From Europe to the Americas: A Comparative Panel-Data Analysis of Migration to Argentina, Brazil, and the United States, 1870-1910 by

In this paper we use annual panel data, 1870-1910, to estimate and compare the determinants of European emigration to Argentina, Brazil, and the United States. We employ the Hausman-Taylor Instrumental Variable technique, which provides for cross-sectional (country) and temporal (annual) controls, accounts for variables that are potentially correlated with the unobservables, and allows the recovery of temporally invariant variables. Although the models are estimated separately, they are linked through the inclusion of migration to the other two countries as a measure of the opportunity cost of migrating to any given country. Migration to the South American countries was dominated by Portugal, Spain, and Italy. Especially after 1900, Italy also became an important source of U.S. immigrants. Nevertheless, the determinants of migration to Argentina and Brazil were similar to those for the U.S. in many, but not in all respects. In general, migration was from low-wage source countries, from countries with large stocks of immigrants in the destination who were born in the origin, and from countries with relatively much recent migration to the destination. Certain determinants differ considerably for the three countries. For example, greater rates of employment growth in the source countries compared to the U.S. seriously deterred migration to the U.S., but the same was not true of respective relative growth for Argentina or Brazil. JEL: F22, N10, O50