Working Paper: NBER ID: w27504
Authors: Hani Mansour; Daniel I. Rees; James M. Reeves
Abstract: This is the first study to examine the effect of experiencing a widespread, deadly epidemic on voting behavior. Using data on elections to the U.S House of Representatives and leveraging cross-district variation in HIV/AIDS mortality during the period 1983-1987, we document the effects of the HIV/AIDS epidemic on votes received by Democratic and Republican candidates. Beginning with the 1994 elections, there is a strong, positive association between HIV/AIDS mortality and the vote share received by Democratic candidates. Congressional districts that bore the brunt of the HIV/AIDS epidemic also saw substantial increases in Democratic voter turnout and contributions made to Democratic candidates.
Keywords: HIV/AIDS; voting behavior; political participation; epidemics
JEL Codes: D72; I18
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| HIV/AIDS mortality rate during the treatment period (1983-1987) (I12) | Democratic vote share (D79) |
| HIV/AIDS mortality rate during the treatment period (1983-1987) (I12) | Republican vote share (D72) |
| HIV/AIDS mortality rate during the treatment period (1983-1987) (I12) | Democratic votes per 100,000 voting-age population (K16) |
| HIV/AIDS mortality rate during the treatment period (1983-1987) (I12) | Republican votes per 100,000 voting-age population (D72) |
| HIV/AIDS mortality rate during the treatment period (1983-1987) (I12) | Democratic vote share (by the 2000 elections) (D72) |
| HIV/AIDS mortality rate during the treatment period (1983-1987) (I12) | Republican vote share (by the 2000 elections) (D72) |