Yves here. In the often dry field of macroeconomics, a new study brings encouraging results. It’s a widely accepted belief that an aging population, often referred to as demographic decline, poses a serious threat to economic growth. The prevailing thought has been that the dependency ratio—representing the number of workers compared to those not actively participating in the economy, such as children and the elderly—is key to sustaining output. The common notion is that an abundance of ‘non-productive’ individuals can hinder progress.
This article highlights an emerging argument: namely, that advances in technology have the potential to counteract the perceived negative effects of an increasing share of the population that isn’t engaged in paid work.
By Daron Acemoglu, Institute Professor in the Department of Economics Massachusetts Institute of Technology; David Autor, Ford Professor of Economics Massachusetts Institute of Technology; Keelan Beirne, PhD candidate in Economics Massachusetts Institute of Technology; and Andrew Scott, Senior Director of Economics Ellison Institute Of Technology, Oxford; Professor of Economics London Business School. Originally published at VoxEU
Over the last seventy years, global birth rates have declined, contributing to slower population growth and an increase in average age. This article argues that, despite widespread economic pessimism, historical data from various countries and U.S. commuting regions indicate that aging populations do not significantly affect overall GDP or earnings growth. Interestingly, nations with lower birth rates experience higher GDP growth per working-age adult, while regions in the U.S. with lower birth rates have faster wage increases. The drag that demographic slowdowns are presumed to have is effectively offset by rapid technological advancement. Even though future demographic shifts will unfold more swiftly and drastically, the historical record suggests that current anxieties are premature.
Editors’ note: This column is based on CEPR Discussion Paper 21673 “Baby busts and growth booms: Demographic change and the macroeconomy”.
In the past seventy years, the worldwide decline in birth rates has been stark, as illustrated in Figure 1. In 1950, the global average crude birth rate was 3.78 births per 100 people, but by 2025, this figure is projected to drop to 1.71. The United Nations anticipates that this will lead to an absolute decrease in the global population during the latter half of the 21st century, marking the first significant decline since the bubonic plague in the 14th century. Currently, nations with the lowest birth rates—like China at 0.63, Japan at 0.60, and South Korea at 0.46—are expected to see their populations shrink by 20-30% by 2050. In addition, the proportion of people aged 65 and older in these countries is projected to reach 30-40%, with older workers (aged 45 and above) making up 60% of the labor force.
Figure 1 Global trends in birth rates
Source: UN Population Prospects 2024.
Such demographic trends evoke significant concern among economists and policymakers. It is often presumed that older workers are less innovative, productive, entrepreneurial, and technologically adept than their younger counterparts. Evaluating these factors, long-term analyses project diminished economic growth (e.g., IMF 2025, OECD 2025) and worsening public finances (e.g., Cooley et al. 2024, Favero and Srivastava 2025).
These concerns mainly rely on predictive models. However, the global economy has already undergone substantial demographic shifts over the past decades, raising questions about the validity of pessimistic forecasts concerning aging populations.
In Acemoglu et al. (2026), we explore two complementary sources of evidence to investigate this issue. The first includes cross-country variations in the timing and extent of birth rate fluctuations. Our sample encompasses all countries with populations exceeding 1 million, excluding tax havens and countries in crisis, covering the period from 1950 to 2020. The second consists of regional variations in the U.S., focusing on 722 commuting zones, with workforces divided into two age categories: ‘younger’ (ages 20–45) and ‘older’ (ages 45–70).
To isolate exogenous variations in fertility that are not influenced by current or anticipated economic conditions, we primarily focus on birth rates from prior decades. For instance, the birth rate in 1950 serves as a baseline to analyze the impact of declining fertility on GDP and wage growth from 1970 to 2020.
Figure 2 illustrates our key findings from the analysis of both cross-national and U.S. regional data: countries with lower birth rates display higher GDP growth per working-age adult, and U.S. commuting zones with declining birth rates experience quicker average wage increases. There is no evidence to suggest that lower birth rates negatively affect overall GDP growth or aggregate earnings. This implies that the increases in per capita GDP and wage growth resulting from falling birth rates are adequate to offset any population decline. The economic significance of this relationship is substantial: a one-percentage-point decrease in the birth rate is linked to an approximate 22% rise in GDP per worker over 40 years and 29% over 60 years.
