Using Italian province-year panel for the years 2010-2014, this paper examines the impact of illicit drug seizures on drug consumption. Specifically, we focus on cocaine market and aim to uncover the relationship and the causal effect of cocaine seizures on cocaine consumption proxied by cocaine-related hospitalization rates. The paper contributes to the existing literature in several ways. Firstly, we build a new panel dataset that has some favourable features as opposed to those used in the previous studies. Secondly, unlike the existing literature that barely touches the endogeneity issue, we address it by resorting to the instrumental variable estimation, using seaports turnover as an instrument for seizures. Our results suggest that there is a stable statistically significant relationship between cocaine seizures and consumption: on average, a one standard deviation increase in a province's cocaine seizures rate is associated with a 0.033 standard deviation decrease in related hospitalization rates; with an instrumental variables approach this effect reaches about 0.093 standard deviation. Finally, we also explore spatial interaction between provinces and find a negative relationship between seizures in key entry points and consumption in the rest of the country, as well as a negative relationship between seizures in adjacent provinces and consumption in a home province.