OBJECTIVE: To evaluate the assumption that moving heightens HIV infection by examining the time-order between migration and HIV infection and investigate differences in HIV infection by migration destination and permanence. METHODS: We employ four waves of longitudinal data (2004-2010) for 4265 men and women from a household-based study in rural Malawi and a follow-up of migrants (2013). Using these data, we examine HIV status prior to migration. Migrants are disaggregated by destination (rural, town, and urban) and duration (return and permanent); all compared with individuals who consistently resided in the rural origin ('nonmigrants'). RESULTS: HIV-positive individuals have significantly greater odds of migration than those who are HIV negative [odds ratio 2.75; 95% confidence interval (CI) 1.89-4.01]. Being HIV positive significantly increases the relative risk (RR) that respondent will be a rural-urban migrant [RR ratio (RRR) 6.28; 95% CI 1.77-22.26), rural-town migrant (RRR 3.62; 95% CI 1.24-10.54), and a rural-rural migrant (RRR 4.09; 95% CI 1.68-9.97), instead of a nonmigrant. Being HIV positive significantly increases the RR that a respondent will move and return to the village of origin (RRR 2.58; 95% CI 1.82-3.66) and become a permanent migrant (RRR 3.21; 95% CI 1.77-5.82) instead of not migrating. CONCLUSION: HIV-positive status has a profound impact on mobility: HIV infection leads to significantly higher mobility through all forms of migration captured in our study. These findings emphasize that migration is more than just an independent risk factor for HIV infection: greater prevalence of HIV among migrants is partly due to selection of HIV-positive individuals into migration.