Figure 2 Birth rates and growth
Notes: This figure reports event study estimates of the impact of birth rates on income growth across countries and U.S. commuting zones, allowing for initial income and the population share of younger and older adults. For more details, refer to Acemoglu et al. (2026).
The observed positive relationship between GDP growth per working-age adult and average wages, combined with a lack of overall impact on GDP or aggregate earnings growth, challenges traditional economic growth models. The standard Solow/Neoclassical growth framework predicts an initial increase in GDP per worker that eventually wanes as the capital stock adjusts, leading to a decline in GDP—a pattern that our findings do not support.
Our estimates account for various factors expected to influence GDP and wage growth, including initial income, wages, and demographic composition. Furthermore, we delve into several plausible explanations for our observations. Lower birth rates might contribute to growth by encouraging higher female workforce participation, shifting child-rearing from ‘quantity’ to ‘quality’ with enhanced human capital investment per child, or reallocating labor from low-productivity sectors like agriculture to higher-productivity industries like manufacturing. However, we find no consistent support for these explanations in either the cross-country or commuting zone data. The persistent nature of our results across multiple decades (Kotschy and Bloom 2023) suggests they diverge from earlier assertions regarding a ‘demographic dividend.’
Instead, the favorable association between falling birth rates and rising GDP per worker appears to be driven primarily by technological advancements spurred by labor scarcity. Countries experiencing the most significant declines in fertility are also witnessing substantial advancements in high-tech exports and employment. Similarly, in U.S. commuting zones, those with lower birth rates are seeing shifts toward R&D-intensive industries and an uptick in labor-saving innovations (refer to Figure 3).
Figure 3 Response of employment across commuting zones
Notes: This figure reports event study estimates of the impact of birth rates on employment composition across U.S. commuting zones. For more details, see Acemoglu et al. (2026).
While lower birth rates lead to both a reduction in population growth and an increase in the number of older individuals, our key findings do not distinguish between the effects of population decline and aging. Historical wars provide a stark example that helps us separate these impacts. For instance, casualties from World War II allow us to differentiate the economic outcomes linked to civilian deaths (which primarily reduce population size without altering age composition) and military fatalities (which both decrease population and raise the average age of the workforce).
Figure 4 reveals that total war deaths correlate with lower future GDP per worker, whereas military fatalities predict higher GDP per worker. This suggests that a scarcity of younger workers is what drives the beneficial productivity response. This pattern aligns with findings by Bergeaud et al. (2025), who show that French regions with higher military casualties during WWI experienced an uptick in labor-saving patenting.
Figure 4 Response of GDP to WWII deaths
Notes: This figure reports estimates of the effect of WWII death rates on future GDP growth across countries. Refer to Acemoglu et al. (2026) for further information.
These findings expand on previous research that demonstrated how shifts in the ratio of older to younger workers can stimulate the development and adoption of labor-saving technologies (Acemoglu 2010, Acemoglu and Restrepo 2017). However, no prior investigations have documented such a noteworthy long-term increase in income per worker coinciding with declining birth rates.
The last seventy years of macroeconomic and demographic evidence do not substantiate the conventional belief—deeply ingrained in much economic discourse about demographic change—that declining fertility leads to falling living standards. However, we cannot entirely dismiss the possibility that forthcoming decades may validate the traditional viewpoint that current evidence fails to support. A significant source of uncertainty lies in the unprecedented speed and scale of demographic transitions occurring in various countries. For example, China is anticipated to see its population aged 15 to 64 years plummet from a high of 1 billion to roughly 300 million by the year 2100.
Notably, China also leads other nations in robotics technology development and application, with its industrial robot count doubling between 2021 and 2024, now comprising nearly 50% of the global total (Financial Times 2026). This indicates that the pathway towards directed technological advancement is indeed active, but it remains uncertain whether, due to the scale of China’s demographic challenges, this will suffice to stabilize GDP growth as our findings suggest has occurred over recent decades.
Additionally, the dramatic demographic transformations currently underway coincide with increasing life expectancy (Scott 2021, 2024), which may initiate institutional changes, policies, and further investments in human capital that support longer life spans.
While our demographic outlook remains relatively stable, our economic prospects are still uncertain and will evolve based on technological advancements and human investments, particularly in artificial intelligence. Drawing from the last seven decades, these factors will play a crucial role in shaping our economic future.
See original post for references.
