Car crashes are a leading cause of death among younger Americans and have become a central concern in the US marijuana policy debate. I construct a novel dataset of marijuana dispensary openings, which I use to present new evidence on the effect of marijuana on traffic fatalities. My intra-state differences-in-differences approach both increases power relative to past analyses and eliminates the potential of time-varying state-level confounding. I find that marijuana dispensary openings increase the rate of fatal car crashes by approximately 5.7%. I use a series of tests to discern between two plausible mechanisms – increased traffic and increased impairment – and ultimately find that the effect is primarily driven by impairment.