The rise of knowledge and the flow of innovation have long been recognized as important factors for economic development. How do increased spatial connections affect the generation and diffusion of ideas? I study knowledge production in Germany in the 19th century, which witnessed an explosion of innovation. The analysis relies on several novel large-scale datasets, including the universe of bibliographic records, covering all published texts in all fields of knowledge, and detailed and comprehensive railway statistics. New ideas — both novel concepts and new combinations of existing ideas — are measured thanks to recent advances in machine learning and topology applied to data analysis. Exogenous delays in railway construction enable the identification of the effect of the railroad network. I show that the railroad contributed to innovation, explaining 11% of the increase in knowledge production. Scholars’ mobility led to the formation of specialized clusters, and thus to cities’ specialization. New ideas are formed by combining ideas coming from cities connected by the railroad. Diffusion of new ideas increases within fields. The findings shed light on the causes for specialization in knowledge production, on the organization of modern science, and on the diffusion of information in dense networks.

The transition to the knowledge economy has been associated with rising regional inequality: urban areas have been growing disproportionately fast compared to peripheral areas, driven by the co-location of knowledge-intensive businesses and highly educated workers. Yet, the geographical concentration of firms and skilled workers is far from uniform across countries; and while the political consequences of regional disparities have been widely studied, and are often linked to the rise of populism and polarization, we know little about the determinants: why is knowledge-intensive growth and innovation very concentrated in certain countries, and less so in others? We stress the role of public policies in determining variation in the agglomeration of innovation and argue that such policies are in turn driven by political coalitions, mediated by institutions. To test our argument we develop a new measure of spatial concentration of patents across countries, as well as a new measure of government coalitions. We show that centralization and the composition of governing coalitions are closely correlated to the spatial concentration of clusters in the knowledge economy. Our argument points to partisan politics as a previously unexplored determinant of the geography of innovation.

How do states integrate peripheral elites and overcome credible commitment problems? This paper argues that elite marriage networks can serve as an “exchange of hostages,” binding central and peripheral elites through kinship ties. We focus on early modern Venice, where the 1630 plague and the fiscal demands of the War of Crete induced a previously closed nobility to admit wealthy mainland families. Using digitized family-level data on marriages, offices, and public debt holdings, we show that families integrated earlier into the noble marriage network were more likely to invest in public debt; integration is also associated with higher provincial tax revenues, increased military spending, and a stronger central institutional presence.

Social science increasingly relies on historical accounts and advanced text analysis to address pivotal questions of state and welfare-state formation, the emergence and transformation of political ideology, and related developments through political speech. Yet historical data in cleanly structured formats remain difficult to obtain. We introduce KaiserParl, a novel corpus of historical speeches comprising debates in German parliaments from 1867 to 1918. It comprises 239,339 speeches across 4,217 sessions. The corpus covers the period before German national formation through the termination of the Kaiserreich, including pivotal events such as unification, the introduction of the welfare state, and the onset of World War I. KaiserParl was OCRed using software capable of processing historical fonts and includes speech-level attributes familiar from more modern parliamentary corpora. Speeches are linked to speaker name, speaker faction, and to an exact day-date. This enables researchers to study political speech across ideological camps, around relevant historical events, or within individual members of parliament over time.

The idea of “rule by the people” is at the heart of definitions of democracy. How “the people” are defined, however, is always a subject of contention. This paper investigates how the meaning and ideological function of the term Volk changed in German parliamentary discourse between 1867 and 1918. Drawing on a novel corpus of over 250,000 Reichstag speeches, we examine when and how political elites invoked the Volk — whether as a democratic, plural, and inclusive subject or as an ethnic, unitary, and exclusionary construct. We use large language models (LLMs) to analyze the semantic context and ideological framing of each instance of Volk, based on a structured typology of progressive and conservative narratives. We then reveal shifts in how the term was used over time and across parties, shedding light on the contested conceptual development of Volk in a period of intense institutional and ideological transformation. The findings contribute to broader debates about semantic change, political rhetoric, and the ideational foundations of democracy and nationalism in modern Germany.

How did the modern welfare state originate, and how did its origin affect its long-term development and efficiency? The literature has mostly pointed to the strength of the labor movement or social cleavages as determinants of welfare provision. Within the Italian post-unitary context, we take a different approach by highlighting that, in its early stages, welfare was not created anew by the state, but had instead been provided for centuries — by institutions affiliated with the Catholic Church. We study the effect of the incorporation of these institutions into the state apparatus in 1890, linking historical welfare provision to the quality of welfare once it was under state control, by focusing on charities that were previously run by the clergy and that had to switch to private or public administration. We expect the effect of the change in administration to be positive in the regions where state institutions originated, and negative in the regions where state institutions were exported and ineffective at the local level. Using detailed historical records of Church-affiliated charities, their services and budgets, we use a difference-in-differences strategy to study the effect of state incorporation on welfare provision, health and literacy outcomes, institutional resource management, and administrative change.

