Group Violence, Ethnic Diversity, and Citizen Participation: Evidence from Indonesia

This paper addresses the impact of violent conflict on social capital, as measured by citizen participation in community groups defined for four activity types: governance, social service, infrastructure development and risk-sharing. Combining household panel data from Indonesia with conflict event information, we find an overall decrease in citizen contributions in districts affected by group violence in the early post-Suharto transition period. However, participation in communities with a high degree of ethnic polarization is less strongly affected and even stimulated for local governance and risk-sharing activities. Moreover, individual engagement appears to be dependent on the involvement of other members from the own ethnic group, which points to emphases on bonding social networks in the presence of violence. Finally, in conflict regions, the wealthier households are more likely to engage into cooperative and infrastructure improvement activities, while they are dropping from security groups. On the contrary, the poorest households get more involved in social service activities and less in infrastructure groups. Our results illustrate the danger of generalizations when dealing with violence impact on community activities. We found a large variety of responses depending on the considered activity and its expected economic or social function. We also found large observed and unobserved individual heterogeneities of the effect of violent conflict on activity participation. Once an appropriate nomenclature of activities is used and intensive controls for observed and unobserved heterogeneity are performed, we found that some activities can actually be stimulated by conflict situations. In this respect, the ethnic configuration of society seems to be central in understanding this type of social capital building.