Do charismatic leaders amplify expectations of reform, and what are the consequences when those expectations are unmet? I study Giuseppe Garibaldi’s 1860 campaign that led to Italian unification. After conquering Sicily, he advanced to the mainland as a liberator and pledged sweeping reforms, but soon ceded power to the monarchy, leaving them unfulfilled. I exploit an unanticipated halt in his planned route to compare municipalities he visited with similar ones he intended to reach but did not. Visited towns initially aligned more with unification and experienced less violent resistance in 1861. Over time, however, they became more politically disengaged: turnout was lower in the 1913 election under universal male suffrage, they sent fewer volunteers to Garibaldi’s corps in 1866, and they were less responsive in 1948 to a party campaigning on his image. Effects were strongest where his redistributive promises were more salient. The results show that charisma acts as an expectations multiplier: it accelerates mobilization but, when delivery fails, magnifies disillusionment, leaving lasting legacies for nation-building, participation, and collective memory.

Working paper available soon!

How do mass media shape gender norms? This paper studies the expansion of local newspapers in Victorian and Edwardian Britain and their impact on women’s labor supply, marriage, fertility, and political mobilization. We link data on the staggered arrival of newspapers across 620 districts to census microdata, petition records, and newspaper text. Using a dynamic difference-in-differences design, we find that, on average, newspaper access reinforced traditional gender roles: female labor force participation declined, marriage accelerated, and fertility rose. Yet these patterns reversed in districts exposed to a major feminist shock—the 1866 suffrage petition for women's suffrage. In these areas, newspapers increased women’s labor market participation, delayed marriage, and reduced fertility. Effects were strongest in places with lower baseline female empowerment. Text analysis reveals that positive framing of the petition independently boosted female labor supply. Moreover, exposure to favorable coverage of the 1866 petition increased local petitioning rates for female political emancipation, suggesting media shaped not only beliefs but collective action. These results show that newspapers acted not just as information sources, but as coordination technologies—amplifying or stalling change depending on timing, content, and narrative salience. Symbolic inclusion can shift social equilibria—but the emancipatory power of media depends on who is visible, when, and how.

Can gender roles and gender equality explain the incidence and brutality of armed conflicts? While qualitative studies have long recognized the impact of unequal gender roles on violence, no empirical study has already explored this relationship. Feminist scholars in international relations have recently made a stand to affirm how the juxtaposition of privileged masculinity and “devalorized” femininity can contribute to explaining the patterns of violence. Indeed, as the sociologist Johan Galtung explains, the mutual relationship between structural and cultural violence helps in understanding how gender-related inequalities can become inescapable rationalizations of war-initiation. This essay aims at quantitatively investigating whether the intra-regional level of gender equality has an impact on the occurrence of internal conflicts and the number of battle-related deaths. I collect fine-grained geographical data on conflict and gender norms from 29 countries of continental Africa. Results indicate that higher gender equality leads to less brutal (but not fewer) conflicts. To establish causality, I exploit the distance to the closest water spring as an instrument for gender roles, by relying on Ester Boserup’s theory of the role of women in human development. Indeed, collecting water is a traditionally female occupation and women have been used to carry heavy loads over long and winding routes, taking away valuable time from income-generating jobs and education. IV estimates confirm that improvements in gender equality reduce the brutality but not the incidence of conflicts: a 1 standard deviation increase in gender equality results in 0.21 s.d. less fatalities in conflicts.