Legal doctrine is often described as evolving case by case while preserving coherent legal traditions. We introduce a method for identifying and tracing such traditions by applying the Louvain community-detection algorithm to the complete U.S. federal judicial citation graph (1790–2024; 760,000 opinions linked by 8.3 million citations). The resulting citation communities are unsupervised, blind to court labels and topical taxonomies, yet they form meaningful empirical groupings: in validation exercises, they recover known institutional structure and capture coherent substantive and procedural approaches to law. Using these communities as empirical units, we study how legal traditions evolve and respond to legislative change. We show that the introduction of pivotal federal statutes shifts citation patterns toward newer precedent, indicating that legislation can disrupt established traditions and redirect the development of case law.

A prominent contemporary phenomenon is “backsliding” of democratic countries into (semi-)authoritarian practices. Importantly, such episodes unfold over time, and often involve uncertainty about the ultimate intentions of governments. Building on recent work (Svolik 2020), we present a model in which a government engages in a reform that may allow for subsequent actions that are inconsistent with the rule of law. Citizens must decide whether to replace the incumbent following the reform. Consistent with existing work, the model suggests that polarization is an important factor in democratic backsliding. More importantly, the model demonstrates that in a dynamic setting, citizens may support incumbent governments even if citizens are fundamentally opposed to authoritarianism. One consequence is that citizens may genuinely regret their electoral choices. We illustrate the model’s implications using a survey experiment in contemporary Poland.

In 2015, Poland began to gradually dismantle institutions of the rule of law that had been carefully put in place after its governmental transition. Moreover, this process has been underway largely with the electorate’s support. This is puzzling because only several years earlier, the rule of law seemed to be all but guaranteed by the country’s accession to the European Union. As Poland is still a member of the European Union, and the European Union is noticing erosion of the rule of law and sounding the alarm, why are “eroders” still being elected? We propose an explanation based on the dynamic model from a 2024 article by Caterina Chiopris, Monika Nalepa, and Georg Vanberg, which posited that voters were deeply uncertain as to whether the incumbent they were about to reelect was introducing policy change because of his genuine ideological commitments or because he is a closet autocrat, for whom policy change is instrumental to usurping power. What is more, the authors find that there is an interactive relationship between democratic commitments and uncertainty. We tested this theory with an experiment around the 2019 nationwide Polish elections and found that citizens with less exposure to democratic rule were more likely to reelect incumbents making sweeping policy changes when they were uncertain about the true intentions of the incumbent’s policy.

We address ten commonly held myths about the racial wealth gap in the United States. We contend that a number of ideas frequently touted as “solutions” will not make headway in reducing Black–white wealth disparities. These conventional ideas include greater educational attainment, harder work, better financial decisions, and other changes in habits and practices on the part of Blacks. While these steps are not necessarily undesirable, they are wholly inadequate to bridge the racial chasm in wealth.

Severe acute undernutrition (SAU) is still a crucial global health issue in the 0–59 months population, increasing the risk of mortality as well as of long-term consequences. In Sudan, 3.3 million children suffered from acute malnutrition between 2018 and 2019. This study was planned to evaluate, in the area of Port Sudan, the prevalence of acute undernutrition after the COVID-19 pandemic and to identify the most important factors favoring the development of acute undernutrition. The available clinical records of all the under-five children (n = 1012) admitted to the Port Sudan Emergency Pediatric Hospital from February 1, 2021 to January 31, 2022 were analyzed. The presence of wasting and kwashiorkor was assessed and children were categorized according to age, gender, place of residence, main reason for hospitalization, and underlying comorbidities. Acute undernutrition was evidenced in 493 (48.7%) children. Of them, only 16 (3.2%) were diagnosed with kwashiorkor. Children with SAU had a higher prevalence of acute gastroenteritis (p < 0.05) and parasitosis (p < 0.05). Infants aged 0–6 months were those with the lowest risk of undernutrition, whereas those aged 7–12 months were those with the greater risk. In these patients, multivariate analysis revealed that SAU and MAU were 2.5 times (OR 2.51; 95% CI, 1.79–3.55) and 5.5 times (OR 5.56; 95% CI, 2.59–18.7) higher. This study shows that the area of Port Sudan is still suffering from an alarming prevalence of severe wasting and the risk of developing acute undernutrition seems strictly related to the introduction of complementary feeding and tends to reduce with increasing age. Measures already in place to prevent acute malnutrition should be reinforced with improvement of mother education on child feeding.