Thirteenth International Conference on Climate Change: Impacts & Responses

Abstract

We use a unique dataset that combines household-level survey data with World Food Programme (WFP) delivery data, and GIS-based measures of weather shocks and climate conditions in order to evaluate the impact of both aid and weather shocks on consumption outcomes. To better understand the pattern of WFP deliveries and potential geographic targeting errors, we evaluate which GIS-based measures of droughts and floods predicted maize yields and aid deliveries. We find that the best predictors of maize yields do not predict WFP deliveries. We next analyze the factors that predict household receipt of food aid under weather shocks, which we instrument in the consumption outcome regressions using WFP deliveries. We find that both weather shocks and WFP deliveries have a positive impacts on aid receipt, though there is also evidence of WFP targeting errors. We then evaluate the impacts of weather shocks and aid on our consumption outcomes. For weather shocks, we control for both a current-period only and sequential weather shocks. Results show that weather shocks have a negative impact on non-food consumption, and that sequential shocks in particular have significant negative impacts on the food-based consumption variables. Aid receipt has fewer significant impacts, but does increase non-food expenditures, the food consumption score and school participation rates under sequential shocks. In some specifications, the value of food consumed and calories per capita are also higher when aid is interacted with sequential shocks.