Subnational Variation in Repression: Spatial, Actor, and Issues

• In Event: Repression and Dissent: New Directions in Research

Abstract

What factors explain subnational variation in repression against nonviolent protests? Existing studies have identified more threatening protests as a contributor to government use of repression. Yet, studies have commonly examined variation in repression at the country-year or event level. In contrast, this study considers subnational region spatial relationships. It specifically hypothesizes that repression in a first administrative district (ADM1) is a function of the extent of protests(t-1) in neighboring ADM1’s. The effect of spatially related protests may be conditioned by two variables. First, we expect protests(t-1) in neighboring ADM1’s to differ in threat perception by issue and actor. In particular, democracy-related protests are expected to be especially threatening to the regime. The effect of protests(t-1) in neighboring ADM1’s is expected to be conditional on whether such protests relate to democracy. Second, excluded ethnic groups may also be threatening to regime. The effect of protestst-1 in neighboring ADM1’s is expected to be conditional on whether such protests relate to an excluded ethnic group protest. Using a sample from Africa, the Middle East, and South Asia and Armed Conflict Location & Event Data, we examine the hypotheses with multilevel OLS and logistic regression models. Our results indicate that governments may repress not simply based on the attributes of a protest but on the attributes of neighboring protests, and highlight the spatial, issue, and actor interdependence of protests and repression.

Authors

• Brandon Ives, Seoul National University

• Jacob S. Lewis, Washington State University