سیاست های انتخابی مهاجرت، انباشت سرمایه انسانی و مدت زمان مهاجرت در افق بی نهایت

An increasing literature fosters selective immigration policies as a tool to increase human capital in both source and destination countries. These policies are supposed to prompt incentives to education, and–if selection is sufficiently severe–to increase the human capital stock in source countries. Nonetheless, when compared to open migration, selective policies make returns to education uncertain, and they may harm incentives to invest in human capital. As a consequence, they may reduce the human capital stock even though selection is “severe”. Moreover, when repeated migration is possible, they backfire on migration duration. We obtain our results in an infinite-horizon model that, unlike the current literature, places no restriction on the number of possible migration spells and allows for the possibility of a forced emigration.

Unlike flows of capital or goods, inflows of immigrants can generate frictions with natives and xenophobia, particularly when combined with high unemployment. National governments are, therefore, highly concerned about immigration control. The eastward enlargement of the European Union is going to add approximately 50 million people to the existing labor force. Large and persistent wage differentials support the incentives for extensive mass migration from low-wage, densely populated countries, to the developed world (Lundborg and Segerstrom, 2002). After these events, the governments of destination countries are urged to raise entry barriers. As reported by OECD, 1999 and OECD, 2001 and by Boeri and Brucker (2005), in recent years many countries have modified their entry regulations to reinforce border controls and restrict the entry, residence and work requirements. These barriers are increasingly taking the form of selective immigration policies based on human capital requirements, and they are renewing the concerns for a brain drain.1 However, the recent “new brain drain economics” (Mountford, 1997 and Stark and Wang, 2002) argues that selective immigration policies may benefit the source countries if selection is sufficiently severe, because they incentivate human capital accumulation and they restrain its outflow at the same time. A common assumption of these contributions is that only permanent emigration is considered. A well-developed literature proves that this assumption is too restrictive (see, for example, Dustmann (2003)). Moreover, most models of temporary emigration suffer from other simplifying assumptions: the return decision is permanent,2 and emigration is always voluntary. From a theoretical perspective, it is not evident why migration should be limited to a single spell. There exists, indeed, clear evidence that even at the end of the 19th century repeated migration spells of 3–4 years were not uncommon (Baines, 1991).3 As for the importance of constrained migration, OECD (2001) emphasizes the role of regional conflicts in increasing population outflows. Emigration waves from Europe in the 1840s were associated with famine and revolutions. Bonifazi and Strozza (2002) describe the huge population relocation that occurred after World War II. We refer to Chiswick and Hatton (2003) for further details on the effects of wars in Africa, Vietnam, Laos, and Cambodia, and the disintegration of Yugoslavia. In spite of its importance, forced emigration is not present in current models of migration duration. Authors such as Kossoudji (1992) and Faini (1996) stress that entry restrictions alter the optimal length of immigration spells, even though they do not develop a formal model.4 Since selective immigration policies are nothing but human-capital based entry restrictions, they tie up the decisions about human capital accumulation and migration duration. In our model we are able to account for the joint determination of these variables. Moreover, we try to overcome the other drawbacks of the literature by placing no restriction on the number of migration spells and by allowing for the possibility of constrained emigration. To meet these requisites, we develop an infinite horizon model including a shock that may force individuals to migrate. Our findings show that selective immigration policies may hinder the individual incentives to human capital accumulation (“brain loss”). The intuition behind our results is that these policies make returns to human capital uncertain with respect to open migration. As a consequence, the claim that selective policies help to build human capital can easily be reversed. Similar concerns are present in Schiff (2005), who lists several reasons to explain why the beneficial effects of the brain gain is overestimated, and presents some empirical evidence for the prevalence of the brain drain. Finally, our results question the consistency of restrictive immigration policies with the objective of reducing the immigrants’ stock: entry closure biases the incentives towards longer migration spells and they increase the share of permanent migrants.5 Our paper is organised as follows. The following subsection reviews some main findings in the literature. The model is developed in Sections 2 and 3. In Section 4 we discuss our results, and a sensitivity analysis is used to illustrate our findings in Section 5. Section 6 contains a comparison of our results to those present in the literature. Conclusions are reported in Section 7. The proofs are gathered in the Appendix. 1.1. Related literature The “new brain drain economics” emphasizes the possible benefits of a brain drain. In an overlapping generations framework there are several mechanisms able to generate a beneficial brain drain, and they rely basically on the existence of externalities on human capital: Vidal (1998) points to enhanced intergenerational transmission of skills and education; Mountford (1997) and Beine et al. (2001) stress the possibility of intergenerational spillovers between skilled workers. The possibility of migration increases the expected returns to human capital, and thus the incentive to education. Stark et al. (1997) distinguish between education and ability: productivity depends on the latter. With asymmetric information about the worker’s ability, the incentive to invest in education and migrating is even stronger for low-ability individuals; however, after their ability (i.e. productivity) is observed, they will find it convenient to return. Stark and Wang (2002) use a static model to state some conditions under which a restrictive immigration policy in the destination country increases the welfare of the sending country: the idea is that entry rationing in developed countries can keep most human capital from flying abroad. Schiff (2005) questions the approach of the “new brain drain” literature, and argues that the benefits of a brain gain are overestimated. He gives a simple example of his idea: since in a steady state all variables must be constant, this has to hold for the number of educated individuals as well. Thus, after emigration, no net brain gain can exist. The results in Schiff (2005) rely on three assumptions: (1) human capital exists in a fixed amount, and individuals choose only whether or not to be educated; (2) entry abroad is based on quotas; (3) emigration is permanent. When education is treated as a continuous variable, it may well be that the number of educated individuals is constant, but the level of their education is higher. Moreover, selective immigration policies are increasingly based on human capital requirements, rather than on quotas. In such a case, it is not true that the probability of entry decreases as the number of skilled workers increases. Finally, overlooking the possibility of return migration may in its turn overestimate the real extent of the brain drain. The literature on return migration adopts mainly life-cycle models. An early contribution to the study of migration duration is in Djajic and Milbourne (1988). They develop a two-period model to study the effect of wage differentials in determining migration flows and their final effect on the equilibrium wages, but they claim that more research is needed to understand why “some migrants make several trips, some stay longer than others, and some never return”. Hill (1987) stresses, interestingly, the importance of “the repetitive character of contemporary labor migration”; in spite of that, his assumption of an identical duration for each migration spell can be deceptive. Temporary migration is studied in several contributions (see, for example, Dustmann, 1997 and Dustmann, 2003).

