Friday, 28 August 2026
61 new items across 4 threads · 169 screened · 34 of 39 sources reached
21 items
Recovers risk preferences from historical crop-choice records by combining a structural portfolio model with machine-learning predicted returns, a template for extracting behavioural parameters from archival data rather than surveys.
digitizednewly digitized
Builds newly digitised twentieth-century US college records to trace assortative mating and mobility, a large-N historical linkage exercise.
Rebuilds Fogel's social-savings question with modern quantitative spatial models, connecting classic cliometrics to current method.
archival
Reconstructs a century of US corporate bond data from archival sources, a new long-run dataset built from primary records.
Maps the Grandes Gabelles of early modern France, turning fiscal geography into usable spatial data on the pre-modern state.
Uses voyage durations to Australia from 1848 to 1885 to measure sailing-ship technical change from shipping records.
digitizednewly digitized
Uses newly digitised Chinese Maritime Customs Service personnel records to trace minority bureaucrats' careers, large-N linkage of historical service records.
Assembles two centuries of systemic bank runs, a long-run crisis dataset built from historical sources.
Uses the Depression and WWII to explain the emergence of the college gender gap, cohort-based historical identification.
Compares the language of historical redlining documents with modern tenant-screening algorithms, text extraction applied across a long time span.
Long-run development in the aftermath of war, on how conflict shapes subsequent growth paths.
Ethnic wealth inequality in England and Wales from 1858 to 2018, built on long-run probate and administrative records.
state capacity
Machine learning applied at scale to Qing dynasty sources to measure state capacity, exactly the historical-extraction-to-institutions pipeline.
Uses the Chinese Exclusion Act as a shock to trace its long-run effect on Western US development.
The Grain Invasion as a natural experiment in the distributional consequences of trade, a classic cliometric question.
Traces Soviet-built Million-Rouble plants in China over the long run, an industrial-persistence design.
New GDP estimates for Italy from 1300 to 1861, historical national accounting of the kind that underpins Little Divergence debates.
Institutional innovation as a precondition for steam adoption, on why technology diffuses when institutions permit it.
Recovers work-limiting disability variables from the 1980s CPS, a data-rescue exercise on an inconsistently coded historical survey.
The 1927 Mississippi Flood as a shock to religious participation, treating religion as informal insurance.
Trade and the end of antiquity, quantifying long-distance exchange at the close of the ancient economy.
14 items
Shows that quietly trying several candidate instruments and reporting the best one inflates IV results, with diagnostics worth running whenever a historical instrument was chosen from a few plausible alternatives.
Handles censoring whose distribution shifts between estimation and target samples, the same overlap problem that arises when a linked historical sample is unrepresentative of the population it came from.
Identification of policy effects from discrete rule changes, generalising the logic behind RDD and bunching designs.
Optimal experimental design and estimation when outcomes are bounded, tightening what can be learned from a fixed sample.
Designs experiments to maximise signal when treatment spills across units, relevant wherever neighbours or networks interfere.
Identification and inference under multi-dimensional unobserved heterogeneity, for settings where selection runs on several latent margins at once.
Shows that experimental sign-up is itself a treatment choice, collapsing two apparently separate selection problems into one bias term.
Identifies unit and period unobserved heterogeneity in nonparametric panels without imposing functional form.
A potential-outcomes treatment of two-period within-subject designs, with tests for the carryover assumptions they rest on.
Inference when observations are dependent through a network defined by graph distance, the relevant case when units are linked by geography or social ties.
Analyses what averaging or long-differencing a panel does to long-run estimates, bearing directly on cross-sectional growth and persistence regressions.
Bayesian inference over identified sets in discrete response models, for cases where point identification fails and bounds are the honest answer.
Identifies policy-relevant effects when units select into treatment and spill over onto neighbours, the standard difficulty in spatial settings.
Cuts simulation bias in multinomial choice estimation under a fixed computational budget.
19 items
Links economic misperceptions to support for democracy, speaking to the belief-based microfoundations of regime legitimacy.
Synthesises how cultural traits evolve and persist, the mechanism underlying most institutional-persistence results.
Models politicians who build support by creating new political goods, a supply-side account of institutional change.
Property tax re-assessment in India as a direct test of what the state can and cannot outsource, squarely a state-capacity question.
Traces how private social clubs feed into political office, an unusually clean measurement of elite capture.
Redistricting and race in local government, on the mechanics by which political institutions convert votes into representation.
Elite political selection across American history, on how family background persists in who ends up holding office.
Tests the link between social structure and wealth inequality across many non-Western societies, extending the institutions-and-inequality literature past the usual samples.
Acemoglu and Robinson propose a systems account of culture as a repertoire interacting with institutions, with discontinuous rather than gradual change.
Uses administrative data on franchise extension to trace its fiscal and educational consequences, a direct test of the democratisation-to-schooling channel.
Separates the economic and political effects of emancipation on slave owners, showing power can persist even as wealth is destroyed.
Argues impersonal rules are what tie democracy and capitalism to equality, a North-style institutional argument.
Reads historical industrial policy through the lens of conflict, on how war shapes what states can direct.
Revisits the legal-origins account of financial development, the canonical institutions-and-growth debate.
A null-calibration evaluation of the eforensics election-fraud model across two decades, testing whether a widely used detection method actually discriminates.
Televised religious epics and the rise of Hindu nationalism, on media as a channel for political identity formation.
Links fallow lengths to the structure of property rights, connecting agricultural technology to institutional form.
Tests whether modern cultural traits trace to historical roots, on the persistence of moral sentiments.
Compares the institutional legacies of Spanish and American rule in the Philippines, a colonial-origins persistence design.
7 items
Compares two tests of sports betting market efficiency, a methodological question about how efficiency is measured rather than assumed.
Uses empty-stadium rugby sevens to separate crowd effects from other sources of home advantage.
Risk-taking under high stakes in Olympic volleyball, using sport as a controlled setting for decisions under pressure.
Revisits penalty shootouts as a natural experiment in performance under pressure, including the order effects the literature disputes.
Experimental evidence of racial bias in selecting a soccer lineup, discrimination measured in a sporting decision task.
Empirical analysis of public expenditure on hosting the FIFA World Cup, on whether mega-event spending returns anything.
Asks whether a structural model beats the closing price as a forecast, an information-aggregation test in betting markets.