Date of Original Version



Working Paper

Abstract or Description

Drug use in a population varies dramatically over time in no small measure due to nonlinear feedback among factors endogenous to the drug system. This suggests that drug policy ought likewise to be dynamic, varying the mix of strategies over time as drug use waxes and wanes. A growing literature that models drug “epidemics” mathematically supports this hypothesis and offers perspectives that may break policy logjams. For example, supply control may be most effective early, in the explosive growth stage of an epidemic. Conversely, treatment and measures to mitigate the consequences of dependent use and flagrant drug markets may have their comparative advantage later, in the endemic stage. Fully harnessing the power of dynamic drug policy will require more research and collection of new types of data, but the promise is worth the effort.