Project Title: Misperceptions of Group Inequality and Support for Redistribution and Sectarian Politics: Experimental Evidence from Lebanon

Name: Maria Clara Souza Rocha

Program: Laidlaw Scholars Leadership and Research Program 

Faculty Advisor: Augustin Bergeron

Supervisor/Mentor: Pearl Anson Tsang

College: Harvard College

Abstract: Economic inequality has risen worldwide, yet in many societies this has not translated into stronger support for redistributive policies or for political coalitions organized around class or economic interest. Instead, political competition remains structured around religious, ethnic, or cultural lines, even where income and poverty are distributed similarly across those groups. This research investigates whether correcting groups' misperceptions about each other's economic conditions can weaken support for identity-based politics and increase support for redistributive, cross-group policies. Lebanon offers a compelling case: political life is organized around sectarian identity, despite poverty and inequality cutting across religious lines in nearly identical ways. In a representative survey of 3,300 adults in Greater Beirut, respondents were randomly assigned to a treatment group that watched a seven-minute video documenting that income inequality between Lebanon’s main religious groups, Sunni, Christian, and Shia, is small, while overall inequality is high. Using large language models (LLMs), we analyzed the treatment group effects on perceived inequality between religious sects, the presence of solidarity across these groups, and the salience of class identity. We also compared our survey sample to nationally representative benchmark surveys and found that both are closely aligned, supporting the relevance of the findings. So far, it has been found that the treatment can successfully correct beliefs, reduce perceived inequality between sects, and increase support for redistribution. However, no immediate change in political preferences or costly civic behaviors, such as support for civil marriage or a non-traditional political figure, was observed. Current theoretical work on whether correcting those misperceptions can shift political preferences or promote political engagement is limited. This research addresses these gaps, providing a unique contribution that tests this mechanism.