View article

We develop an integrated framework for testing theories of consumption behavior by examining how agricultural households in rural India respond to anticipated and unanticipated seasonal income fluctuations. Using information on village‐level rainfall surprises, we estimate idiosyncratic unanticipated income shocks by allowing weather risk to impact households differently depending on observable characteristics. In applying our methodology to unique panel data on consumption, debt, and intrahousehold gift exchanges, we address a number of shortcomings in previous tests of theories of consumption behavior. Yet, we find no evidence against the hypothesis that households smooth idiosyncratic fluctuations in their income. Our approach also uncovers several interesting features of rural credit and insurance markets.