Skip to main content
SearchLoginLogin or Signup

The Importance of Cost-Benefit Analysis of Crime Reduction Programmes

There are many possible methods of reducing crime (see e.g. Tonry & Farrington, 1995). There is criminal justice or correctional prevention, which involves deterrence, incapacitation and rehabilitation. There is community prevention, which involves targeting community risk ...

Published onMar 05, 2023
The Importance of Cost-Benefit Analysis of Crime Reduction Programmes


There are many possible methods of reducing crime (see e.g. Tonry & Farrington, 1995). There is criminal justice or correctional prevention, which involves deterrence, incapacitation and rehabilitation. There is community prevention, which involves targeting community risk factors and social conditions, such as community cohesiveness or disorganization, or using the community as the context for intervention. There is situational prevention, which involves reducing opportunities for offending or increasing the risks of being caught in the physical environment, for example by using closed-circuit television (CCTV) cameras or improved street lighting. And there is developmental or risk-focussed prevention, which involves targeting early child or family risk factors for offending, and enhancing protective factors.

How can governmental and other policymakers decide which of these approaches is best? For example, should there be more imprisonment? More police? More rehabilitative programmes in prison or on probation? More community penalties or community programmes? More situational crime prevention? Or more developmental crime prevention?

One obvious criterion for the success of a crime prevention programme is the extent to which it reduces crime. However, this cannot be the only consideration. Our argument is that, in choosing between programmes, it is important to consider all their monetary costs and benefits, and to choose the programmes with the greatest benefit-to-cost ratios. Knowledge about value for money is also vital in trying to improve evidence-based crime prevention policy and practice. We need to know what works (and for whom) and what saves money.


In general, government policymakers do not understand effect sizes such as Cohen’s d and think that a 5% absolute decrease in conviction rates is very small. However, they do understand money in and money out. They are impressed by evidence suggesting that, for example, £5 is saved for every £1 spent on a programme, even if the savings are in the future. For example, in 1997 we were invited to go to London to HM Treasury to brief senior officials on cost-benefit analyses of crime prevention programmes (see Welsh & Farrington, 2015). Under the recently elected government of Prime Minister Tony Blair, the Treasury was spearheading an ambitious Comprehensive Spending Review, to review public expenditure across all areas of government. After our meeting, the Treasury’s involvement was a major factor in launching the government’s ambitious Crime Reduction Programme, which had an unprecedented level of funding of £250 million (see Dhiri et al., 2001).

As another example, David Farrington and Leena Augimeri met with the Deputy Minister and Assistant Deputy Minister of Public Safety Canada in November 2013 and briefed them about cost-benefit analysis and the SNAP (Stop Now And Plan; see later) early developmental prevention programme (see Augimeri et al., 2007; Farrington & Koegl, 2015). At this meeting, the Deputy Minister committed $6.3 million to extend SNAP all over Canada.

The metric used in measuring the effectiveness of programmes is important. The Washington State Institute for Public Policy (WSIPP) provides the greatest range of information about benefit-to-cost ratios of crime prevention and crime control programmes. For example, for restorative justice conferencing, they reported Cohen’s d = 0.072, based on the UK research of Shapland et al. (2008; see Assuming a 50% reconviction rate (see Farrington et al., 2022), this value of d corresponds to a 3.3% absolute reduction in reconviction rates, or a 6.6% relative reduction – apparently small effects. However, based on 2018 costs and benefits of this programme, WSIPP estimated that the benefit-to-cost ratio was 2.05 (cost per participant = $1,166; benefits per participant based only on reduced crime = $2,391). Therefore £2.05 was saved for every £1 expended on this programme, which does not seem a small effect.


It is important to consider the financial costs of each programme. There are initial capital costs versus operating costs, and there are average costs versus marginal costs. For example, the marginal cost of placing one more person in a prison may not be very great. It is also important to take account of inflation, if the costs are immediate and the benefits are long delayed.

