Abstract
Climate change is expected to increase the frequency and severity of extreme weather events, which will have significant implications for economic behaviour, including decisions to migrate. Although an increasing body of literature examines the relationship between climate and migration, most studies rely on aggregated data, raising concerns about ecological fallacies when inferring individual behaviour. Therefore, complementary micro-level approaches have focused on the role of environmental risk perceptions in shaping migration intentions. However, the measures used in these studies are often based on self-reported perceptions, which are subject to measurement error and endogeneity. In particular, unobserved factors —such as institutional quality or perceived environmental safety— may influence both risk perceptions and migration intentions together, which can bias estimates. This study addresses these challenges by conducting a micro-level analysis in Indonesia that focuses on a small island context which is highly vulnerable to climate-related risks, including extreme weather events and rising sea levels. To address potential endogeneity concerns, we construct a novel measure of coastal resilience in local communities and use it as an instrument for environmental risk exposure. Our findings aim to provide more credible evidence of the behavioural mechanisms linking climate risk perceptions to migration, offering policy-relevant insights for vulnerable regions. Therefore, our work contributes to the literature on migration and environmental economics by providing a novel measure and applying it to analyse highly vulnerable regions.