This thesis consists of three chapters. While the chapters address distinct questions using different historical settings, and as such should be read as independent pieces of research, some commonalities can be identified. The chapters share the overarching themes of democratisation and the functioning of early democracies. In their combination of historical data and modern econometric techniques, the three chapters additionally have a common methodological foundation. Finally, though these chapters are works of economic history, it has been my goal throughout to not lose sight of questions of interest within economics more broadly conceived. In the first chapter, I examine how interaction costs shape the diffusion of social movements. During the thirty-year period 1881-1910, Swedish society underwent two transformative developments: the large-scale roll-out of a national railway network and the nascence of social movements which came to dominate social and political spheres well into the twentieth century. Using event-study and instrumental variables methodologies, I document the causal impact of proximity to the newly constructed railway network on the spread and growth of membership in these social movements. Well-connected municipalities were more likely to host a local movement organisation and saw more rapid growth of membership per capita. By reducing least-cost distances between municipalities, railways intensified the influence exerted by neighbouring concentrations of membership, thereby enabling spatial diffusion. What determines the efficacy of identity-based propaganda, and how long-lasting are its effects? In the second chapter, I shed light on these questions by studying the impact of the Swedish State Institute for Race Biology’s popularisation of race biology on right-wing ideology in the short and long run. In a popular book edition of its systematic classification of the Swedish population according to “Nordic purity”, the Institute identified particularly “pure” areas of the country. Implementing a differences-in-differences strategy, I document the effect of the publication on rightwing ideology: following the publication, election districts of above-median “purity” exhibit a 3.4 percentage point relative increase in the vote share of right-wing parties. This effect is concentrated in areas with little immigration, suggesting that insular communities may be particularly susceptible to this type of racial rhetoric. Finally, the rightward turn appears to persist over time: present-day municipalities in formerly above-median “purity” regions exhibit a higher relative vote share for the Sweden Democrats, a populist party with roots in the extreme right. In the third chapter, which is joint work with Sascha O. Becker, Andreas Ferrara and Luigi Pascali, we study the effect of warfare on the development of state capacity and representative institutions using novel data on cities and territories in the German lands between 1200 and 1750. More specifically, we show that cities with a higher conflict exposure establish more sophisticated tax systems, and also develop larger councils, councils that are more likely to be elected by citizens, and more likely to be independent of other local institutions. These results are consistent with the idea of a trade-off between more efficient taxation and power sharing proposed in earlier work. We make headway on establishing a causal role of wars by using changes to German nobles’ positions within the European nobility network to instrument for conflict.