We analyze financial contracting versus social contracting as household risk management tools, employing U.S. agriculture data. First, we document that households in communities that are endowed with more risk exposure, measured by inferior soil quality, exhibit higher religious adherence rates. Second, we analyze the Federal Crop Insurance Reform Act of 1994, which reduced the cost of crop insurance significantly, and find that financial contracting displaced social contracting. Finally, our comprehensive analysis of the economic effects of such substitution reveals that financial contracting resulted in smaller risk pools (i.e., church congregations) and less diversification, but without a commensurate increase in productivity, pointing to more intense moral hazard problems. The evidence has implications for public policy as well as our understanding of households’ cost-benefit analysis of available risk management tools. Joint with Mitch Warachka and Frank Yu.