Working Paper: CEPR ID: DP17598
Authors: Veronica Frisancho; Evi Pappa; Chiara Santantonio
Abstract: Every day, three women are murdered in the United States by a current or former partner. Yet policy action to prevent gender-based violence has been limited. Previous studies have highlighted the effect of female political representation on crimes against women in the developing world. This paper investigates whether the election of a female politician reduces the incidence of gender-based violence in the United States. Using a regression discontinuity design on mixed-gender races, we find that the election of a female House Representative leads to a short-lived decline in the prevalence of femicides in her electoral district. The drop in femicides is mainly driven by a deterrence effect that results from higher police responsiveness and effort in solving gender-related crimes.
Keywords: gender-related violence; close elections; police effort; regression discontinuity
JEL Codes: D72; J12; J16
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Female House representative elected (K16) | decline in femicides (J12) |
| Female House representative elected (K16) | increase in reported rapes (K42) |
| Female House representative elected (K16) | increased police responsiveness (J45) |
| increased police responsiveness (J45) | decline in femicides (J12) |
| Female House representative elected (K16) | changes in societal attitudes towards women (J16) |
| changes in societal attitudes towards women (J16) | increase in reported rapes (K42) |