Are migrants able to use the migration experience to their benefit, that is to improve their livelihoods, and is this result nuanced by whether migrants are poor or non-poor? This paper explores these questions quantitatively using data on migrants and non-migrants from Ghana and Egypt. It describes the main challenges in the empirical literature and introduces a conceptual model to explore the links between migration and poverty. The empirical model accounts for the direct effects of migration on poverty and for the role of migration in moderating the dynamics of poverty. Results show the selectivity of migration with respect to subjective poverty and that migration can have a significant impact on helping people improve their livelihoods. The paper further finds that selectivity with respect to human capital depends on "reasons for migration" and visa status. These findings enrich existing empirical studies by providing a clear estimation of sequential events and enable policymakers to better understand the processes behind migration and poverty.