The main benefits of a crime prevention programme accrue from a reduction in the costs of crime. Therefore, it is important to measure all the financial costs of crime, which may be borne by the government (i.e. the taxpayer), the victim or the programme participant (see e.g. Cohen, 2020; Welsh & Farrington, 2000; Welsh et al., 2001). For example, the governmental costs include criminal justice costs of police, courts, prisons and probation, as well as health, medical, education and welfare costs. The costs to victims include tangible or out-of-pocket costs (e.g. property loss or damage, medical expenses, lost wages) and intangible costs (e.g. distress, suffering, fear, lower quality of life). Assigning monetary values to intangible costs can be controversial but is necessary (see e.g. Cohen & Farrington, 2021). The most common method of estimating intangible costs is to ask people how much they are willing to pay to avoid a hazard, such as being burgled or robbed; more information about this is presented in the article by B. Raffan Gowar, David Farrington and Maria Ttofi. The costs to programme participants (and their families) accrue from the effects on their lives (e.g. in education, employment or relationships) of being apprehended, convicted and punished.

The first governmental attempt to estimate the cost of crime in England and Wales was completed by Brand and Price (2000), based on 1999 prices. They estimated, for example, that each homicide cost, on average, £1.1 million, each sexual offence cost £19,000, and each domestic burglary cost £2,300. Their work has been updated since, and the latest report was published by Heeks et al. (2018), based on 2015-16 prices. This estimated that each homicide cost about £3.2 million, each rape cost £39,000, and each domestic burglary cost £5,900.

In order to decide on the effectiveness of a crime prevention programme, it is essential to estimate what would have happened in the absence of that programme. Therefore, it is essential to compare those who received an intervention (individuals or areas) with a (comparable or near-equivalent) control condition which did not receive the programme. The control condition is needed in order to estimate how many offenders and crimes were prevented by the programme, as well as the effects of the programme on other outcomes (e.g. education, employment or relationships). It is important to ensure that the intervention condition is similar to the control condition before the intervention, for example by randomly allocating individuals or areas to conditions or by matching, preferably with before and after measures of criminal and other outcomes.


The most important crime prevention methods are developmental, situational, community and correctional. As an example of a cost-benefit analysis of developmental prevention, the SNAP programme (discussed above) in Toronto, Canada, was evaluated by Farrington and Koegl (2015). The main aim of this programme, targeted on children at risk, is to help them to think before they act, and to choose desirable actions in provoking situations. Based on the best available evaluations, the researchers estimated that the programme caused a relative decrease in offending of between 18% and 33%. From knowing what types of crimes were committed (leading to convictions) by these young people (boys) between ages 12 and 20, and from knowing the average cost of each type of crime, the researchers estimated that the cost saving per boy (at 2012 prices) was between $9,493 and $17,404. Since the cost of the programme per boy was $4,641, the benefit-to-cost ratio based on reduced convictions was between 2.1 and 3.8 to 1. However, convictions are only the “tip of the iceberg” of offending. From knowing the “scaling-up factors” from convictions to self-reported offences of different types, the researchers estimated that the benefit-to-cost ratio based on actual offences committed was between 17.3 and 31.8 to 1. This means that, for each dollar spent on the programme, between $17 and $32 was saved to the government, crime victims and the participants themselves.

As an example of a cost-benefit analysis of situational prevention, Painter and Farrington (2001) evaluated the effect of improved street lighting on crime in Dudley, in the UK. The lighting was improved in an experimental area and this was compared with a comparable control area which had unchanged lighting. Before and after victimization surveys were carried out in each area to assess changes in crime rates. The results showed that crime decreased by 41% in the experimental area and by 15% in the control area. Altogether, there was a net reduction of 509 crimes in the experimental area in one year, saving £339,186 according to UK Home Office estimates of the cost of different types of crimes (adjusted to 1993 prices). Since the capital cost of the improved lighting was £54,815, the benefit-to-cost ratio was 6.2 to 1 after one year. If it was assumed that the capital cost would be paid off over a 20-year period, it was estimated that this would cost £4,611 per year (taking account of the annual cost of the lighting). Assuming a continued lower crime rate, this yielded a remarkable benefit-to-cost ratio of 121 to 1. A similar evaluation in Stoke-on-Trent yielded benefit-to-cost ratios of 5.4 to 1 after one year and 54 to 1 after assuming that the capital cost was paid off over 20 years.

