Technological innovation is a key driver of long-run economic growth (Romer 1990, Aghion and Howitt 1992). The forces that drive technological change are complex. On the one hand, innovation depends on incentives, markets, and strong institutions. At the same time, innovation systems also rely on having people – scientists, engineers, and skilled workers – who can generate, develop, and share new ideas and knowledge.

One question that has garnered considerable interest from economists and policymakers alike is: what happens to innovation when societies experience events such as wars, demographic shocks, and mass migrations?

Recent work has emphasised how the resulting labour shortages can incentivise firms to invest in developing and adopting labour-saving technologies (Voth et al. 2023). As a result, a scarcity of labour can sometimes foster faster technological change (Acemoglu 2002, Hanlon 2015). But this neglects the fact that innovation also depends on the supply of ideas and skills. When shocks disproportionately remove highly educated or technically trained individuals, the consequences for innovation may be very different. The loss of scientists, engineers, and other skilled workers may disrupt knowledge networks and weaken local innovation ecosystems. In turn, this can harm the productivity of established inventors and reduce the emergence of future innovators.

In a new working paper (Repetto et al. 2026), we examine this human capital side of adverse shocks using one of the great events of the 20th century: the deaths of young men during World War I.

The ‘lost generation’

World War I caused enormous damage across Europe. In Britain alone, more than 750,000 servicemen died during the conflict. Because the war was fought largely abroad, this massive shock to the country’s human capital was not accompanied by the physical destruction experienced elsewhere in many continental economies.

The loss of so many young men on the battlefields had many social, economic, and cultural legacies (Carozzi et al. 2024). Of particular note is that Britain began the war without conscription and instead asked for volunteers. Vast numbers of educated and talented young men answered the call and, in the absence of established allocation systems, soon found themselves in frontline combat roles. Mortality rates would be particularly high for alumni of universities and public schools, many of whom as junior officers led their troops ‘over the top’. Contemporaries quickly recognised the social consequences of the loss of the most talented men of the cohorts and referred to them as a ‘lost generation’ (Winter 1977).

Despite the prominence of this idea in cultural and historical discussions, we still know relatively little about how the lost generation affected economic outcomes over the long run. Did the loss of so many young men from social groups that disproportionately produced engineers, scientists, and future innovators reduce the capacity of local communities to generate new ideas? Or did labour scarcity instead encourage the development of new technologies?

To answer this question, we combine several newly digitised historical datasets covering England and Wales. To measure how heavily different communities were affected by wartime losses, we use military records from the Commonwealth War Graves Commission and other archival sources that identify the place of origin of soldiers who died (Figure 1). We then link these records to a comprehensive dataset of patents filed in Britain between 1895 and 1979, geolocated to the communities where inventors lived (Figure 2). The resulting dataset covers roughly 750,000 patents and more than 50,000 inventors.

This approach allows us to examine how innovation trajectories evolved over the decades in places that experienced different levels of wartime mortality. Importantly, by comparing communities within the same county, we can estimate the effects of very localised deaths on innovation while holding fixed the market-wide changes in incentives to innovate brought about by a general scarcity of labour.

WWI mortality decreases communities’ innovation potential

Our results show that communities that lost more soldiers in World War I experienced persistent declines in innovation (Figure 3). A 10% increase in wartime deaths reduced the probability that a community produced any patent in a given period by roughly 0.09–0.12 percentage points. Patenting at the local level was relatively rare, so this represents a meaningful decline in innovative activity. Importantly, these effects persisted for decades. Communities that experienced heavier wartime losses continued to produce fewer patents throughout the 20th century.

Figure 3 Effect of WWI mortality (log) on probability of producing any patent

The effect was even stronger for high-impact inventions (Figure 4). When we focus on patents that rank among the most important innovations of their time, the negative effects of wartime mortality become roughly three times larger. In other words, the loss of human capital did not simply reduce the number of inventions. It also disproportionately affected breakthrough innovations.

Figure 4 Effect of WWI mortality (log) on probability of producing any patent, by patent importance

Inventors working at the frontier were affected the most

Why did wartime mortality have such persistent effects? One possibility is purely mechanical: if inventors themselves died in the war, innovation would naturally decline. However, our analysis shows that this explanation cannot fully account for the results. Even when we exclude inventors who died during the war, we still find that inventors living in heavily affected communities became less productive. These inventors produced fewer patents after the war, and the decline was especially pronounced for complex or highly innovative inventions.

