Association for Public Policy Analysis & Management

Return Migration is an intrinsic part of the cycle of international migration and development. Although in recent years, there is growing attention towards border protection, limited research has been conducted on understanding the impact of return migration on the origin country. Some studies have found that return migrants tend to earn higher wages relative to those that never migrated, are more likely to start their own business, and reduce violence and shift social norms in their communities. The US-Mexico corridor represents one of the most massive migrant movements in human history. Since 2007 the trend has reversed with more Mexican migrants returning to their homes than those moving to the US. The experiences gained abroad benefit not just the return migrants but may also cause spillover effects that enhance the living conditions of their entire communities. In this study, we attempt to establish the causal effect of return migration on the education outcomes of the Mexican municipalities. We use various sources to construct a dataset for all 2456 municipalities. Our main dependent variable, education index, is created using the expected and mean years of schooling, following the methodology provided by the United Nations Development Programme (UNDP).

Our principal independent variable, return migration, from the National Population Council (CONAPO), refers to the percentage of households in the municipality with migrants who lived abroad in 2005 and had returned to Mexico before 2010. This variable, in our case, maybe potentially endogenous if migrants return to municipalities that are more developed and have better education outcomes. To address this issue, we use two instruments for return migration: US state-level immigration laws acting as push factors and distance to the border to account for the cost of relocation.

We use the control function approach to assess the causal relationship of return migration on education outcomes. The benefit of using this approach is that when there is no endogeneity, there is no loss of efficiency since the second stage uses the observed return migration variable instead of the predicted values. Our results suggest a significant association between improvement in the education index and return migration in the Mexican municipalities. Specifically, a one percentage point increase in return migration rate is associated with a 0.020-0.039 points increase in education index. We further find that the illiteracy rate, the percentage of the population not attending school, and those with incomplete primary education decrease in municipalities with more return migrants.

We speculate these improvements to be driven by the migrants' experience and accumulation of human capital in the US. Our paper suggests return migration from a developed to a developing country as a mechanism through which migrant flows may benefit origin developing countries around the world. Our results have important policy implications for return migrants in developing countries, especially in Mexico, with debates surrounding the "border." Policies aimed at assisting the reintegration of return migrants in local markets and employing the human, physical, and social capital accumulated from abroad may be substantial in improving the quality of education.