As an example of a cost-benefit analysis of a community intervention, Fagan et al. (2019) evaluated “Communities That Care” (CTC). This programme aimed to reduce offending by implementing prevention strategies to reduce risk factors for offending and enhance protective factors against offending. In the evaluation, 24 communities across the United States were placed into 12 matched pairs, and one community in each pair was chosen at random to receive CTC. The effectiveness of CTC was assessed using surveys of over 4,000 students from grade 5 (age 10-11) to grade 12 (age 17-18). Kuklinski et al. (2015) found that the programme reduced delinquency, alcohol use and smoking cigarettes, saving an average of $4,809 per youth (at 2011 prices). Since the average cost of the programme per youth over five years was $556, and taking account of inflation, they reported a benefit-to-cost ratio for CTC of 8.22 to 1.

As an example of a cost-benefit analysis of correctional treatment, Farrington et al. (2002) evaluated a “High Intensity Treatment” programme in a UK Young Offender Institution (YOI). The programme was an intensive military style regime that had significant rehabilitation components (e.g. cognitive-behavioural treatment, drugs education), and it was delivered to young offenders aged 18-21; 175 experimental offenders were compared with 127 control offenders who were in other YOIs. Pre-existing differences between experimental and control offenders were taken into account using predicted reconviction scores. A two-year follow-up showed that the average savings per inmate (based on UK Home Office estimated costs of crimes, adjusted to 1999 prices) was £2,480, compared with the extra cost of the regime of £2,441 per inmate. Therefore, the benefits just outweighed the costs after two years. However, Jolliffe et al. (2013) then followed up the offenders for 10 years in criminal records, and estimated that the benefit-to-cost ratio increased from 1.1 after two years to 3.9 after 10 years. This was the result of continued relative reductions in re-offending by the experimental group offenders compared to their control group counterparts.

This special issue includes new reviews of cost-benefit analyses of developmental crime prevention programmes (by Christopher Koegl, David  Farrington and Brandon Welsh) and correctional treatment programmes (by Steven Zane, Jhon Pupo and Brandon Welsh). In addition, there is a cost-benefit analysis of the famous Montreal Longitudinal and Experimental Study, by Adam Vanzella-Yang and colleagues. Last but not least, there is a new review of economic evaluations (including cost-benefit analyses) of mental health interventions in criminal justice, by Martin Knapp and Gloria Wong.


Cost-benefit analyses of crime prevention programmes have previously been reviewed in detail by Welsh et al. (2015). They found that developmental, situational, and community prevention methods are each effective in reducing crime, as well as in mitigating other social problems. Additionally, the evidence shows that these crime prevention methods often have large benefit-to-cost ratios over short and longer time frames. Welsh et al. (2023) concluded that benefit-to-cost ratios were generally greater for early prevention than for imprisonment. However, more research is needed, and the articles in this special issue draw attention to key research priorities for the years ahead. It is our hope that the new scientific knowledge reported here will lead to more rational, evidence-based expenditure on crime prevention and to increased public safety. This would be in the interests of everyone.


Data sharing is not applicable to this article as no new data were created or analysed in this study.


Augimeri, L. K., Farrington, D. P., Koegl, C. J., & Day, D. M. (2007). The SNAP Under 12 Outreach Project: Effects of a community based program for children with conduct problems.  Journal of Child and Family Studies, 16, 799-807.

Brand, S., & Price, R. (2000). The economic and social costs of crime. London: Home Office (Research Study No. 217).

Cohen, M. A. (2020). The costs of crime and justice (2nd ed.). London: Routledge.