The effects were strongest in technologically advanced sectors such as electricity, machinery, transportation, physics, and chemistry (Figure 5). These were the sectors where innovation relied most heavily on specialised skills, collaboration, and technical expertise. This pattern suggests that the war disrupted the local networks and complementary skills that supported innovation. When communities lost engineers, technicians, and other skilled workers, the productivity of surviving inventors declined.

Figure 5 Effect by inventor’s sector (of first patent)

The losses also affected the next generation of innovators. Communities that experienced higher wartime mortality produced fewer new inventors in subsequent decades. This decline cannot be explained simply by population changes. Even after accounting for differences in local population size, heavily affected areas generated fewer successful new patentees.

We find that the loss of particularly skilled individuals played an important role. Communities that lost engineers or graduates from elite educational institutions experienced especially large declines in the emergence of new inventors. This pattern is consistent with the idea that talented individuals play an important role not only in generating new ideas themselves, but also in inspiring and mentoring future innovators.

Despite these negative effects, some inventors were able to adapt. Inventors exposed to heavy wartime losses were more likely to relocate to other regions. Those who moved to more dynamic innovation environments were able to partially offset the negative effects on their productivity (in line with previous evidence on exposure effects, e.g. Bell et al. 2019). Collaboration also helped. Inventors working with co-authors were better able to maintain their innovative output, while solo inventors experienced larger declines. This finding highlights the importance of collaborative networks in sustaining innovation.

However, these adjustments only partly compensated for the initial shock. At the community level, innovation declined persistently in areas that experienced the greatest wartime losses.

These findings contribute to a growing literature on the long-run economic consequences of war (Kóczán and Chupilkin 2022). Previous research has shown that conflicts can have lasting effects on economic development, institutions, and social outcomes (Davis and Weinstein 2002, Voigtländer and Voth 2013). Our results highlight another important channel: the loss of human capital can permanently reshape the geography of innovation.

While labour scarcity can sometimes stimulate technological change, our evidence suggests that the loss of highly skilled individuals may instead weaken a society’s capacity to generate new ideas. Innovation depends on networks of skilled individuals and complementary expertise. When these networks are disrupted, the effects may persist for decades.

Lessons for today’s world

Although World War I took place more than a century ago, the lessons remain relevant today. Conflicts around the world continue apace. In many cases, wars lead to large losses among young cohorts, which may include engineers, scientists, and other skilled professionals. Our findings suggest that the deaths of talented individuals may have consequences that extend far beyond the immediate tragedy and destruction. By reducing the supply of talent and weakening innovation networks, wars may also reduce a country’s capacity to generate new technologies and sustain economic growth. In this sense, the economic legacy of war may include not only destroyed capital and disrupted institutions, but also the loss of ideas and innovations.

References

Acemoglu, D (2002), “Directed technical change”, Review of Economic Studies 69(4): 781–809.

Aghion, P, and P Howitt (1992), “A model of growth through creative destruction”, Econometrica 60(2): 323–51.

Bell, A, R Chetty, X Jaravel, N Petkova, and J Van Reenen (2019), “Who becomes an inventor in America? The importance of exposure to innovation”, Quarterly Journal of Economics 134(2): 647–713.

Carozzi, F, E Pinchbeck, and L Repetto (2024),“The experience of war shapes communities and future wars”, VoxEU.org, 20 January.

Davis, D R, and D E Weinstein (2002), “Bones, bombs, and break points: The geography of economic activity”, American Economic Review 92(5): 1269–89.

Hanlon, W W (2015), “Necessity is the mother of invention: Input supplies and directed technical change”, Economic Journal 125(587): 1456–93.

Kóczán, Z, and M Chupilkin (2022), “The economic consequences of war”, VoxEU.org, 14 December.

Repetto, L, D Cipullo, E Pinchbeck, and J Bietenbeck (2026), “Human-capital shocks and innovation: Evidence from Britain’s Lost Generation”, CESifo Working Paper 12529.

Romer, P M (1990), “Endogenous technological change”, Journal of Political Economy 98(5, Part 2): S71–S102.

Winter, J (1977), The Great War and the British people, Macmillan.