Research in regional and labor economics has established that economic incentives play a significant role in the process of internal migration. The most common approach is to view migration as a form of human capital investment undertaken by individuals who expect to benefit from the standpoint of increased earnings. One of the central concepts in these models is self-selection. Individuals who self-select the option of migration tend to differ from the nonmigrant population in ways that are not measured in most data sets. The contribution of this paper is in its distinction between two aspects of migrant selection. On one hand, some workers possess unmeasured traits that might simultaneously affect their wages and their propensity to engage in risky human capital investment such as migration. On the other hand, measured earnings might exert a direct effect on migration. Based on samples of employed Swedish males and females at two points in time, this study seeks first to examine whether migration between the two periods occurs in the presence of correlation between unmeasured factors present in both earnings during the first period and the subsequent decision to migrate. Second, it looks for an explicit role of earnings per se in the migration decision. Results of the study provide support for selection based on unmeasured traits for both genders. For females, there is also evidence of selection based on measured earnings. © 2008 Wiley Periodicals, Inc.