Australian Institute of Criminology

Skip to content


There has been a great deal of interest among policymakers and criminal justice practitioners in the use of locally targeted strategies to reduce Indigenous overrepresentation in the criminal justice system, particularly in prisons. Approaches such as justice reinvestment, which aims to direct resources away from the prison system towards locally based crime prevention initiatives, have been highlighted a way of reducing this overrepresentation. These approaches are underpinned by research that demonstrates strong links between risks of engagement with the criminal justice system and residential locations.

In an environment of limited resources and competing policy interests, there is a critical need for crime prevention interventions to be cost effective and directed towards those populations and areas with the greatest need. Being able to identify those communities and locations where chronic and persistent offenders are most likely to reside, and understanding the factors that tend to produce the interconnections between place and offending, has enormous potential for the effective targeting of crime prevention initiatives.

This paper provides a valuable contribution to the evidence base supporting effective crime prevention targeting. By showing that chronic offenders, who account for a large proportion of all offences, are not randomly distributed geographically, those locations where interventions and resources can be effectively directed are highlighted. These locations typically have very high levels of social and economic disadvantage, pointing to the need for programs and social interventions that can effect change at the community level. Interventions likely to yield demonstrable outcomes are those that work with families and that operate holistically to address the multiple dimensions of disadvantage. While there is much still to be done to develop and implement interventions with lasting benefits, the methodological approach and findings reported in this paper will go a long way to informing these efforts.

Adam Tomison