1 October 2025, 15:15 CET
Seminar Room 1.204
Katharina Werner
Universität zu Köln
In the context of civil conflicts, particularly those rooted in ethnic or religious divisions, effective post-conflict reconciliation strategies are critical for restoring social cohesion and mitigating the risk of recurrent violence. While existing literature affirms that intergroup contact can promote cooperation and trust, the precise mechanisms driving these outcomes remain largely underexplored. Moreover, the effectiveness of digital contact has not been empirically established yet, but digital contact offers a cheap and scalable alternative to faceto-face contact. To address this gap, we conduct a controlled experiment in a post-conflict setting in Indonesia, designed to isolate the effects of different forms of digital intergroup contact. Participants are exposed to one of three treatments: (1) free communication, (2) conflict-focused discussion, and (3) collaboration on a joint project, with a baseline condition of no communication. We measure intergroup cooperation using a multilevel public goods game and out-group trust through a trust elicitation task. Our results show that free communication is the most effective in enhancing intergroup cooperation and out-group trust, consistently outperforming the other treatments. In contrast, conflict-focused discussion and joint project collaboration show weaker and more variable effects. These findings challenge the traditional contact theory assumption that a common goal is necessary for effective interactions. Sentiment analysis and machine-learning evaluation of communication protocols suggest that free communication generates more positive emotional responses, which may underlie its effectiveness. This study contributes to the understanding of the micro-foundations of intergroup contact and offers valuable insights for designing evidence-based policies aimed at post-conflict social and economic recovery.
Galina Vysotskaya
Universidade Nova de Lisboa - Nova School of Business and Economics
This paper studies how the salience of ethnic conflict changes ethnic identities and social capital by focusing on the universe of ethnic-related conflict events in SubSaharan Africa between 2005 and 2023. I match ACLED data on ethnic violence events with Afrobarometer survey data on social cohesion outcomes, such as ethnic identity and trust. I compare respondents interviewed after a violent event that involves their ethnic group with respondents of the same ethnic group interviewed before the violent event, using a high-dimensional fixed effects approach. I find that individuals interviewed in the days after an ethnic-related conflict event have their ethnic identification strengthened; as for social capital, while generalized trust increases as conflict becomes more prominent, trust in known individuals does not change. Shedding light on the mechanisms behind the findings, I show that the direction of the effect on ethnic identification depends on being co-ethnic with the group perpetrating violence or with the group being victimized, which suggests further amplification of ethnic conflict.
Andrea Caflisch
International Organization for Migration
Post-conflict recovery programs often seek to restore economic activity and rebuild fractured social ties. Yet interventions that promote economic revitalization can also generate unequal gains, potentially exacerbating tensions in fragile environments. We study a high-value capital grant program for nascent entrepreneurs in Iraq implemented as part of an effort to support recovery in communities affected by conflict. To assess how such programming affects not only beneficiaries but also social dynamics with indirectly exposed community members, we embedded a randomized controlled trial into a business creation grant program administered by the International Organization for Migration. Over the 2024–25 program cycle, we experimentally assigned 1,087 grants across 2,132 applicants in 18 communities recovering from violence and displacement. To capture indirect exposure effects, we use a pre-treatment referral sampling design to enroll non-applicants (N = 2,981) with social or professional ties to applicants. We present pre-registered analysis of economic outcomes at six-months and seven conceptually distinct dimensions of social cohesion: competition, grievances, interdependence, contact, civic engagement, trust, and inclusive attitudes.