The selection of migrants and returnees in Romania: Evidence and long-run implications

This paper uses census and survey data to identify the wage earning ability and the selection of recent Romanian migrants and returnees on observable characteristics. We construct measures of selection across skill groups and estimate the average and the skill-specific premium for migration and return for three typical destinations of Romanian migrants after 1990. Once we account for migration costs, we find evidence that the selection and sorting of migrants are driven by different returns to skills in countries of destination. Our identification strategy for the effects of work experience abroad permits a cautious causal interpretation of the premium to return migration. This premium increases with migrants' skills and drives the positive selection of returnees relative to non-migrants. Based on the compatibility of the results with rationality in the migration decisions, we simulate a rational-agent model of education, migration and return. Our results suggest that for a source country like Romania relatively high rates of temporary migration might have positive long-run effects on average skills and wages.