Security measures intended to deter crime may unwittingly displace it to neigh- boring areas, but evidence of displacement is scarce. We exploit precise information on the timing and locations of all bank robberies and security guard hirings and firings in Italy over a 10-year period to estimate the deterrence and displacement efeects of guards. We find that hiring a security guard lowers the likelihood that a bank is robbed by 35-40 percent, though over half of this reduction is immediately displaced to nearby banks that are unguarded. A simple theoretical model of dis- placement reveals ambiguity in policies to mitigate these spillovers. Our findings suggest that policies that restrict the use of guards in sparse, rural markets and that require the use of guards in dense, urban markets could be socially beneficial.