Cohen, M. A., & Farrington, D. P. (2021). Appropriate measurement and use of “costs of crime” in policy analysis: Benefit-cost analysis of criminal justice policies has come of age. Journal of Policy Analysis and Management, 40, 286-293.

Dhiri, S., Goldblatt, P., Brand, S., & Price, R. (2001). Evaluation of the United Kingdom’s “Crime Reduction Programme”: Analysis of costs and benefits. In B. C. Welsh, D. P. Farrington, & L. W. Sherman (Eds.), Costs and benefits of preventing crime (pp. 179-201). Boulder, CO: Westview Press.

Fagan, A. A., Hawkins, J. D., Catalano, R. F., & Farrington, D. P. (2019). Communities That Care: Building community engagement and capacity to prevent youth behaviour problems. New York: Oxford University Press.

Farrington, D. P., Ditchfield, J., Hancock, G., Howard, P., Jolliffe, D., Livingston, M. S., & Painter, K. A. (2002). Evaluation of two intensive regimes for young offenders. London: Home Office (Research Study No. 239).

Farrington, D. P., Gaffney, H., & White, H. (2022). Effectiveness of 12 types of interventions to reduce juvenile offending and antisocial behaviour. Canadian Journal of Criminology and Criminal Justice, in press.

Farrington, D. P., & Koegl, C. J. (2015). Monetary benefits and costs of the Stop Now And Plan program for boys aged 6-11, based on the prevention of later offending. Journal of Quantitative Criminology, 31, 263-287.

Heeks, M., Reed, S., Tafsiri, M., & Prince, S. (2018). The economic and social costs of crime (2nd ed.). London: Home Office (Research Report No. 99).

Jolliffe, D., Farrington, D. P., & Howard, P. (2013). How long did it last? A 10-year reconviction follow-up study of High Intensity Training for young offenders. Journal of Experimental Criminology, 9, 515-531.

Kuklinski, M. R., Fagan, A. A., Hawkins, J. D., Briney, J. S., & Catalano, R. F. (2015). Benefit-cost analysis of a randomized evaluation of Communities That Care: Monetizing intervention effects on the initiation of delinquency and substance use through grade 12. Journal of Experimental Criminology, 11, 165-192.

Painter, K. A., & Farrington, D. P. (2001). The financial benefits of improved street lighting, based on crime reduction. Lighting Research and Technology, 33, 3-12.

Shapland, J., Atkinson, A., Atkinson, H., Dignan, J., Edwards, L., Hibbert, J., Howes, M.,

Johnstone, J., Robinson, G., & Sorsby, A. (2008). Does restorative justice affect reconviction: The fourth report from the evaluation of three schemes. London: Ministry of Justice (Research Series 10/08).

Tonry, M., & Farrington, D. P. (Eds.) (1995). Building a safer society: Strategic approaches to crime prevention. Chicago: University of Chicago Press.

Welsh, B. C., & Farrington, D. P. (2000). Monetary costs and benefits of crime prevention programs. In M. Tonry (Ed.), Crime and Justice, vol. 27 (pp. 305-361). Chicago: University of Chicago Press.

Welsh, B. C., & Farrington, D. P. (2015). Monetary value of early developmental crime prevention and its policy significance. Criminology and Public Policy, 14, 673-680.

Welsh, B. C., Farrington, D. P., & Raffan Gowar, B. (2015). Benefit-cost analysis of crime prevention programs. In M. Tonry (Ed.), Crime and Justice, vol. 44 (pp. 447-516). Chicago: University of Chicago Press.

Welsh, B.C., Farrington, D.P. & Sherman, L.W. (Eds.) (2001). Costs and benefits of preventing crime. Boulder, CO: Westview Press.

Welsh, B.C., Paterson, H., & Farrington, D.P. (2023). Early prevention as an alternative to imprisonment: The research evidence on monetary costs and benefits.   In B.C. Welsh,  S.N. Zane, & D.P. Mears (Eds.), The Oxford handbook of evidence-based crime and justice policy. New York: Oxford University Press, in press.

No comments here
Why not start the discussion?