Research

My research asks how political and educational institutions shape long-run economic development, with a primary focus on nineteenth-century France. I am particularly interested in how institutional variation at the local level (in suffrage rules, schooling access, and administrative structure) translates into differences in labour markets, industrialisation, and regional growth. Empirically, I construct original historical datasets from archival sources, combining quantitative methods with GIS, OCR, and machine learning tools for large-scale data extraction. Methodologically, I am interested in how historically persistent territorial patterns condition what can be credibly identified from policy variation, and what those patterns themselves reveal about long-run development.

Master’s Thesis

“Enrichissez-vous !”: The Effects of Primary Schooling and Political Participation on Economic Development in 19th-century France

Master’s Thesis · Sciences Po · 2025 Sciences Po Prize for Best Master’s Thesis (2024–2025)

Jury: Roberto Galbiati, Emeric Henry (supervisors), Kevin O’Rourke

Abstract This paper studies the joint role of primary schooling and local democratic participation in shaping industrial development in 19th-century France. I exploit two population-threshold reforms enacted under the July Monarchy - the Municipal Law of 1831, which expanded local voting rights, and the Guizot Law of 1833, which mandated primary schools in municipalities above 500 inhabitants - using a newly assembled arrondissement-level dataset covering 1830–1865. A static OLS design finds that pre-reform male primary education is positively associated with industrial wages in the 1840s. A dynamic IV specification, instrumenting schooling changes with exposure to the Guizot threshold, detects no statistically significant medium-run wage effect. Municipal suffrage exposure shows no independent or interactive effect in either design. A central finding is methodological: because commune-size distributions are historically structured and spatially correlated with prior development, threshold-based designs in this setting face identification challenges that are themselves historically informative.

Predoctoral Research

As a predoctoral fellow at UZH under Hans-Joachim Voth, I contribute to the following projects.

(Untitled) H.-J. Voth, M. P. Squicciarini, N. Voigtländer
I contribute to the early-stage data construction of this project, set in Revolutionary France, integrating archival, textual, and geospatial sources into a unified research pipeline — including historical geocoding, GIS integration, and the preparation of administrative data for econometric analysis.
Image(s) H.-J. Voth and D. Yanagizawa-Drott · R&R, Quarterly Journal of Economics
This project uses machine learning applied to over 14 million high school yearbook portraits (1930–2010) to measure cultural change through stylistic choices, constructing granular measures of individualism, cultural persistence, and style novelty across US commuting zones. I contributed to the computer vision pipeline: image preprocessing, labeling and validation, supervised model training, as well as feature and name extraction.
Fighting for Growth: Labor Scarcity and Technological Progress During the British Industrial Revolution H.-J. Voth, B. Caprettini, A. Trew · Conditionally accepted, American Economic Review
I contributed to the replication package by writing clean, documented code to recreate a subset of the variables used in the analysis.

For details on additional research assistance and replication experience, feel free to consult my Resume.

Selected Sample Work

A selection of assignments from my graduate coursework. Browse the full Sample Work page for more.

  • Railroads, Market Access, and Urban Fringe Dynamics: Extensions to Donaldson & Hornbeck (2016) Research Proposal · Urban & Regional Economics | Sciences Po · 2025 This proposal outlines two extensions to the Donaldson-Hornbeck market-access framework: one incorporating dynamic spatial adjustment and urban fringe land-use equilibria (drawing on monocentric city theory), and one adapting the approach to 19th-century France using railway maps, agricultural inquiries, and historical census data. [PDF]
  • Primary Schooling and Crime in 19th-Century France: A Cohort-Based Identification Strategy Research Proposal · Topics in Political, Public & Organizational Economics | Sciences Po · 2025 Proposes a causal identification strategy for the effect of the 1833 Guizot Law on adult crime using cohort exposure as an instrument for literacy, estimated in a panel fixed-effects framework with 19th-century administrative data. [PDF]
  • Replication of Monge-Naranjo, Sanchez & Santaeulalia-Llopis (2019) Replication Package · Computational Economics | Sciences Po · 2024 · with Lionel Chambon and Paulo Gugelmo Cavalheiro Dias Full Julia replication package for “Natural Resources and Global Misallocation” (AEJ: Macroeconomics), reconstructed from scratch — the authors provided raw data and a single merge file but no code for tables or figures. The full replication package, documentation, and reproducible workflow are available here.