Can organizations that sustain collusion also promote growth? We study German industrialization, when the country reached the technological frontier and legally enforceable cartels came to organize more than half of industrial output. We link the universe of cartel contracts and court rulings on their enforceability to annual data on all joint-stock firms, input–output linkages, and censuses before 1933. Our model formalizes a trade-off between the static losses from market power and the investment gains that cartel rents may support under financial frictions. Vertical contracts between cartels and their buyers can reduce these losses. To estimate these effects, we use staggered cartel entry and unexpected changes in cartel enforceability. Cartel members earned more, invested more, upgraded, and scaled up, especially in capital-intensive industries. Upstream cartelization initially lowered downstream performance, but this effect diminished as vertical contracts, facilitated by banks and the state, expanded. Districts with more cartel members recorded higher employment, while markets with a higher cartel share saw fewer entrants and greater concentration. The results suggest that organizations sustaining market power can finance investment when financial markets are incomplete; the cost is a market structure that favors incumbents over entrants.

This paper studies the role of fiscal capacity in European state consolidation. Our analysis is organized around novel data on the territories and cities of the Holy Roman Empire in the early modern period. Territories implementing an early fiscal reform were more likely to survive, increased in size, and achieved a more compact extent. We provide evidence for the causal interpretation of these results and show key mechanisms: revenues, military investments, and marriage success. The imposition of Imperial taxes, quasi-random in timing and size, increased the benefits of an efficient tax administration on the side of rulers, driving the implementation of fiscal centralization. Within territories, Chambers became the dominant administrative institution, tilting the consolidating states toward absolutism.

We examine how the Thirty Years’ War (1618–48), the largest conflict in pre-modern Europe, gave rise to absolutism. We use planned troop movements from secret military communications to estimate the effect of town-level troop presence on the expansion of ruler-controlled fiscal and military capacity and the dismantling of parliaments. During the war, troop supply demands raised the value of delegating resource mobilization to rulers. After the war, rulers used this capacity to form coalitions with landed elites and weaken parliaments. Courts and city leagues limited parliamentary decline, consistent with lower contracting frictions and stronger local outside options. With parliaments eliminated, militarized absolutist regimes persisted for centuries. Our findings highlight a dynamic trade-off between dictatorship and disorder during states of emergency.

How does culture respond to institutional change? We study the collapse of the Holy Roman Empire (1789–1815), when half of Central Europe changed rulers. Using 15 million birth records from hundreds of cities between 1700 and 1850, we measure cultural traits in real time. Cities that experienced ruler change saw greater naming turnover, dispersion, and novelty. We construct control groups using diplomatic records to isolate these effects, which emerged immediately and persisted. The collapse of hegemonic authority weakened state-aligned identities while strengthening religiosity and nationalism. These shifts undermined subsequent state building, highlighting challenges of ideological integration after regime change.

We digitize and analyze the near-universe of National Socialist German Workers’ Party (NSDAP) membership records and link them to population and industrial censuses. We use these data to document the geography of party membership, the selection of members, and the dynamics of entry. We find that ninety percent of the variation in membership arises between neighboring villages within counties, with one-quarter recording no members. Early party members were socially selected, but these differences disappeared as the party grew. SS members remained distinct: younger, more educated, and more likely to exhibit Nazi insignia in membership portraits. Entry occurred in discrete waves, with later members disproportionately following earlier members from the same communities, workplaces, and families. These patterns suggest that social interaction and coordination transformed a selective political organization into a mass movement. Consistent with this interpretation, while selection of members diminished over time, local membership levels diverged after the arrival of an early joiner across otherwise similar matched municipalities. Finally, this variation in party membership is associated with subsequent deportations of Germany’s Jews.

Many states exhibit high degrees of capacity without the fiscal resources necessary to fund a modern bureaucracy. We argue that they achieve this by exploiting features of the social structure of the societies they govern to motivate individuals to engage in bureaucratic and governance tasks without pay. We develop and illustrate the concept of the “Embedded State” using a unique survey of British urban government from 1835. Since British local authorities had few resources, only two-thirds of positions were paid. We first show that unpaid positions were significantly more productive than paid ones. We then show that unpaid positions conveyed prestige and were ‘stepping stone’ positions, provided different on-the-job incentives, and were taken up by the socio-economic elite. We also show that the successful Embedded State featured patronage and corruption and could not fully motivate unpaid bureaucrats to implement onerous tasks.

