Résumé
Does exposure to a more risky environment affect risk preferences? Going beyond single‐case study evidence, we report results from five surveys conducted in three countries and link this with administrative data to study whether a link between exposure and preferences is detectable and widespread. We find no evidence for endogenous preferences in Norway and Tanzania, but relatively strong evidence in Chile, where differences in risk exposure are most pronounced. Moreover, we make a first pass at disentangling selection from adaptation as potential mechanisms. For Tanzania and Norway, the data speaks for selection, while it speaks for adaptation in Chile.