We study whether exposure to a more risky and more social work environment has an effect on risk- and social preferences. Combining experimental and administrative data from selected fisheries in three countries bridges a gap between existing case-study evidence and global estimates from historical data. While we do not find strong evidence for endogenous social preferences, we do find evidence for endogenous risk preferences, especially in Chile, where the differences in risk exposure are most pronounced. Making use of the fact that we have repeated observations from some fishers, we make a first pass at disentangling selection from adaptation as potential mechanisms that make preferences endogenous. For Chile, we find suggestive evidence for an adaptation process within fishers, while for Tanzania and Norway, the data speaks more towards a selection process that changes the composition of the population in line with risk exposure.