Graduate Economic History Seminars 2026-27

Autumn Term seminars are Wednesdays, 1.00 - 3.00pm

Venue: KSW 212 (unless otherwise stated)

30 September

Johannes Karge, Paris School of Economics
Bonds, Business Cycles, and Financial Crises
Abstract: Do corporate bond markets function as an aggregate “spare tire” when banks are impaired? I study this question using a new long-run dataset of primary bond issuance by issuer sector for 19 countries from the nineteenth century to the present. The data distinguish non-financial corporations, financial corporations, and governments, allowing me to trace how primary bond-market financing is reallocated during systemic banking crises. I find that non-financial corporate bond issuance contracts sharply and persistently after crisis onset rather than expanding to replace impaired bank lending. Three years after a crisis, issuance is roughly one-half below its pre-crisis level, while government issuance rises strongly and financial-sector issuance increases only temporarily. The contraction remains pronounced when issuance is scaled by GDP, investment, or pre-crisis bank credit, and is stable across historical periods, crisis definitions, and alternative specifications. Combining the issuance data with historical credit spreads shows that crises accompanied by larger credit-market repricing experience substantially deeper subsequent contractions in non-financial corporate issuance, consistent with tighter market conditions contributing to the decline. Countries with deeper pre-crisis bond markets subsequently experience stronger investment recovery, but this relationship is not distinctly stronger during banking crises. The results reveal an aggregation wedge between firm-level substitution and aggregate financing: selected firms may retain access to bond markets when banks retrench, but this margin is not large enough to stabilize non-financial corporate financing in systemic crises.

7 October

Noah Sutter, LSE
Surnames and Distinction: Multi-outcome Social Mobility Estimates for French Elites (1791-1902), Testing a Multiple-Capital-Forms Interpretation of the Latent Factor Model
Abstract: Bi-generational estimates of intergenerational wealth mobility imply that economic advantages dissipate within only a few generations. Surname-based methods, by contrast, uncover much higher rates of intergenerational persistence. I extend the surname-based method to 19th century France, using a newly digitised dataset of 1.8 million observations of wealth at death. I find high rates of persistence comparable to the ones for England during the same period - between 0.74 and 0.93. I use data on 22 elite registers linked at the surname level, yielding more than 110,000 observations, to independently test these results and confirm the persistence levels. I do not find evidence for the economic decline of the ancien régime nobility. While sorting on wealth gives a persistence rate of 0.85 (0.09) for the highest ventile, sorting on wealth and hereditary noble titles results in no detectable regression to the mean. I propose the hypothesis that the latent factor can be interpreted as an durable endowment with social and cultural capital.

14 October

John Zhang, LSE
Welfare Retrenchment and Social Unrest
Abstract: Does welfare retrenchment cause social unrest? I study this question using parish-level expenditure on poor relief and events of social unrest in post-Napoleonic England, drawing on the welfare reforms of 1818-9 to motivate instrumental variables, regression discontinuity, and event study research designs. The reduced-form analysis finds a causal effect of welfare retrenchment on social unrest between 1816 and 1834, attributable to an exacerbation of relative deprivation that stems from comparisons with higher-expending parishes. Simulations from a counterfactual analysis assuming greater accountability for these grievances highlight the importance of equitable social welfare provision in minimising the political costs of austerity.

21 October

Kangle Zhu, Pompeu Fabra
Gods of the Coast: Religious Networks and the Making of Nationhood in Taiwan
Abstract: Can decentralized religious networks contribute to national identity outside direct state-led nation-building? We combine newly assembled historical records on the establishment and lineage of Mazu temples in Taiwan with repeated cross-sectional survey data. To address endogenous temple formation, we exploit historical variation in hierarchical religious diffusion and demand for maritime protection. Greater long-run exposure to the Mazu network increases Taiwan-centered national identification. Organizational structure is central to this relationship: localities with stronger lineage ties and greater network embeddedness exhibit stronger Taiwan-centered identity, consistent with repeated interaction across geographically dispersed communities rather than doctrinal belief alone. These patterns vary across governing periods and extend to stronger support for Taiwanese independence. The findings show how decentralized social networks can provide durable foundations for national identity beyond formal state institutions.

28 October

Marco Cokic, LSE
Making Friends - Diplomacy and Trade in Cold War Poland
Abstract: Economic considerations play a crucial role in the foreign policy of countries around the world, yet the empirical literature on economic diplomacy is almost entirely confined to market economies, where diplomatic activity reduces informational and contractual frictions between private firms. To what extent diplomacy matters in a centrally planned setting, where export quantities were administratively determined and prices politically set, remains unresolved. This paper provides the first systematic quantitative evidence on the trade effects of economic diplomacy under central planning.
The paper draws on a novel dataset covering Poland's bilateral state visits and treaty signings between 1945 and 1989 across over 150 partner countries, matched to bilateral trade flows transcribed from Polish national statistical yearbooks. Visits are disaggregated by delegation rank and by whether the encounter was scheduled or incidental, such as a state funeral or the Olympic Games.
The empirical strategy proceeds in four steps. A PPML structural gravity equation with bilateral pair and exporter-importer-time fixed effects establishes the baseline association. Incidental multilateral encounters are exploited as quasi-random variation, complemented by propensity score matching to address the endogeneity of planned visits. Further, visits are disaggregated by attendee rank and by first versus subsequent contact to test whether the extensive margin carries different trade returns.
The paper finds that diplomacy promotes trade even under central planning, with important heterogeneity. Visits to Western partners show larger effects than intra-COMECON visits, where relationships were already politically mandated. First visits carry small but consistent effects with a fixed cost of relationship formation and lower-ranking delegations produce stronger trade responses. The findings speak to how political institutions shape the returns to commercial diplomacy and offer a historical perspective on contemporary debates about geopolitical fragmentation and diplomatic engagement as a tool of economic statecraft.

11 November

Anna Arkhina, Bocconi
Emancipation, Industrialization, and Assortative Mating in Moscow Governorate, 1830–1910
Abstract: This article examines estate homogamy in Moscow Governorate between 1830 and 1910, asking how emancipation, industrialization, migration, and urbanization reshaped marital boundaries in imperial Russia. It draws on more than 66,000 marriage entries from Orthodox parish registers for Moscow city and the surrounding uezds, supplemented by benchmark years around the abolition of serfdom and the Revolution of 1905. Spouses are classified into seven estate-based groups, with occupational classifications used as a robustness check. Log-linear models distinguish changes in marital association from shifts in the marginal composition of local marriage markets and compare general, group-specific, status-distance, and hypergamous patterns. Estate homogamy remained a persistent feature of marriage formation throughout the period, but its strength and trajectory varied sharply across space and social groups. Moscow exhibited lower homogamy than the predominantly rural uezds, while differential homogamy models consistently provided the best fit. Net of compositional change, marital closure weakened over time, beginning earlier in Moscow and declining sharply in rural districts around emancipation. Peasants and the military estate accounted for much of this decline, whereas clergy, merchants, and ranked officials retained strong endogamy. A comparison of industrial and agrarian uezds shows that pre-emancipation homogamy was higher in agrarian areas, where landlord authority and village institutions more tightly constrained partner choice; after 1861, this gap narrowed as mobility, wage labor, transport, and market integration expanded. The findings provide conditional support for modernization theory. Institutional reform and socioeconomic development weakened inherited barriers but did not produce a uniform or linear dissolution of social closure. Instead, modernization reconfigured the bases and geography of assortative mating, leaving elite and corporate boundaries resilient even as traditional rural constraints eroded. The study extends comparative research beyond Europe’s Hajnal line and demonstrates the value of estate status for understanding stratification and partner selection in imperial societies undergoing uneven change.

18 November

Andrés Irarrazaval Garcia Huidobro, LSE
The Long March Towards Equality: Theory and Evidence from Surplus Distribution
Abstract: Has inequality changed over the long run? The world average top 1% share of national income was 18% in 1820, 20% in 1920, and is 18% today, suggesting that elite power over resources has been remarkably resilient to historical change. This paper reveals a Great Levelling. When nations were poorer, much of national income was tied to subsistence. Transferring this income to the top 1% would leave the other 99% below what they need to survive. I factor this subsistence constraint into inequality measurement and study how the surplus — income above subsistence and therefore transferable — is distributed. Doing so shows that the world average top 1% share of national surplus fell from 53% in 1820 to 40% in 1920 and 19% today, a 2.8-fold decline in elite control over transferable resources. Growth, by lifting incomes above subsistence, expanded the share of income that could accrue to the top 1%, from 38% in 1820 to 90% today. Yet actual top income shares have not risen: the surplus that growth liberated accrued mainly to the 99%. Case studies also point to the importance of factoring in subsistence constraints: top income shares, paradoxically, make former US slave states under Jim Crow appear more equal than non-slave states, colonial India as equal as democratic India, and low-income nations as equal as high-income ones today. In all three, elites control most of a limited surplus while exhibiting an unremarkable top income share.

