LSE - Small Logo
LSE - Small Logo


February 21st, 2019

Are non-custodial sentences a credible and cost-effective substitute to incarceration?

0 comments | 28 shares

Estimated reading time: 5 minutes


February 21st, 2019

Are non-custodial sentences a credible and cost-effective substitute to incarceration?

0 comments | 28 shares

Estimated reading time: 5 minutes

Nick Cowen, Siddhartha Bandyopadhyay, and Juste Abramovaite analyse the effects of custodial and non-custodial sentences on recorded crime in England and Wales. Their results suggest that non-custodial sentences can be an effective alternative to custody when it comes to reducing property crime but their effect is less consistent when looking at violent crime.

As the criminal justice system struggles with government cuts, a recent report by the Centre for Justice Innovation has suggested that judges have lost faith in community sentences because of a lack of communication with recently privatized probation providers. Yet the human toll and economic costs of prison are well-established and recognized by judges and magistrates. Can there be cost-effective, more humane alternatives to custody?

To answer this question we examined how variations in sentencing across police force areas from 2002 to 2013 affect police recorded crime for the key categories of theft, violence, sexual offences and robbery. We find that the criminal justice system typically plays a critical role in reducing crime, and that community sentences and other alternatives to custody can be particularly useful sanctions for courts aiming to protect their communities. Since they are often less expensive than prison as well, it appears that there are ways of protecting the public more cost-effectively than relying on custody.

Measuring the performance of public services is always difficult. The challenge is compounded with the criminal justice system because one of the key things we are interested in measuring is how much crime is prevented, not just how it is dealt with once committed. For most people not having to deal with the criminal justice system at all is the surest sign of its success. Thus, research evaluating sentencing effectiveness has to try and work out what would have happened without that particular sanction being handed out. The problem cases, repeat offenders for example, are typically more visible than the successes. This may motivate politicians to insist on reform without necessarily understanding what is already working within the system.

There are several mechanisms through which the criminal justice is theorized to reduce crime:

Incapacitation: incarceration in prison or greater surveillance in the community can prevent an individual offender from re-offending during a sentence.

Rehabilitation: a sentence can be an opportunity to address an offender’s problematic behavior that leads to them committing crime through therapeutic interventions, especially to address drug addiction, and the provision of education and training.

Specific deterrence: the experience of being convicted and punished is unpleasant and is meant to discourage people from committing crime in the future.

General deterrence: the threat of being detected and then punished deters not only offenders who directly experience the sanction, but also potential offenders in the community from committing crimes that they would otherwise attempt.

However, there are also ways in which sanctions can produce perverse criminogenic outcomes that lead people to commit more crime than if they hadn’t been sanctioned at all. This can apply especially to prison sentences. Incapacitation can prevent offenders from accessing legitimate sources of employment, education, and supportive social structures such as their families. This can make them less likely to reintegrate successfully at the end of a sentence. The social experience of punishment may actively bolster their ‘deviant’ identity. Offenders may also learn new skills that allow them to avoid detection when they commit future crimes. They may meet experienced criminals with whom they can cooperate and form gangs. They may learn that punishment is not as unpleasant an experience as they had feared, or their tastes may adapt to be more accepting of future sanctions. For the truly desperate, the prospect of prison may offer respite from homelessness and insecurity. Such cases, if they can be identified, can be handled more effectively with social support than with criminal sanctions.

Thus, there are plausible reasons to be both optimistic and sceptical about the effects of criminal sentencing on crime. At the same time, the range of channels through which sentences change behaviour means that research on individual offenders is likely to miss some of the broader social impact. This is precisely what our data and approach can help to identify.

In order to get an intuitive handle on our results, we have estimated the association between a 1% increase in one sentence type for an offence category and the following year’s crime rate. Our results attempt to control for some critical socio-economic factors (age composition and unemployment) that determine the environment in which people see committing crime as a viable option. Our stylized example is based on recorded crime figures for 2014 based on a hypothetical nationwide change in sentencing activity in 2013. We use statistically significant estimated coefficients keeping in mind that these predictions have upper and lower bounds that can be computed from the standard errors.

Property crime: Sentencing 1% more offenders to prison for property offences (including theft and handling) reduces next year’s recorded crimes by 2,693. However, a similar 1% increase in community sentences reduces these offences by 3,590. In this example, 1% is going to be approximately 320 extra offenders sentenced for a property offence. Thus, it appears there is scope to reduce property crime (72% of recorded crimes in our analysis) more cost-effectively and humanely through greater use of community sentences instead of prison.

Violence: Sentencing 1% more adult violent offenders to custody reduces next year’s violence against the person offences by 1,153. 1% more suspended sentences reduces such offences by 649. This suggests that there is scope to tackle violent crime more efficiently with less costly suspended sentences (that are often combined with community orders) and less reliance on immediate custody.

Sexual offences: Sentencing 1% more adult sexual offenders to prison reduces the following year’s recorded sexual offences by 94.

Robbery: in contrast to the other offences, robbery appears to be harder to tackle more effectively through variations in existing sanctions. Sentencing an additional 1% of adult robbery offenders to prison appears to increase the following year’s robbery numbers by 29. Sentencing an extra 1% of adult robbers to community sentences reduces the next year’s number of robberies by around 8. This may imply that we are near the limits of what variations in sentencing can do for preventing robbery victimisation.

Our results suggest that the criminal justice system in England and Wales generally helps to reduce crime and that alternatives to custody (particularly community sentences) can be less disruptive to offenders and more cost-effective than prison in many cases. We think that re-establishing trust between courts and community sentence providers is an urgent priority. Good policy should offer courts a range of appropriately resourced, credible sanctions that can channel both existing and potential offenders away from committing crime and into more socially cooperative and personally beneficial activities.


Note: the above draws on the authors’ published work in The British Journal of Criminology.

About the Authors

Nick Cowen is Classical Liberal Institute Fellow at New York University.




Siddhartha Bandyopadhyay is Professor of Economics at the University of Birmingham.



Juste Abramovaite is Research Fellow at the University of Birmingham.




All articles posted on this blog give the views of the author(s), and not the position of LSE British Politics and Policy, nor of the London School of Economics and Political Science. Featured image credit: Pixabay (Public Domain).



Print Friendly, PDF & Email

About the author


Posted In: Featured | Public Services and the Welfare State

Leave a Reply

Your email address will not be published. Required fields are marked *

Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
This work by British Politics and Policy at LSE is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported.