| US-educated Indian engineers played a major role in the establishment of the "Silicon Valley of Asia" in Bangalore. The experience of India and other countries shows that returning well-educated emigrants, despite their small numbers, can make a difference. The first part of this dissertation builds a model of "local" knowledge spillovers, in which migration of a small number of highly skilled individuals greatly affects country-level human capital accumulation. All economic activity occurs in pairs of individuals randomly matched to each other. Each pair produces the consumption good; the skills of the two partners are complementary. At the same time, the less skilled partner increases human capital by learning from the more skilled colleague. With poor institutions at home, highly skilled individuals leave the country seeking better opportunities abroad. On the contrary, improved institutions foster return migration of emigrants who have acquired more knowledge while abroad. These return migrants greatly amplify the positive effect of better institutions.;In the second part of the dissertation, I empirically analyze the propensity of US immigrants to return. Today, little is known about the returnees: who are they and how do they compare to those who did not return? How does their decision to return depend on economic situation at home? To identify return migration, I use the method adopted from Van Hook et.al. (2006). The method is based the U.S. Current Population Survey (CPS) which interviews households for two consecutive years. About a quarter of foreign-born individuals drop out of the sample between the first and the second years, due to various causes including return migration. After eliminating all other causes of dropout, I estimate the propensity of immigrants to return, depending on personal and home country characteristics. I find that the difference between recent immigrants and other immigrants is greater than the difference between men and women, or skilled and unskilled migrants. Thus, assimilation differentiates immigrants more in their decision to return than education or gender. In particular, distance to home country negatively affects return propensity of those who arrived over 10 years ago, and has no effect on recent immigrants. |
