Most interpretations of prevalent counterinsurgency theory imply that increasing government services will reduce rebel violence. Empirically, however, development programs and economic activity sometimes yield increased violence. Using new panel data on development spending in Iraq, we show that violence reducing effects of aid are greater when (a) projects are small, (b) troop strength is high, and (c) professional development expertise is available. These findings are consistent with a "hearts and minds" model, which predicts that violence reduction will result when projects are secure, valued by community members, and implementation is conditional on the behavior of non-combatants.