Cautioning Cannabis Offenders Instead of Arresting Them Costs Half as Much with No Difference in Use | Research

The takeaway

An Australian study found that police cannabis cautions were half the cost of arrest ($388 vs. $733) with no statistically significant difference in subsequent cannabis use between the two groups.

Readers interested in cannabis policy and alternatives to criminal penalties.

Caution cost $388 vs. arrest cost $733, with identical cannabis use outcomes

What the researchers found

Researchers compared two police responses to minor cannabis offenses in Australia: formal caution (with referral to education/treatment) versus traditional arrest and charge.

Among 544 matched offenders (195 arrested, 355 cautioned), there was no statistically significant difference in self-reported cannabis use days before and after the police intervention between the two groups. Both groups showed similar patterns of continued or changed use regardless of whether they were arrested or cautioned.

However, costs were significantly different. The mean cost for arrest was $733 compared to $388 for caution, after matching on age, prior criminal conviction, cannabis consumption, employment status, criminal activity, and severity of dependence. Cautioning saved roughly half the cost per offender while producing equivalent outcomes.

Why it matters

This study provides economic evidence for cannabis diversion programs. If cautioning produces the same outcomes as arrest at half the cost, the policy argument for diversion becomes straightforward: it saves public resources without increasing cannabis use. This evidence is relevant to the global debate about decriminalization and alternatives to criminal penalties.

The numbers in context

544 matched offenders. 195 arrested, 355 cautioned. Mean cost: arrest $733 (SD 151), caution $388 (SD 111). No significant difference in cannabis use pre- and post-intervention.

How the study worked

Purpose-built nationwide online survey of detected cannabis offenders in Australia. Propensity score weighting and doubly robust regression addressed differences between groups. Cost data included policing, court, penalties, assessment, treatment, and educational sessions.

What this study cannot tell us

Self-selected online survey of detected offenders, which may not represent all cannabis offenders. Self-reported cannabis use is subject to bias. The study could not account for deterrence effects on the broader population (people who were never detected). Cost estimates did not include long-term consequences of criminal records on employment and housing.

How to read the evidence

Moderate evidence from a nationwide survey with propensity score matching, limited by self-report and self-selection.

When this study was published

Published in 2017. Australia continues to evolve its cannabis diversion policies.

The bigger picture

Australia has been a pioneer in cannabis diversion programs, and this study validates those approaches from a cost-effectiveness standpoint. The finding aligns with broader evidence that criminal penalties for minor cannabis possession do not reduce use but do impose substantial costs on both offenders and the justice system.

Questions still open

Would including the long-term costs of criminal records (lost employment, housing barriers) make the case for cautioning even stronger? Does the type of educational intervention offered during cautioning affect outcomes? How do these findings translate to other countries' legal and policing contexts?

Common questions

Does arresting people for cannabis possession actually reduce use?

This study found no difference in cannabis use between people who were arrested versus cautioned. Both groups showed similar patterns, suggesting that the threat of arrest does not produce additional deterrence compared to a formal warning with education referral.

What is a cannabis caution?

In Australian cannabis cautioning schemes, police issue a formal warning and refer the offender to education and/or treatment instead of arresting and charging them. It avoids criminal records for minor possession while still providing an intervention.

Read the original research

Alternate policing strategies: Cost-effectiveness of cautioning for cannabis offences.

The International journal on drug policy, 41, 140-147

View on PubMed ↗Publisher’s paper ↗

Citation

Shanahan, Marian; Hughes, Caitlin Elizabeth; McSweeney, Tim; Griffin, Beth Ann. (2017). Alternate policing strategies: Cost-effectiveness of cautioning for cannabis offences.. The International journal on drug policy, 41, 140-147. https://doi.org/10.1016/j.drugpo.2016.12.012