Adaptation, spatial effects, and targeting: Evidence from Africa and Asia. (March 2021)

• Record Type: Journal Article
• Authors: Lim, Krisha; Wichmann, Bruno; Luckert, Martin
• Is Part Of: World development. Volume 139(2021)
• Publication Date: 2021-03
• Subjects: Agriculture -- Adaptation -- Spatial targeting -- Spatial multiplier -- Spatial spillover; Economic history -- 1990- -- Periodicals; Economic assistance -- Developing countries -- Periodicals; 330.9
• DOI: 10.1016/j.worlddev.2020.105230
• Languages: English
• ISSNs: 0305-750X

Abstract

Highlights: The ability of farmers to adapt to changing environments is an important determinant of welfare. Neighbors may enhance adaptive capacity through spatial spillover adaptation effects. Spatial interactions amplify the marginal effects of adaptive capacity elements by 50 percent. Targeting the adaptation of few but central households is enough to generate spatial spillovers. Abstract: The ability of farmers to adapt to changing rural environments in developing countries is an important determinant of welfare. But the adaptive capacity of farmers may be constrained because economic resources, information, and institutions are often weak or missing. Farmers living close to one another face similar environmental and economic conditions, but may have different levels of adaptive capacity. These spatially connected farmers (or neighbors) can potentially ease economic constraints and facilitate adaptation by acting as conduits of information and resources. This paper estimates spatial effects on adaptation: how neighbors' adaptations influence farmers' adaptation decisions. Our data contains information from more than two thousand households located across 12 countries in Africa and Asia. We find that neighbors significantly influence adaptation decisions, and that spatial interactions amplify the marginal effects of adaptive capacity elements by 50 percent. We also propose a targeting approach that decomposes the total spatial effect into effects coming from different subsets of households that are selected according to their spatial centrality. We find that targeting the two households with highest eigenvector centrality in each village is enough to generate a statistically significant spatial multiplier effect.