New research shows that people released from prison are significantly less likely to reoffend if they are subject to supervision and other licence requirements – especially first-time prisoners.
By contrast, those with five or more previous prison spells show little behavioural change.
The study, by the University of Strathclyde, found that supervision and licence conditions reduced reoffending by 15% in the first four weeks after release – when monitoring is most intensive – and by 5.5% in the three years afterwards, indicating an enduring effect.
Part of the decline stems from prison recalls, often for 14 days, issued at probation officers’ discretion when licence conditions are breached. These recalls can temporarily suppress offending because individuals are back in custody.
Behavioural change
The researchers emphasise, however, that this mechanism cannot explain longer-term reductions. Probation officers can only issue recalls in the initial months following release, meaning lower reoffending years later reflects genuine behavioural change.
Licence conditions – requiring good behaviour, a fixed residence and regular contact with a probation officer – now apply to all individuals released from prison in England and Wales.
Yet reoffending rates remain stubbornly high with recent Ministry of Justice figures showing that more than half of adults released from custodial sentences of less than one year are re-convicted within a year.
Dr Markus Gehrsitz, the lead researcher, said: “Our research suggests that supervision can change the trajectory of offenders who have had limited prior interaction with the justice system. On the other hand, it seems to have little effect on repeat offenders with several prison spells.”
The project, funded by Administrative Data Research (ADR) UK, allowed researchers at Strathclyde’s Fraser of Allander Institute and Department of Economics to use a new Ministry of Justice dataset to follow tens of thousands of de-identified offenders through the criminal justice system.
Intensive supervision
The study – which has yet to be published in a peer-reviewed journal – also found that supervision is more effective for those released from longer prison sentences. Offenders who served six to 12 months had markedly lower reoffending rates than those released after sentences of two months or less, largely due to longer and more intensive supervision.
Researcher Dr Sam Grant said: “Offenders released from long sentences have a longer licence period. That shows up as lower reoffending.”
The policy mainly reduced theft and other offences that typically do not carry a prison sentence. It also led to fewer violent crimes, but this effect dissipated over time: a “zero” effect could not be ruled out one year post-release. Effectiveness of supervision did not vary by gender, ethnicity or age.
Dr Gehrsitz added: “Imposing licence conditions and supervision makes a tangible difference. That said, recidivism remains very high among those released from short sentences – in our data, around two thirds of people reoffended within one year. Offender supervision mitigates the problem but is far from a complete solution.
Still, our results are encouraging. They suggest supervision can stop the ‘revolving door’ for at least some offenders and that investing resources in first-time offenders is likely to have the greatest impact.
The economists also found that the benefits of supervision –through reduced crime – outweigh its costs.
Dr Grant said: “Every additional pound spent on supervision avoids more than £2 in crime-related costs, including policing, court time, and victimisation.
“However, supervision and licence conditions for prison leavers are unlikely to address prison overcrowding. Because most of the crimes prevented are ones that typically would not have resulted in imprisonment, and because recalls can increase prison numbers, the net effect on capacity is limited.”
The researchers drew on anonymised administrative data from the Ministry of Justice’s Data First Cross-Justice Datasets, which track tens of thousands of individuals through prisons, courts, and probation.
Future research
The project was supported by ADR UK, an Economic and Social Research Council (ESRC) investment, and the Office for National Statistics, which ensures safe and secure research access to government data.
A legacy of the project is that the code used to generate the dataset will be made available to other researchers.
Dr Gehrsitz added: “Our study shows the potential of large administrative datasets to evaluate policies affecting hundreds of thousands of people. This data can now support a wide range of future research on criminal justice interventions.”