Our main contribution is to separate the effect of high-skilled and low-skilled immigrants, by exploiting the different geography and timing of the inflows of these two groups of immigrants. We find that an increase in the first type of immigrants decreases the share of the Republican vote, while an inflow of the second type increases it. These effects are mainly due to the local impact of immigrants on votes of U.S. citizens and they seem independent of the country of origin of immigrants.They see high-skilled immigrants causing Democratic voting among "U.S. citizens" whereas what I believe is happening is that high-skill immigrants cluster in high tech hubs where tech firms get them H-1b visas. Those places already vote Democrat and continue to do so.
By contrast, an influx of low-skilled immigrants into areas where they directly compete with locals for low-skilled jobs becomes a source of citizen dissatisfaction. That dissatisfaction moves people to vote for Republicans who today are less receptive to low-skilled immigration.
