Published online by Cambridge University Press: 22 September 2015
Conventional wisdom recognizes the prevalence of intergroup clashes during political transition. Most explanations of ethnic riots, however, are based on clashes in mature democracies, and are therefore silent on the dynamics at work during democratic transition. Using district-level data in Indonesia from 1990 through 2005, this article argues that riots tend to occur in ethnically divided districts with low electoral competition because uncompetitiveness in the first democratic elections signals continued regime entrenchment and local political exclusion. As such, riots often follow uncompetitive elections, and dissipate after elections become more competitive and opposition candidates secure electoral victory.
Yale-NUS College, Singapore (email: risa.toha@yale-nus.edu.sg). I would like to thank Ashutosh Varshney, Patrick Barron, Sana Jaffrey and Blair Palmer for sharing the UNSFIR data. This research has been supported by the Mustard Seed Foundation, UCLA Institute for Social Research and the Pacific Rim Research Grant. A visiting fellowship at APARC, Stanford University, allowed the completion of this manuscript. Stephen Worthington at IQSS at Harvard University provided excellent technical help. My special thanks go to Merry Alianti, for diligent research assistance. I am grateful to Barbara Geddes, Michael Ross, Daniel Posner, Daniel Treisman, Andreas Wimmer, Don Emmerson, Michael Buehler, Dan Slater, Tom Pepinsky, Eddie Malesky, Allen Hicken, Joel Selway, S.P. Harish, and participants of seminars at Northern Illinois University, UCSD IRPS, SEAREG, Ash Center for Democratic Governance and Innovation at Harvard Kennedy School, the editor and four anonymous reviewers at the British Journal of Political Science for their comments. All errors are mine. Online appendices are available at http://dx.doi.org/doi:10.1017/S0007123415000423 and replication files are stored at http://dataverse.harvard.edu/dataverse/BJPolS.