Abstract:

The impact of immigration on redistribution policies has been a major topic of interest in political economics. However, in the previous literature, assuming migration level as exogenous, both theoretical prediction and empirical investigations have yield mixed results. This study explicitly takes into account the potential reverse causality between immigration level and redistribution and evaluates the impact of immigration on state income tax rate, using a constructed balanced panel dataset of 50 states in the U.S. from 1994 to 2005. Two-stage lease squares (2SLS) method coupled with fixed effect regression is conducted. The results show that there is a time lag in the median voters’ response to immigration: while the immigration level in the preceding year is negatively associated with the tax rate, an increase in the current share of immigrant in the population in contrast leads to a concomitant increase in the tax burden.