Working Paper: CEPR ID: DP15587
Authors: Axel Dreher; Sarah Langlotz; Johannes Matzat; Chris Parsons
Abstract: We provide causal evidence showing that migration increased the polarization of politicians campaigning for the House of Representatives between 1992 and 2016. Our polarization measures derive from ideology data based on 3 million campaign contributions. Our shift-share estimates hold over the medium-run, although they wane over time. These effects are strengthened should counties host similarly educated or more culturally distant migrants. Contributors' race, employment status and occupations play important roles. Our results hold when focusing specifically upon refugees, where we exploit the spatial and temporal variation stemming from the opening of refugee resettlement centers for the sake of causal identification.
Keywords: migration; refugees; polarization; political ideology; united states
JEL Codes: J15; F52; F63
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Greater cultural distance between immigrants and natives (J69) | More pronounced effects of immigration on political attitudes (K37) |
| Similar educational backgrounds (Y80) | More pronounced effects of immigration on political attitudes (K37) |
| Increased migration (F22) | Greater polarization among politicians (D72) |
| 1% increase in the share of new immigrants in a county (J11) | 0.89% increase in difference between extreme and moderate campaign contributions (D79) |
| Increased immigration (K37) | Shift rightward in ideology of winning candidates (D72) |
| Increased immigration (K37) | Higher probability of conservative Republican candidates winning (D79) |
| Increased immigration (K37) | Decrease in probability of liberal Democrats winning (D79) |
| Larger immigration inflows (J69) | Political spectrum shifts to the right (P26) |