OBJECTIVES. Long-distance social relationships have been a feature of human evolutionary history; evidence from the paleoanthropological, archaeological, and ethnographic records suggest that one function of these relationships is to manage the risk of resource shortfalls due to climate variability. We should expect long-distance relationships to be especially important when shortfalls are chronic or temporally positively autocorrelated, as these are more likely to exhaust local adaptations for managing risk. Further, individuals who experience shortfalls not as rare shocks, but as patterned events, should be more likely to pay the costs of maintaining long-distance relationships. We test these hypotheses in the context of two communities of Bolivian horticulturalists, where climate variability – especially precipitation variability – is relevant to production and access to long-distance connections is improving. METHODS. Data on individuals’ migration histories, social relationships, and other relevant variables were collected in 2017 (n=119). Precipitation data were obtained from the US National Center for Atmospheric Research, allowing us to estimate participants’ exposure to drought and excess precipitation. RESULTS. Exposure duration, proximity in time, and frequency did not predict having a greater number of long-distance relationships. Males, extraverted individuals, and those who had traveled more did have more long-distance relationships, however. CONCLUSION. Another function of long-distance relationships is to access resources that can never be obtained locally; ethnographic data suggest this is their primary function in rural Bolivia. We conclude by refining our predictions about the conditions under which long-distance relationships are likely to help individuals manage the risks posed by climate variability.