Despite recent political backlash in Europe and the United States, international migration remains a formidable force with wide-ranging consequences in destination and origin countries. The presence of a migrant family member living in a developed country could have multifaceted implications for the welfare of the left behind. In our recent paper we attempt to examine the impacts of international migration on the education of family members left behind in Nigeria. Specifically, we study the net effects of the presence of a migrant family member living in a foreign country on the educational attainment of family members back at home at both secondary and post-secondary levels. We then examine the role of the prospect of future migration, inspired by the presence of a family member abroad, as a potential channel mediating the effect of emigration on family education. By conducting such analysis for Nigeria, the most populous country on the African continent, we hope to shed light on the human capital implications of migration in one of the poorest regions in the world.

The link between emigration and education is often viewed from the point of the view of the emigrant without much consideration for spill-over effects on those left behind. That is why the issue of migration from developing countries tends to raise the spectre of brain drain. This is a legitimate concern given the selectivity involved in attracting skilled migrants from developing countries that do not share borders with developed countries. However, this concern has been countered with the argument that the prospect of migration may lead to a net increase in the stock of human capital of the origin country by encouraging more individuals than who will eventually emigrate to invest in education. In this regard, the potentially positive impact of migration on the education of the left behind might not be limited to household wellbeing. It could also extend to improving aggregate human capital despite initial loss through brain drain.

The presence of a migrant family member in a foreign, and usually, more developed country could influence the education of the left behind through a number of potential channels. The first potential channel is remittances. Remittances may help to relax the credit constraint that is often behind underinvestment in education in many developing countries. As a second potential channel, the absence of family members from home may affect education of the left behind negatively by depriving them of proper guidance and role models or burdening them with extra household responsibilities. The third potential channel is the improved probability of future migration of the left behind. The argument for this channel rests on two assumptions. First, the presence of family members and other social networks in destination countries plays an important role in encouraging new emigration to those destinations. Second, on average, there is higher return to education in destination countries than in origin countries.

In our recent paper, the primary objective was identifying the net effect of the emigration of a family member on education that may have been transmitted through a variety of channels. But, as a
secondary objective, we also test the effect of future own emigration as a channel. We focus on the probability of future emigration as a potential channel for two reasons. First, the prospect of future emigration is deemed to affect the expected returns to education more directly than competing channels such as remittances which exert influence through the budget constraint. Second, as far as our geographical focus is concerned, more migrants from African countries move to the US and the European Union through the help of family reunification programmes than any other means, lending credence to the importance of family network for migration.

Although the key propositions we test are rooted in human capital theory, the scope of the study was mainly empirical. We used completion of secondary school and attendance of postsecondary education as alternative outcome variables. Completion of secondary schooling has become an important indicator of social progress in developing countries since the expansion of primary schooling in most of these places in recent years has led to churning out of students eligible for secondary education. Moreover, secondary education is arguably the minimum requirement for most migrants to be able to cope up with life and work in foreign countries. For example, the United States Diversity Visa programme that grants 50,000 permanent residence visas annually via lottery requires winners to have attained a minimum of 12 years for formal schooling. In this regard, we expect that migration of a family member has a positive impact on completion of secondary education by the left behind. Postsecondary education, on the other hand, may involve more strategic considerations than secondary schooling. For instance, people might defer investing in postsecondary education in their home countries if they expect to migrate to another country in the near future. But, it could also be the case that postsecondary education increases both the chance of migration and the return to education in prospective destination countries. Thus, the effect of migration on postsecondary education depends on a number of factors including timing, perceived probability of migration and the comparative quality and cost of education in origin and destination countries.

We employ household and individual level data from the migration and remittances survey conducted by the World Bank in 2009 to do the empirical analysis. Making causal inference regarding the link between migration and education requires identifying an exogenous source of variation for migration. We use proportion of international migrants form the same town as the respondent and distance to a foreign mission/church in 1921 as instruments for the presence of a migrant family member. Accordingly, we use two-stage least squares estimation to measure the net effect of migration on the education of the left behind. Once we have estimated the main coefficients of interest, we then test the relevance of the prospect of future migration of the left behind as a channel. This is done by estimating the predicted probability of migration as a function of the presence of a migrant family member. We then estimate the effect of the predicted probability of future migration on educational attainment.

Firstly, we find a positive and significant impact of having a migrant member of the household on the probability of completing secondary school, and attending post-secondary school. This result is robust to various specifications and estimations techniques and is validated by using various exogenous measures of migrant networks as instruments. Secondly, we find that being in a migrant household increases the
probability of own future migration. Finally, we find that the probability of own future migration is positively correlated with the probability of completing secondary school or attending post-secondary school. Our results help understand the dynamics and channels through which migration influences human capital development of those left behind.