Pins and Pistons: Gender and Patenting during the French Industrialization

Final Seminar

Youssouf Merouani

2025-12-05

Why Study Women and Technology?

Why 19th-Century France?

  • France and England had similar levels of economic performance until the 17-th century (Ridolfi and Nuvolari, 2021)

  • French industrialization: from “failure” narratives emphasizing backwardness (Clapham, 1921; Landes, 1949) to revisionist accounts stressing rational specialization in quality production (O’Brien and Keyder, 1978; Roehl, 1976)

  • French patent system (1791). 1844 reform introduced installment payments, lowering barriers (Baudry, 2020; Galvez-Behar, 2019)

  • Law: ‘the administration is not authorized to inquire about the civil capacity of the petitioner. If the request comes from a married woman, a minor, or a legally incapacitated person, the administration must issue the patent’.

Sources

I) How to identify Women?

Brevet de quinze ans, 22 janvier 1891; par la dame Durand, née Morel, Veuve Durand, et la demoiselle Morel, à Paris — Chaussures à …

Women-linked Patents, 1791-1900

II) And how do you classify technology at scale?

Automatic Patent Technology Classification of 390,000 patents at an expert level through Machine Learning.

What happens when inventors collaborate?

Do they become more productive? (Kreutzer and Taalbi, 2025; Lee and Bozeman, 2005)

Do they have wonderful new ideas? (Fleming et al., 2007; Uzzi et al., 2013)

Or more influential ideas? (Uzzi et al., 2013; Wuchty et al., 2007)

I) How to identify unique inventors? II) And how to measure novelty and influence of ideas?

~500,000 Inventor-Observations ➔ Scarce Information ➔ Probabilistic + Machine Learning Algorithm ➔255,000 unique inventors. Over 4,000 are individual women inventors

Novelty is asking: How similar is an idea to previous art?

While Influence is asking: How similar is an idea to future art relative to how derivative it is at birth?

What happens to inventors that switch from Solo to Collab? Compared to those that have not switched-yet and never switch.

No productivity or novelty. Some influence. Mostly it's about moving to new sectors

How does the network structure (t–1) predict diversification (t)?

  • Cross-class collaborators strongly predict Diversification.

  • Non-redundant ties provide better access to different knowledge.

  • Big networks alone do not cause exploration

  • Family ties and home-tech ties keep you anchored in your home field

Who are these women inventors?

No available digitized census data on the national level

Algorithm pipeline that exploits available civil-records and metadata in genealogical search engines

Automatically builds full biographies for 545 married women inventors with high reliability, ~95% accuracy, for all life-event timings.

OLS analysis shows that (birth) regions with high literacy rates and low share of women in agriculture have higher women inventors per capita

Invention happens on average in the early forties, after marriage

… and childbearing

Main Findings: Four Papers

  • Micro-level data is solving puzzles about gender and long-run development

  • 1.8% of French patents female-linked (1791-1900), higher on average than Britain or US. Stable across the century.

  • Not Elite: 56% middle and low-skilled backgrounds

  • Women patented across all sectors from “pins” to “pistons”, with higher likelihood in sectors they typically worked in.

  • Women relied heavily on family ties for collaboration with higher number of connections in male-dominated sectors

  • Collaboration persistently shifted inventors across technology fields

  • Women’s trusted cross-class partnerships generated especially large diversification returns

  • Women entered mid-life (41), after marriage and after childbearing

  • Higher women inventor rate in counties with higher literacy and lower agricultural employment