The study of migration duration is receiving increasing attention, as well as the effect of the chances of migrating on human capital accumulation. In our work, we have attempted to connect these streams of literature. Our findings question the effectiveness of human-capital based selective policies. First, we reproduce in a more general and rigorous framework the result that border closure backfires on the optimal migration duration. Second, we stress the impact of crises on the decision to accumulate human capital and to stay abroad longer. Third, we conclude that selective policies are a double-edged weapon: they can both foster and harm the equilibrium level of human capital. With respect to open immigration, they make returns to human capital uncertain, and they can have ambiguous effects on education, by causing a “brain loss”, rather than a brain gain. This result suggests that selective immigration policies should not be used unconditionally, and that the logic behind the “brain gain with a brain drain” is not always correct. The policy implications of this finding are quite important. When the sign of View the MathML source∂h∗∂Ψ is different for different individuals, selective policies can increase human capital dispersion within a country. When the sign of View the MathML source∂h∗∂Ψ is different for different source countries, selective policies can increase cross-country human capital dispersion. This can have worrisome effects on the possibility of convergence in the long run. Finally, it is worth recalling that economic policies can affect pp: plans aimed to reduce macroeconomic risk in developing countries reinforce the incentives to return. Though it may be difficult to modify these characteristics of an economy, there are no theoretical reasons why international, coordinated development policies should be less effective or more costly than enforcing strict border closure.