Can drug policies modify cannabis use starting choice? Insights from criminalisation in Italy

A key question in the ongoing drug policy debate is whether legalising cannabis leads to an increase in cannabis use. In Europe although no country has yet moved to legalisation, many have decriminalised personal possession. However, some jurisdictions are still discussing increased sanctions or have further strengthened penalties for the possession of illicit substances in order to deter widespread cannabis use. This is the case in Italy, where a law introduced in 2006 and repealed in 2014 de facto criminalised personal drug possession, and a potential increase in penalties is currently being debated as a policy option. Despite the intense public debate surrounding the legal status of cannabis, limited empirical research has been conducted in Europe to assess the population-level effects of drug policy reforms, mainly due to data availability constraints. In this study, we analyse the effect of criminalisation on the age of onset of cannabis use using an unique dataset that combines seven waves (2001-2017) of the nationally representative Italian Population Survey on Alcohol and other Drugs with relevant socio-economic data. The final dataset comprises 77,650 observations. Leveraging the rare opportunity to examine the effects of a policy that remained in force for a limited period, our empirical investigation employs a Complementary Log-Log model to analyse the starting rate, that is, the transition rate from non-use to use. To do so, we use self-reported data on the age of first cannabis use. Our results suggest that the implementation of stricter punishments has a significant effect in reducing the likelihood of early cannabis use initiation. The observed impact of criminalisation is limited in younger ages and diminished as adulthood approaches. This paper also discusses other considerations related to the social costs of criminalisation, which should also be taken into account in the ongoing policy debate.