The central question faced by policy makers contemplating decriminalization of cannabis is whether such a move will lead to an increase in use, and if so, by whom and by how much. We address this question by investigating the impact of decriminalization on the decision to start using cannabis. Our analysis is based on individual level information from a general population in Australia.
Australia provides an interesting case study for examining this issue because it has decriminalized the use of cannabis in half of its states and territories. In modeling cannabis uptake, we use a discrete-time hazard model and account for unobserved dierences between states that decriminalize and those that do not. We find that decriminalizing cannabis shifts the age distribution of uptake towards younger age groups while leaving the proportion of those who will start using cannabis unchanged. This suggests that decriminalization effects when individuals start using cannabis, rather than whether or not they start.
