Political Parties and the State in Civil Wars

Locations of security force fatalities during the Civil War in Punjab.

Abstract

Do democratically-elected legislators from nationally-dominant parties reduce ethnic separatist conflict? Existing theory remains unclear. Such legislators might lack credibility among insurgent sympathizers, limiting their ability to mitigate ethnic separatist insurgency in their constituencies, or their symbolic affiliation with ``outsiders'' could incite more violence. Alternatively, state representatives from parties with pan-national presence may benefit from superior influence over the security bureaucracy, rendering them more effective against insurgents than legislators from other parties. We test these competing explanations using a regression discontinuity design, and granular conflict data from a decades-long separatist insurgency in the Indian state of Punjab. Findings suggest that representatives of the nationally-dominant party durably reduced insurgent violence within their legislative constituencies. Mechanism analyses indicate that reductions occurred in constituencies with above-average information and communication technology access and through relatively stronger influence over security forces.

Publication
Under Review
Aidan Milliff
Aidan Milliff
Assistant Professor