We study empirically how past exposure to conflict in origin countries makes migrants more violent prone in their host country, focusing on asylum seekers in Switzerland. We exploit a novel and unique dataset on all crimes reported in Switzerland by nationalities of perpetrators and victims over the period 2009-2012. Causal analysis relies on the fact that asylum seekers are exogenously allocated across the Swiss territory by the federal administration. Our baseline result is that cohorts exposed to civil conflicts/mass killings during childhood are on average 40 percent more prone to violent crimes than their co-nationals born after the conflict. The effect is stable through the lifecycle and is attenuated for women, for property crimes and for low-intensity conflicts. Further, a bilateral crime regression shows that conflict exposed cohorts have a higher propensity to target victims from their own nationality - a piece of evidence that we interpret as persistence in intra-national grievances. Last, we exploit cross region heterogeneity in public policies within Switzerland to document which integration policies are able to mitigate the detrimental effect of past conflict exposure on violent criminality. In particular, we find that offering labor market access to asylum seekers and fostering social integration eliminates all the effect.