Legislative Incumbency and Electoral Repression and Violence

• In Event: State Repression and Human Rights I

Abstract

State repression in the context of elections is a common and particularly pernicious form of repression. According to the NELDA data, one in five elections since 1990 has been characterized by state repression. Electoral repression is perpetrated by many different actors, including party agents, organized criminal associates, formal state security forces such as police or military officers, or party-affiliated militias. Agents of electoral repression are often rewarded with state patronage or forbearance, which is easier to access for incumbents. On the other hand, legislative incumbents may be less reliant on violence because they have access to other tools to win elections. Does holding an additional legislative seat in a district increase or decrease the incidence of electoral repression in that district? We use a close-election regression discontinuity design that combines data from a large proportion of the violent elections since 2010 to test whether legislative incumbents are associated with more or less election violence. We disaggregate the results based on state and non-state perpetrators, and test how the effects of legislative incumbency on electoral repression differs depending on the political context, such as in majoritarian versus proportional systems. An underlying goal of this project is to bridge findings from comparative politics and international relations and explicitly test the generalizability of findings from single-country quantitative case studies by testing our theory across the universe of cases of electoral repression.

Authors

• Lauren E Young, University of California, Davis

• Marika Miner, University of California, Davis