Essays in political economy

This dissertation explores the interplay between norms, preferences, and information across three distinct chapters. It investigates how information derived from commemoration, news, or social media interacts with social norms, such as trust and pro-social behavior, and political preferences, including regime support in autocracy and demand for regulation. Chapter 1 explores the transmission of social norms through collective memory in transient communities. Utilizing novel data on online donation ads for small personal items as a proxy for pro-social behavior, the study reveals that individuals are less likely to engage in such behavior when reminded about past repression through commemoration. This emphasizes the crucial role of collective memory in shaping historical legacies, even in transient communities. Examining the demand for state regulation during the COVID-19 pandemic, Chapter 2 challenges existing theories by incorporating fear alongside trust. Analyzing survey data from 61 Russian regions, the study finds that fear of the virus increases demand for regulation. The findings highlight a critical scope condition: the impact of trust on regulation is conditional on fear, with high levels of fear decreasing the effect of trust. This offers insights into how fear of social threats shapes support for state intervention, especially in crises. Concluding the dissertation, Chapter 3 establishes an empirical link between exposure to information on casualties, contrasting war propaganda, and war and regime support in Russia after the full-scale invasion of Ukraine in February 2022. Analyzing the changes in social media engagement in response to verified information on Russian war fatalities in Ukraine, the study reveals that accurate information on the human cost of war disrupts the spread of war propaganda and has the potential to erode support for the autocrat at war. This underscores the broader implications of countering false narratives in the context of independent media and misinformation.