Research Highlights Article
November 6, 2026
Labor shortages and industrial advancement
Wartime labor scarcity pushed English farmers to adopt laborsaving technology during the Industrial Revolution.
A 1780 caricature of a press gang in Britain
Source: Public domain, via Wikimedia Commons
Economists have long debated why the Industrial Revolution began in Britain. One prominent account, associated with economic historian Robert Allen, holds that British labor was expensive and coal was cheap, so it paid to invent and adopt machines that replaced workers. A competing account, advanced by Joel Mokyr, emphasizes Britain’s unusual supply of skilled mechanics and tinkerers who could build, maintain, and improve those machines.
In a paper in the American Economic Review, Bruno Caprettini, Hans-Joachim Voth, and Alex Trew explain how they found evidence for both explanations.
“We didn’t think that there was any reason why the two theories could not be right at the same time,” Caprettini told the AEA in an interview.
To test Allen’s hypothesis, the economists needed to find examples of labor shortages in Britain that arose for reasons unrelated to local economic conditions. They found such a labor shock in the French Revolutionary and Napoleonic Wars of 1793 to 1815, when the Royal Navy grew from roughly 16,000 sailors to about 150,000.
At the height of mobilization, more than 350,000 men served in the army and navy combined, equivalent to 10 to 14 percent of Britain’s adult male labor force. Because experienced seamen were scarce, the navy took in large numbers of untrained “landsmen,” some of whom were forcibly recruited by press gangs.
Contemporary observers noticed the consequences for local economies. An 1812 agricultural survey of Dorset reported that farmers were installing threshing machines chiefly because laborers were scarce, and it expected that scarcity to be most severe in maritime districts.
Some of the solutions [to aging populations] may be just to let the market do its job. There will be fewer people, higher wages, and then this is going to create this incentive to adopt these new technologies.
Bruno Caprettini
To measure these effects, the authors assembled several new datasets. On the labor side, they digitized the navy’s crew lists, known as muster rolls, covering more than 95,000 sailors from 263 ships, and combined them with existing army enlistment records, tracing the home locations of more than 45,000 recruits.
With the help of large language models, they extracted more than 1,500 local wage observations from roughly 20,000 pages of government agricultural reports. They also constructed an index of local mechanical skill from apprenticeships in trades such as millwrighting, mechanical patents, and newspaper references to machinery.
To track technological progress, the authors identified nearly 4,900 agricultural machines listed in newspaper advertisements for farm sales. They then classified each machine as laborsaving, such as threshers and horse rakes, or nonlaborsaving, such as turnip cutters, which were were useful in producing more livestock fodder from the same produce, but required additional labor.
To isolate the cause and effect between labor shortages and machine adoption, the authors exploited a physical constraint on the navy’s warships. These ships were so large that they typically needed waters that were 10 to 15 meters deep to anchor. As a result, recruiting parties reached coastal communities near deep waters far more easily than those facing shallow estuaries. The authors restricted their analysis to areas within 15 kilometers of the coast and compared places equally distant from the shoreline but at different distances from deep water.
The parishes of Lesbury and North Ferriby in northeast England illustrate the approach. In 1801 they were similar in size and economic structure. Lesbury, however, faced the deep waters of the North Sea, while North Ferriby sat on the shallow waters of the Humber estuary. The navy recruited in Lesbury, which later had four laborsaving machines to North Ferriby’s one. Across the full coastal sample, distance to deep water strongly predicted naval recruitment but not army recruitment.
The results indicate that recruitment produced real labor shortages, as evidenced by more significantly skewed sex ratios where recruitment was more intense. Across the sample, farm wages roughly doubled between 1790 and 1811 while living costs rose 64 percent.
Those labor shortages, in turn, led to more technology adoption. Areas with heavier recruitment had more laborsaving machines, with no comparable increase in nonlaborsaving equipment. The authors estimated that a one-standard-deviation increase in naval recruitment led to a 0.6 to 1.0 standard-deviation increase in laborsaving machines.
Military recruitment and technology adoption
The chart below shows the average number of agricultural machines in areas grouped by their level of military recruitment per capita, based on data from 1790 to 1820. Panel A shows that areas with heavier recruitment had more laborsaving machines, while panel B shows little relationship for non-laborsaving machines.
Finally, the effect of labor scarcity on adoption was larger, on average, where mechanical skills were more abundant, as Mokyr’s account of the Industrial Revolution would suggest.
The consequences may also have outlasted the war, as mechanization continued after soldiers and sailors returned home. Regions with more wartime technology adoption later produced more productive threshing machines and shifted workers out of farming more quickly.
The findings may also shed light on the forces that are shaping the evolution of technology today. In particular, they offer empirical support for theories of directed technological change, which predict that scarce labor encourages technologies that economize on it. For contemporary economies, Caprettini sees a measure of reassurance for countries facing aging and shrinking populations.
“Some of the solutions may be just to let the market do its job,” he said. “There will be fewer people, higher wages, and then this is going to create this incentive to adopt these new technologies.”
Overall, the research suggests that the two leading explanations for Britain’s industrial takeoff are not mutually exclusive. Scare labor and skilled mechanics worked together to facilitate the adoption of the technologies that spurred the Industrial Revolution.
♦
“Fighting for Growth: Labor Scarcity and Technological Progress during the British Industrial Revolution” appears in the October 2026 issue of the American Economic Review.