I use an instrumental variable approach to estimate the effect of excise taxes on cigarette smuggling. The IV approach addresses the potential endogeneity of excise taxes while controlling for other determinants of smuggling. I use panel data on 47 states from 1990-2009. The main results confirm the validity of the instrument, the percent of Democrats in the upper house of state legislatures, but do not reject exogeneity of excise taxes. Robustness tests using an alternative measure of cigarette smuggling find the opposite result. All models find that per capita income and the number of federal police per 100,000 residents are significant determinants of smuggling.