. This often takes many years or decades, which is what we see in the data. I grant that sociological studies on criminology are very difficult to attenuate noise, but we don’t and cannot have controlled studies. In this case the methodology looks good, and it’s extremely well cited. Indeed, this is a far higher level of rigour than I am used to seeing in criminological research. If you wish to throw out this study on the premise you argue, we may as well throw out the entire school of sociology and criminology.

There is a lot going on in this article, and it would far to long to analyze everything in-depth right now. So, instead, I will provide two more recent studies which contradict certain aspects of the findings here. I could do more but I need to work.

This 2019 study found that, contrary to the article being discussed, legalizing abortion had a large effect in reducing over all crime.

This report found only a 5% effect of increased incarceration in the drop in violent crime from 1990 to 2000, and after that no effect at all. Additional to this, the exact same drop in violent crime in Canada was observed over the same time period, with no increase in incarceration rate, casting doubt into the effect it played in a global trend.

The research is encouraging, and I hope we see more of it. However it is hardly conclusive, and it certainly doesn’t indicate we should ignore sentence severity as a necessary and critical component of deterrence.

That was not the only study. This one discusses the effect of in-prison physical and mental health to recidivism.. This review concludes that post-release programmes have a positive impact on recidivism in most western countries.. This Scottish review concludes that, while there are many factors, “Rehabilitative interventions with the strongest evidence base for reducing reconviction rates are cognitive-behavioural programmes which address criminogenic needs.”

Overall, there is not, in my opinion and the opinion of the Department of Corrections, the evidence that increasing the severity of punishments reduces crime, and that the greatest effect is addressing the causes of crime in the first place.