Classic migration theory predicts that individual and household migration decisions are partially responsive to economic pushes from origin communities and pulls from destinations. Recent theorizing argues that this basic relationship is fundamentally influenced by the experiences accumulated within migrant streams, connecting potential migrants with future migrants between origin and destination. Drawing upon a 16-year study of migrant departures and returns from 22 villages in northeastern Thailand, we extend current knowledge about these fundamental relationships before, during, and after Thailand's economic crisis of 1997. We answer the following questions: How are migrant departures from the origin affected by the crisis, how are migrant returns to origin communities affected by the crisis, and how do migrants' accumulated experiences connecting origin and destination moderate these relationships? We examine effects separately for men and women since village and destination economies are sufficiently sex differentiated. We find that migrant selectivity partially explains year effects: that is, earlier periods are more highly selective. Migrant cumulative experiences facilitate migration throughout the time period and modestly influence the migration decisions during economic downturns, but these effects are far more important for women than for men. For return migration, year effects emerge only for the post 1997-98 period and only after controlling for migrant social capital and occupational sector. Origin-based migrant social capital slightly, but significantly, reduces the odds of return migration throughout the period of observation. However, migrant social capital does amplify the likelihood of return migration after the Asian Financial Crisis. Construction workers are the most likely to return to their origin villages after the Asian Financial Crisis, while manufacturing, service, and agricultural workers show little change in behavior.