22nd September 2021
An article published in Crime and Justice by Petrich et al. in September 2021 entitled ‘Custodial Sanctions and Reoffending: A Meta-Analytic Review’ conducts a meta-analysis of various studies that seek to find a correlation between custodial sanctions and reoffending rates. The sample was primarily taken from studies in the United States (78.3%), but other countries were also included such as the UK (5.3%), Nordic Countries (4.6%), and The Netherlands (3.8%). The authors note that policy changes in the United States between the mid-1970s to the 1990s that increased the use and length of prison sentences were not based on empirical evidence. They also state that this research is more important than ever given the increasing scrutiny of racial disparities in the US criminal justice system, as well as the COVID-19 pandemic which has put the current prisoner population at increased risk.
Imprisonment: Deterrence or a Criminogenic Environment?
The article distinguishes between two different outlooks on prison; those who view it as a deterrence, and those who view it as a criminogenic environment.
There is a cost-benefit rationale behind the idea that prison acts as a deterrent to crime. This logic assumes that those who are considering committing a crime will weigh up the possible risks against the perceived benefits – and where there are stronger sanctions for committing the crime, they are less likely to go through with it. The authors put forward three holes in this argument – reoffending rates indicate that people will make the costlier decision repeatedly; the model does not factor in the probability of being caught into the cost-benefit analysis; and it has been proven that some people prefer shorter terms of imprisonment to longer community service orders, especially if they grow accustomed to the prison environment.
Many other scholars view imprisonment as a criminogenic environment that increases the risk of reoffending upon release, however. The article refers to studies that back up this theory – including studies that show people who are put in prison with others who have similar offending patterns will be more likely to reoffend (particularly for skills-based crimes such as theft). They also cite Agnew’s general strain theory, which argues that the removal of positive stimuli and the introduction of negative stimuli in the transition to prison increases offending risks. For example, they highlight the fact that “weakening social bonds with family” caused by terms of imprisonment can stabilise offending. Finally, they highlight research done on the “labelling effect” of those who have a criminal record, where those persons with a criminal record are less likely to find employment, for example.
IPRT’s research on Community Service in Ireland (available here) also discusses the negative impacts of short terms of imprisonment on family ties. It refers to studies that show rates of recidivism decline when people maintain family contact while in prison, and in comparing the impact of Community Service Orders to short terms of imprisonment, finds through qualitative interviews that even short terms of imprisonment have a negative effect on family relationships.
The article then moves on to consider the previous failings of other studies to reliably track the correlation between custodial sanctions and reoffending. They highlight the lack of heterogeneity in the methodological quality of previous studies, the discrepancies in the characteristics of sanctions assessed, and the different sociodemographic factors that were considered in the broad range of studies.
The review conducted by the authors in the study – described by them as the “first comprehensive meta-analytic review of the literature in over a decade,” allowed them to measure whether a slew of different research methods, sanctions considered, and sociodemographic factors created a different set of results.
The overall result of the study found that, no matter what variables were considered, there was always either a null or criminogenic effect of imprisonment on reoffending. This was true no matter what research design was employed. Through the disaggregation of the data, they also found some interesting results – such as that prison seemed to be more criminogenic for juveniles rather than adults, but that the criminogenic rate on offending was the same for both males and females in prison (although the study had few female subjects), and was also the same across different countries.
The authors state that based on past research and this meta-analysis, “the limited effects of custodial sanctions on reoffending should be viewed as a criminological fact.” They state that the highest-quality studies may reduce the effect of the criminogenic environment on reoffending, but none eliminate it entirely. There is therefore a consensus that custodial sanctions do not reduce reoffending.
In response to the idea that reforms to the criminal justice system will be costly, they state that some reforms merely require more professionalism instead of more funding – providing effective services costs no more than ineffective ones. Finally, they also state that regardless of cost, keeping the system as it is comes at a huge opportunity cost, as reoffending continues, and more crimes are committed in society.
Read the study in full here.