Migrants ' Social Ties in the u.S. and Investment in Mexico"

Using a sample of 1,112 heads of household from the Mexican Migration Project (MMP), I develop a multinomiallogit regression to test the hypothesis that migrants with social ties to other migrants in their place ofsettlement will have a greater risk of investing their remittances and savings on a productive activity or a home than of spending them on consumption. I find that migrants who live with family members or townspeople during their last migration trip have a higher risk of investing their savings on a productive activity or a home relative to spending them on consumption. Migrants who belong to a social club with other migrants have a higher risk ofinvesting their remittances on production or housing than ofusing them for consumption. Studying what conditions lead migrants to invest their remittances is of great practical importance because of the enormous sums of money migrants send to their countries of origin, estimated at $75 billion worldwide (Taylor et al. 1996). In 1999 migrants sent $6.8 billion in remittances to Mexico. This sum is greater than all Mexican agricultural exports, almost equals the country's income from tourism, is more than two-thirds the value of oil exports, and is more than 50 % the size of the foreign direct investment in Mexico (Multilateral * I would like to acknowledge the assistance of participants in the Princeton University Department of Sociology's empirical seminar of 1999-2000, especially ProfessorFrank Dobbin. Paul Bern and Dan Edelstein of Princeton's Data and Statistical Services provided crucial programmingsupport. Dana Glei and Aaron Sparrow ofthe Office ofPopulation Researchassisted in refining my models. I would also like to thank Doug Massey for his helpful comments on a