Are there agglomeration economies in crime? The positive correlation between city size and crime rates is well-known. This paper establishes a positive relationship between youth cohort size and marijuana use rates. It further demonstrates a negative association between youth cohort size and marijuana prices, youth drug possession arrest rates, and both overall and youth sales arrest rates. Cohort size affects demand by lowering possession arrest probabilities, but this factor explains less than 10 percent of the relationship. The main effect shown here, accounting for at least a quarter of the relationship, is on the supply of marijuana. Larger youth cohorts yield thicker drug markets that, through lower sales arrest risk and informational economies, generate cost-savings in drug distribution.