25 November

David Teeters, LSE
Alpha: Information, Networks, and Investor Returns. Evidence from the Universe of Named British Financial Market Participants, 1752-1827
Abstract: Who made money in early modern British capital markets and how? Drawing on newly-digitised archival records, this paper offers the first population-scale empirical answer to a historiographical question long debated in institutional and qualitative terms. The English financial revolution (Dickson (1967), Neal (1990), Fetter (1965), O'Brien (1988)) established the origins of modern financial markets over the long eighteenth century: a liquid secondary market for public debt, a network of broker-dealers, and a broad investor base. The paper provides the novel empirical foundation for analysing the distributional consequences of this transformation.
Alpha is the first in a series of six articles examining the macrofinancial foundations of modern capital markets. It exploits the Financial Lives of the British (FLB) dataset (approximately fifty million transactions among more than five hundred thousand named market participants, 1694–1827) to construct the first complete cross-sectional distribution of lifetime individual portfolio returns for any historical market.
This paper combines two empirical strategies that the modern finance literature can rarely observe together at population scale. First, the trading network among named market participants is reconstructed from pairwise counterparty transactions and each investor's centrality measured. Where recent literature infers trading relationships, a financial market history approach allows direct observation from official primary archival records. Second, the informational content of individual investors' trades is measured by the direction and magnitude of subsequent price movements, following Kyle (1985). The joint distribution of network centrality and informational advantage, and its relationship to lifetime portfolio returns, is the primary focus of this paper.
Anticipated conclusions: neither investing behaviour nor preexisting wealth alone predicts membership in the right tail of lifetime returns; investors central to the trading network outperform on both counts; the trades of central investors carry significantly greater price impact than those of peripheral investors, consistent with informational advantage as the channel by which central network position translates into returns.

2 December

Guillem Blasco-Piles, Barcelona
The Great Inner Divergence: TFP and Manufacturing Dualism in Industrializing Empires before WWI
Abstract: This paper provides the first aggregate and disaggregated comprehensive Total Factor Productivity estimates for manufacturing in the Ottoman, Qing and Russian Empires before their collapse, incorporating both the traditional industry and capital estimates. Previous studies relied on modern-only establishments and labour productivity estimates, masking the role of capital and inner economic dynamics, which become essential during structural transformation processes. Using industrial censuses from 1908-1913 and regional reports combined with a novel reconstruction methodology for the traditional industry TFP, our results document extreme internal productivity dualism. Mechanized establishments achieved close to British efficiency levels while traditional non-mechanized plants operated at one-fifth to one-third of the industrial leader. At the aggregate level, lower-productivity traditional establishments seem to determine the aggregate productivity due to their vast weight in the manufacturing landscape. These findings suggest the persistence of the Great Divergence stemmed not from technological adoption incapacity but from the inability to diffuse new technologies beyond modern industrial enclaves—a pattern that illuminates persistent dualism in developing economies today.

9 December

Liza Brover, Yale
Economics of Repression
Abstract: Between 1937-1938, over 1.5 million people were arrested in the Soviet Union as part of Stalin's Great Terror. Existing theories to explain the targets of the Great Terror suggest political motivations in line with those articulated in Executive Order 00447, which outlined repression quotas for criminals, former wealthy peasants, and anti-Soviet elements. This paper provides evidence for an economic motivation to explain the within-region and within-enterprise patterns of repression. A decrease in economic plan fulfillment (an increase in economic tautness) is associated with lower levels of repression. This is because the shadow cost of repressing a worker in a region where economic plan targets were not met was higher than in regions where the production constraint was met or overfulfilled. I provide both aggregate and enterprise-level evidence that is consistent with this explanation. Using newly digitized individual-level factory worker surveys, I find that workers from enterprises where economic plans were filled or overfulfilled (not taut) were more likely to be repressed, even when controlling for political motivations. However, higher value workers were more likely to be shielded from repression, in line with the explanation that their shadow cost of repression was higher. This finding is new to the literature as it is the first to provide empirical evidence of economic decision-making behind the Great Terror.