Editors’ note: This column is based on CEPR Discussion Paper 21548 “Fighting for Growth” (forthcoming in American Economic Review). 

In every rich country today, as populations age and few youngsters enter the labour force, employers and politicians bemoan labour or skill ‘shortages’ – typically meaning that they would like to hire more at present wages (van Herck et al. 2026). But what happens if massive labour shortages suddenly hit an economy? In this column, we turn to historical evidence to find out. Between 1792 and 1815, Britain and France fought each other for global dominance. Many men who had previously tilled the soil or tended machines found themselves marching and fighting or sailing the high seas. 

More than one out of every ten British men ended up in the army or navy. This created acute labour shortages, and upward pressure on wages. It also led to massive productivity gains through the adoption of new machines. By the time the war was over, Britain’s transition to mechanised agriculture was well underway.  

The debate on the Industrial Revolution

Why did Britain industrialise first? Allen (2009) famously argued “the Industrial Revolution… was invented in Britain … because it paid to invent it there”. British coal was abundant and cheap, while British labour was scarce and dear. It made sense for entrepreneurs to adopt machines that used lots of coal but little labour.

The theory has not gone unchallenged. Mokyr (2021) suggests that high wages reflected high worker productivity; Ó Gráda and Mokyr (2023) argues instead that what set Britain apart was the quality of its workforce: an unusually large supply of skilled mechanics who could develop and maintain complex machines.

Adjudicating between these competing views is hard. Only one country industrialised first. Analysis of cross-country divergence can be analytically disappointing; comparing Britain to its contemporaries hardly leads to credible quantitative analysis. 

Wartime labour scarcity

In a new paper (Voth et al. 2026), we look at variation within England and ask: did labour-saving technology spread faster where the Napoleonic Wars induced labour shortages? By 1809, the army and militia numbered 300,000 men, with another 141,000 serving aboard Royal Navy ships. Together, they accounted for nearly 15% of England and Wales’s prime-age male population. Napoleon’s Grand Armée – relative to France’s population – was only about one-third as large. Crucially, British recruitment was highly uneven: some regions sent many more men than others (Figure 1). With low labour mobility (Smith 1776), these localised shocks created substantial regional variation in labour costs.

Figure 1 Army and Navy recruitment (log thousands recruits per capita)

Sources: Dancy (2015) and Muster rolls.

Contemporaries realised that military recruitment created labour shortages. From Yorkshire, Tuke (1800) observed: “an advance from twenty to twenty-five per cent has generally taken place [in wages], arising … from the great consumption of men in the navy and army, and consequent present extreme scarcity of hands for agricultural labour.” Systematic analysis of over 20,000 pages of Parliamentary wage reports confirms a strong positive association between recruitment intensity and wages.

Figure 2 Binscatters plotting log total recruitment per capita against agricultural wages in winter (left) and summer (right)

Sources: Dancy (2015) and muster rolls (recruitment), British Census of 1801 (population) and General View of Agriculture (wages).

Technology in the fields

We study how new technology spread in British agriculture – a sector that became one of the most dynamic of the era, adopting machines like mechanical threshing machines. Unlike manufacturing, which was geographically concentrated, farmers cultivated almost every corner of England, giving us the full geographic variation in recruitment to work with. Threshers were not new, but had existed – largely unused – for decades.

Labour shortages drove technology adoption: observers at the time already realised that “a considerable number of thrashing machines have been erected in this county … the principal inducement for using them is a scarcity of labourers, which, in a state of warfare, may be expected to be felt most in the maritime districts” (Stevenson 1812). To substantiate the link, we collect detailed data from historical newspapers advertising farm sales and leases – more than 20,000 articles. Whenever one mentioned a labour-saving machine, we geolocated it and added it to our dataset. The positive association between recruitment and machine adoption is striking (Figure 3). 

Figure 3 Map of labour-saving machines mentioned in historical newspaper advertisements (left) and binscatter plotting labour-saving machines against log total recruitment per capita (right)

Sources: British Newspaper Archive (machines), Dancy (2015) and muster rolls (recruitment) and British Census of 1801 (population).

Deep water and the Royal Navy

Correlation is not causation, and technology may have been adopted without recruitment causing it. Labour-replacing machines might have displaced rural workers, pushing them to enlist – reversing the direction of causation. Or a third factor, such as proximity to urban centres, might have induced both higher recruitment and faster adoption.

To make progress, we focus on naval recruitment, which was both massive and shaped by natural constraints. During the Napoleonic Wars, the Navy faced what Dancy (2015) calls a “problem of arithmetic”: while the Navy could draw on seamen on merchant ships, there weren’t enough of them. To fill the gap, the Navy recruited hundreds of thousands of inexperienced “landsmen” – perhaps as many as 280,000.

