CJB258
Why are Aboriginal adults less likely to receive cannabis cautions?
Download this publication
Summary
Background
"The Cannabis Cautioning Scheme was introduced in NSW in April 2000, creating a formal diversion pathway for adult offenders who meet several eligibility criteria relating to prior offending, cannabis quantity and admission of guilt, among others.
We use a dataset of 38,813 observations involving 27,127 adult offenders proceeded against for a cannabis use/ possession incident between January 2017 and February 2020.
We first explore differences in Aboriginal and non-Aboriginal adults’ eligibility for the Scheme. Then we apply a Kitagawa-Oaxaca-Blinder (KOB) decomposition to a sample of 18,395 observations involving 15,869 adult offenders who met the eligibility criteria for a cannabis caution. This splits the gap in cannabis cautioning rates into components attributable to offender and offence characteristics, and an unexplained component representing any unobserved differences in how similar offenders are treated."
Key findings
Figure 1. Offender characteristics which explain the disparity in Aboriginal vs non-Aboriginal cannabis cautioning for eligible offenders
"There was a 32.2 p.p. difference in cautioning rates between all Aboriginal and non-Aboriginal people charged with cannabis use/possession (11.7% vs 43.9% for Aboriginal and non-Aboriginal people respectively). Eligibility is a major source of this disparity. Around 78% of Aboriginal people proceeded against for a cannabis offence were ineligible to receive a caution compared with 45% of non-Aboriginal people. Considering only eligible offenders, we find a 34.4 p.p. difference in cautioning rates (39.5% for Aboriginal offenders vs 73.9% for non-Aboriginal offenders).
Figure 1 presents our KOB decomposition results, showing the proportion of this disparity between Aboriginal and non-Aboriginal offenders that can be explained by observed factors. We find that the difference in cautioning rates is explained by:
• higher levels of prior offending and imprisonment among Aboriginal offenders (24.5 p.p. or 71% of the disparity);
• police jurisdiction level variation in cautioning rates explain (5 p.p. or 15%); and
• demographics (2 p.p. or 6%).
The remaining 2.9 p.p. (8% of the gap) is unexplained and may arise either because of a difference in how Aboriginal offenders are treated or in other unobserved factors not included in the model."