This paper examines whether early 20th century state-level prohibitions on the sale and manufacture of alcohol influenced entrepreneurial responses to the COVID-19 crisis. During the period 1900-1920, approximately 1,600 counties in the United States voluntarily went dry (i.e., they prohibited the sale and manufacture of alcohol through local referenda or local ordinances), while an almost equal number of counties were forced to comply with state-level prohibitions. I argue that being forced to become dry fostered a long-lasting latent culture of adaptation and informal entrepreneurship. Using county-level data from 2010–2023, I find that exposure to forced prohibition is associated with a surge in new business applications =in the aftermath of the COVID-19 shock. This effect appears to be more pronounced in areas with low remote-work feasibility and stronger anti-prohibition sentiment as, for instance, measured by references to bootlegging and speakeasys in historical newspaper accounts and anti-prohibition party vote shares. My results suggest that informal entrepreneurial norms rooted in the resistance to prohibition had an enduring impact on economic adaptability and entrepreneurship.

Natural disasters often shape political environments by emphasizing governance dynamics during crises and influencing voter behavior. Understanding dynamics in competitive authoritarian regimes is important since state control over resources and media can strengthen the political impacts of crises. Using a difference-in-differences and event-study approach and leveraging spatial variation in earthquake intensity, collapsed buildings, and proximity to the epicenter, I analyze the effect of earthquake exposure on incumbent vote share. The findings show a significant increase in the votes received by the incumbent, Recep Tayyip Erdoğan, especially in the areas most affected by the earthquake, despite the many complaints about how the government handled the disaster. I explore potential mechanisms driving this effect, including state-led aid distribution and media control. The findings are robust to alternative exposure measures and regional subsamples. Interestingly, the earthquake increased the incumbent support without affecting voter turnout or the probability of political turnover, suggesting mechanisms such as rally-around-the-flag dynamics and scapegoating as a central role. This research contributes to the literature on crises, voting behavior, and competitive authoritarian regimes by examining one of Turkey’s most catastrophic natural disasters.

Can cultural orientation be at the root of new labor and entrepreneurial trends triggered by the Covid-19 shock in the US? The frontier conditions shaping early US history are known to foster traits such as self-reliance, inventiveness, and individualism, all of which may be influential drivers of changes in labor supply decisions under unexpected circumstances. Using novel measures developed by Bazzi et al. (2020) on the frontier experience of US counties/ states and up-to-date information on job quits and new business applications, I argue that residents in locations with wider frontier experience are more likely to quit their jobs and become self-employed following this health shock. These findings hold even after considering alternative explanations relying on resistance to vaccination mandates and the impact of government transfers during the pandemic. These results contribute to the literature on the long-run effects of culture on economic outcomes by adding a labor supply dimension to the analysis. They also provide a cultural perspective on the emergence of the so-called Great Resignation phenomenon in the US.

This project examines how exposure to warfare during childhood shapes long-run gender norms and attitudes. I compile novel historical datasets on WWII exposure, linked with contemporary survey data on gender attitudes and behaviors. Preliminary analysis suggests that early-life exposure to war correlates with persistent differences in gender role attitudes. Ongoing work focuses on refining exposure measures and testing mechanisms. The historical dataset will be available upon completion of the project.