Naval recruitment was influenced by geography. The Royal Navy had no central HR office that would recruit and assign men. Every captain raised his crew as best he could, looking for volunteers, pressing seamen from passing merchant vessels, or drawing on the Impress Service centres. For an Englishman to be exposed to Navy recruitment, location mattered: the Navy’s largest warships – three-deckers bristling with cannons – needed deep water to anchor. We exploit this fact as an instrument: within a narrow strip of land within 15 km of the British shoreline, distance to the deep sea predicts naval recruitment but is otherwise unrelated to local economic characteristics.

This induces as-good-as-random variation: Consider two ancient hundreds in Norfolk – Clackclose and Smithdon – just 32 km apart, both on fertile coastal land near the Great Ouse. In 1801, each was home to around 1,000 people, mostly rural workers. They were similar along all observable characteristics. All except one: Smithdon faces the deep sea of East Anglia while Clackclose lies on the shallow waters of the Great Ouse estuary, 16 times further from the deep sea. The Royal Navy recruited in Smithdon but not in Clackclose, resulting in greater gender imbalances (1.05 versus 1.00 women per man) – and twice as many labour-saving machines (14 versus 7).

Our instrumental variable strategy generalises this logic across the entire coastal sample. A doubling of recruitment led to one additional machine. Since seabed depth is plausibly unrelated to local labour market dynamics in otherwise similar coastal areas, the link between recruitment, labour scarcity, and technology adoption is likely causal.

The role of skilled mechanics

Was labour scarcity all that Britain needed to take off? Machines could replace the men fighting Napoleon – but someone had to build and maintain them. To measure the role of skilled mechanics, we count how many young men apprenticed as millwrights, wheelwrights, and watchmakers in the years before the wars. These apprentices learnt the mechanical skills essential to construct the complex machines we study: their presence, in theory, should facilitate technology adoption.

This is what we find: areas with at least one mechanic adopted significantly more than those without (Figure 4, left). More strikingly, these mechanics also magnified the impact of recruitment on adoption (Figure 4, right): in areas with skilled mechanics, labour scarcity led to much more adoption than areas without it. In sum, labour scarcity and skill abundance both promoted technological progress – indeed, they reinforced each other. 

Figure 4 Machine adoption across three categories of mechanic presence (left) and binscatter of machine adoption against log recruitment per capita across three categories of mechanic presence (right) 

Sources: British Newspaper Archive (machines), British Census of 1801 (population), Dancy (2015) and muster rolls (recruitment) and records for the fees paid by apprentices (mechanics).

Conclusion

Why did England industrialise? Our research is inspired by what contemporaries saw and wrote: a giant military conflict, lasting almost a quarter of a century, pulling in vast quantities of manpower, induced labour shortages. Employers, faced with too few hands at high wages, looked to machines to keep up production. This pattern played out clearly in agriculture. Machines spread that could do the work of several men over months in a few weeks. The labour shortage–mechanisation nexus was not unique to agriculture. And without mechanical skills, no new machines could be adopted.

Our results suggest a unified interpretation of the British Industrial Revolution. Labour scarcity and skill abundance complemented each other in driving technology diffusion. These forces were already at play before the Revolutionary and Napoleonic Wars: during the “long 18th century”, Britain fought a war in one year out of every three. The labour shortages this induced caused massive take-up of labour-saving machinery, facilitating the release of rural labour to the cities – and boosting British productivity in the long run.

The lesson? Governments should stop worrying and learn to love labour shortages. Scarce labour and high wages can be a shot in the arm of the economy as long as the skills and capital necessary to adopt new machinery are there. It’s the playbook of the world’s first ‘Great Escape’ from poverty, hunger, and early death. 

References

Allen, R C (2009), The British Industrial Revolution in Global Perspective, Cambridge University Press.

Dancy, J R (2015), The Myth of the Press Gang, Boydell & Brewer.

Kelly, M, J Mokyr, and C Ó Gráda (2023), “The Mechanics of the Industrial Revolution”, Journal of Political Economy 131(1): 59–94.

Mokyr, J (2021), “The Holy Land of Industrialism”, Journal of the British Academy 9: 223–47.

Stevenson, W (1812), General View of the Agriculture of the County of Dorset, McMillan.

Tuke, J (1800), General View of the Agriculture of the North Riding of Yorkshire, McMillan.

Van Herck, K, A Kiss, and A Turrini (2026), “The Rise and Fall of EU Labour Shortages”, VoxEU.org, 26 March. 

Voth, H J, B Caprettini, and A Trew (2026), “Fighting for Growth”, CEPR Discussion Paper 21548 (forthcoming in American Economic Review).