This study examines the effects of military base closures on educational expenditures and student outcomes with a national panel data set of school districts between 1990 and 2002. We adopt difference-in-differences estimation in combination with propensity score matching and instrumental variables techniques to estimate these effects. We find that per-pupil spending increases by 25.2 Percent in the first year, where it remains. We also find a substantial decrease in graduation rates, but an improving trend occurs in the years after the closure. The author(s) 2013.