This note represents a progress report on the digitization of city directories in Munich, 1845–1914. With information on names, occupations, and residential locations, directories provide a valuable source of microdata especially in a context where individual-level census records did not survive. We construct linked, complete-count data of over one million household-years based on 15 directories. Additionally, we precisely geolocate all historical addresses, drawing on a wide range of supplementary data. We discuss the construction of this dataset and present a novel approach to classify occupational standing based on the noun components of occupation titles. Finally, we show a series of descriptive findings on city growth, spatial inequality and social mobility that shed light on city development in one of the largest and fastest-growing, industrializing urban centers of Central Europe during the 19th and early 20th centuries. We view our approach as a pilot study towards the use of city directories as a source of rich individual-level microdata to study the economic and social history of cities.

The Nazi dictatorship brought German industry under political command: after 1933, the state gained the power to impose cartels and compel membership, and by 1944, 94 percent of industrial joint-stock firms recorded membership in such a compulsory industrial group. We document this transformation in an annual panel of 27,949 joint-stock firms matched to 5,651 cartel contracts. Linking the panel to internal government records from the Reich Chancellery, we trace the transfer of authority over industrial organization from firms to the state and recover ties between firms and officials. We digitize records on camps, production sites, and associated firms, including firms listed under cover names, and connect changes in industrial organization to the allocation of forced labor. We test whether the structures that once coordinated private industry became instruments of wartime mobilization.

How did globalization shift private market risks onto the state? By the 1870s, British commerce depended on distant sea routes and imported inputs whose security no individual firm could provide. I model a government allocating protection across trade routes to reduce expected disruption costs, with protection increasing in disruption risk, rerouting costs, and dependence on hard-to-substitute imports. Because steamships could travel only where coal was stocked, geocoded coaling stations show the feasible route network and how it expanded. I apply this framework to the Russo-Turkish War of 1877–1878, which unexpectedly increased concern in Britain about the security of overseas commerce. In a global grid-cell panel from 1800 to 1920, I measure state attention with places mentioned in confidential correspondence and military maps, and state action with coaling stations, defended naval stations, and annexations. State activity shifted toward more exposed locations after the war, consistent with a state response to risks private actors could not bear. This state presence outlasted the shift to oil-fired ships: bases acquired to secure nineteenth-century trade routes remained in British hands and became centers of military activity during World War I.

What is the electoral impact of banning a political party? We assemble data on all recorded party bans since 1900, 2,015 bans in 150 countries. We link these data to election results and pair them with detailed case-study data on two bans and a survey on the proposed ban of the German AfD. We find, first, that bans in democracies do not change the vote share of the ideological faction of the banned party. A model with fixed voter demand and adjusting party supply accounts for this retention of votes. When re-entry is free, a successor inherits the support of the banned party almost one for one. When re-entry is blocked, neighboring parties absorb its voters and adopt its language. We illustrate each case with our case studies. Second, party bans change who governs. The banned party is typically excluded from government, so a ban often moves votes to nearby parties eligible for government. This substitution raises the probability that the banned faction leads the government by 11 percentage points, from a pre-ban rate of 17 percent.

How do new scientific ideas spread: by changing the views of established researchers or through the entry of new researchers and journals? After 1850, British scientists increasingly described race as a biological category that was fixed, heritable, and ranked. We develop a text-based measure of how scientific fields conceptualize race and apply it to 227,000 articles published in 110 British and Irish journals between 1800 and 1900, linking each mention to its journal and author. Interest in race rose after Darwin’s On the Origin of Species (1859), then fell within a decade. But the share of mentions framing race in biological terms doubled and remained high through 1900. New journals, mostly in anthropology, and new authors each account for about three-quarters of this lasting shift, while changes in the writing of established authors explain less than one-tenth. Successive cohorts of new authors used biological language more frequently. These findings suggest that the shift toward a biological conception of race came less from established scientists changing their minds than from new journals and authors entering the field.