References

Baudry, J. 2020. A Politics of Intellectual Property: Creating a Patent System in Revolutionary France, Technology and Culture, vol. 61, no. 4, 1017–44
Chanteux, A. 2023. “Les filles d’Athéna : Femmes et brevets au XIXe siècle en France,” These de doctorat, Paris, EHESS
Clapham, J. H. 1921. The economic development of France and Germany, 1815-1914, Cambridge [Eng.] The University press
Fleming, L., Mingo, S., and Chen, D. 2007. Collaborative Brokerage, Generative Creativity, and Creative Success, Administrative Science Quarterly, vol. 52, no. 3, 443–75
Galvez-Behar, G. 2019. The patent system during the French industrial revolution: Institutional change and economic effects, Jahrbuch für Wirtschaftsgeschichte/Economic History Yearbook, vol. 60, no. 1, 31–56
Goldin, C. 1990. Understanding the Gender Gap: An Economic History of American Women: National Bureau of Economic Research gold90-1, date last accessed August 12, 2022, at https://www.nber.org/books-and-chapters/understanding-gender-gap-economic-history-american-women
Khan, B. Z. 2016. Invisible Women: Entrepreneurship, Innovation, and Family Firms in Nineteenth-Century France, The Journal of Economic History, vol. 76, no. 1, 163–95
Kreutzer, P. J. and Taalbi, J. 2025. Collaboration for the Bioeconomy
Landes, D. S. 1949. French Entrepreneurship and Industrial Growth in the Nineteenth Century, The Journal of Economic History, vol. 9, no. 1, 45–61
Lee, S. and Bozeman, B. 2005. The Impact of Research Collaboration on Scientific Productivity, Social Studies of Science, vol. 35, no. 5, 673–702
O’Brien, P. and Keyder, Ç. 1978. Economic growth in Britain and France 1780 - 1914: Two paths to the twentieth century, London, Allen & Unwin
Oldenziel, R. 1999. Making technology masculine: Men, women and modern machines in America, 1870-1945, Amsterdam, Amsterdam University Press
Ridolfi, L. and Nuvolari, A. 2021. L’histoire immobile? A reappraisal of French economic growth using the demand-side approach, 1280–1850, European Review of Economic History, vol. 25, no. 3, 405–28
Roehl, R. 1976. French industrialization: A reconsideration, Explorations in Economic History, vol. 13, no. 3, 233–81
Stanfors, M., Leunig, T., Eriksson, B., and Karlsson, T. 2014. Gender, productivity, and the nature of work and pay: Evidence from the late nineteenth-century tobacco industry, The Economic History Review, vol. 67, no. 1, 48–65
Tilly, L. A. and Scott, J. W. 1989. Women, Work and Family, New York, Routledge
Uzzi, B., Mukherjee, S., Stringer, M., and Jones, B. 2013. Atypical Combinations and Scientific Impact, Science, vol. 342, no. 6157, 468–72
Wajcman, J. 2010. Feminist theories of technology, Cambridge Journal of Economics, vol. 34, no. 1, 143–52
Wuchty, S., Jones, B. F., and Uzzi, B. 2007. The Increasing Dominance of Teams in Production of Knowledge, Science, vol. 316, no. 5827, 1036–39

Appendix: Collaboration paper

Outcome Outcome variable Sample (treated group) Comparison group Fixed effects Detrending? N (inv-yr)
Productivity Number of main patents inventor i files in year t Switchers Never-collaborators + Not-yet-treated switchers Inventor FE; Calendar-year FE; Stack FE No 3,568,668
Novelty Mean novelty score of all main patents filed that year Switchers Never-collaborators + Not-yet-treated switchers Inventor FE; Pre-event home-class × year FE; Stack FE No 1,057,035
Influence Mean influence score of all main patents filed that year Switchers Never-collaborators + Not-yet-treated switchers Inventor FE; Pre-event home-class × year FE; Stack FE No 1,057,062
Diversification Share of patents outside inventors’ pre-event home class (residuals after detrending) Switchers Not-yet-treated switchers Inventor FE; Pre-event home-class × year FE; Stack FE Yes (within-inventor linear trend in OutsideClassShare removed) 98,052

Variable Meaning in plain English
Cross-class partner share (t-1) Fraction of your collaborators last year who worked in different technology classes than you.
Degree (t-1) Number of collaborators you had last year (how big your personal network is).
Non-redundancy (t-1) Share of your neighbors who are not connected to each other. This is the classic Burt structural holes / brokerage measure.
Class reach (t-1) How many distinct technology classes your collaborators collectively span.
Family tie share (t-1) How many of your collaborators are family members.
Repeated outside tie share (t-1) Repeated ties to collaborators who are outside your entry class.
Repeated home-only tie share (t-1) Repeated ties to collaborators who are only in your entry class (deep inside your home field).
Neighbors’ outside output (t-1) How much your neighbors (collaborators) worked outside their own home field last year (a measure of exposure to explorative people).