How Can We Help Farmers When They Are Already Clever

The ability of farmers to adapt to changing rural environments in developing countries is an important determinant of welfare. However, farmers' adaptation may be constrained by their adaptive capacity because economic resources, information, and institutions are often weak or missing in these areas. Networks of relationships can potentially ease these constraints and facilitate adaptation by acting as conduits of information and resources. The contribution of this thesis is three-fold. First, using the number of farming practices households have changed over the last ten years as our measure of adaptation, we investigate network effects on farmers' adaptation decisions. We use spatial econometric techniques to estimate the effects of adaptive capacity elements and neighbors' adaptation on farmers' adaptation. Second, we propose an approach that analyzes whether or not the adaptation of a subset of neighbors also generates significant network effects. We decompose the total network effect into network effects coming from the most central household, the two most central households, and so on. Third, using the number of food secure days in a year as a measure of households' welfare, we show how the spatial attributes of households suggests instrumental variables that can be used to address the endogeneity issue in welfare analysis. We use a rich dataset that contains information from 2,095 households located across 12 countries in Africa and Asia, which provides robust and generalizable findings. Our data allows us to examine the importance of network effects, in addition to traditional adaptive capacity elements, including access to information, human capital, financial resources, physical assets, farm and household characteristics, and farming and crisis experience. We find that neighbors significantly influence adaptation decisions, and network interactions amplify the marginal effects of adaptive capacity elements by 50 percent. In addition, we find that there are benefits to targeting fewer, but more central, households. Finally, we find that one additional farming practice changed increases welfare by 5.5 food secure days. Our results imply that investing in adaptation programs that relax adaptive capacity constraints could help farmers improve their welfare, and network effects not only catalyzes impacts of policy interventions but also offers a targeting strategy.