This project analyses two main research questions on the long-term consequences of immigration.
The first question is whether historical immigration affects attitudes of natives towards immigrants, measured as voting for parties supporting anti-immigration policies in recent elections.
The second question addresses whether historical immigration can affect the ability of immigrants to assimilate into the native’s society. Specifically, the aim is to analyze whether in locations with more historical migration recent immigrants tend to assimilate better, because of the existence of a more multicultural society. In this project, the focus is on economic assimilation.
Moreover, the mechanisms through which historical immigration affects the outcomes of interest are analyzed.
The setting is the United States. In this context, the historical migration wave considered corresponds to the age of mass migration in the United States, which happened between 1850 and 1914.
The methodology makes use of recently-developed machine learning methods for causal inference, which are particularly helpful in settings with a potentially large number of (possibly nonlinear) confounders, and to explore heterogeneous treatment effects.
The United States is a suitable setting to study the proposed research questions, as it experienced a large historical migration wave. However, the results of this project are informative for other countries that experience large inflows of immigrants, as many of the countries in the European Union. The analysis of mechanisms is especially important for external validity. By highlighting which variables drive the effect of historical immigration on the outcomes, this research is instructive on whether other countries can expect similar effects to those found in this setting.
