The Wealth of Nations Was Always a Fertility Story

2026-04-18

This essay was originally published in Fertility Intelligence, Fertiligent™'s newsletter on the demographics, science, and economics shaping how we build families — and how families, in turn, shape everything else. It is reproduced here in full.

The half-sentence we forgot

In 2024, the total fertility rate in Seoul fell to 0.64 children per woman.¹ Not the country — the capital. South Korea's national TFR rose marginally to 0.75, up from an historic low of 0.72 the year before.² These numbers are not a statistical curiosity. Seoul is the first major city in human history to post a TFR that low outside of famine or war.

When Adam Smith published An Inquiry into the Nature and Causes of the Wealth of Nations in 1776, he made an assumption so obvious he barely argued for it: a nation's prosperity and its population rose together. More people meant more specialization, more specialization meant more productivity, more productivity meant more wealth. Labor was the denominator of everything. "The most decisive mark of the prosperity of any country," Smith wrote, "is the increase in the number of its inhabitants."

Two and a half centuries later, we have discovered the half-sentence Smith did not write. Prosperity and population do rise together — until they don't. And once they decouple, the compounding engine that built the modern economy starts running in reverse.

This is the defining economic story of the century. Not AI. Not geopolitics. Not the energy transition. Those are accelerants. The underlying reaction is demographic.

In this issue of Fertility Intelligence, I want to make the argument plainly: collapsing fertility is the hidden economic headwind of the 21st century, and the policy world is still operating on a mental model that expired in the 1990s.

The Becker moment, and the moment after

To understand how we got here, start with Gary Becker. In 1960, Becker published "An Economic Analysis of Fertility" and proposed what came to be called the quantity–quality tradeoff: as wages rise, the opportunity cost of raising children rises with them, and parents substitute toward fewer, better-resourced children.³ The theory earned him the 1992 Nobel and a generation of followers. It also made an empirical prediction that, by the late twentieth century, looked airtight: richer countries have fewer children.

The prediction was correct. The global TFR has fallen from 3.31 live births per woman in 1990 to 2.25 in 2024, barely above the 2.1 replacement threshold.⁴ In the wealthiest economies it fell much further — to 1.35 in Germany, 1.18 in Italy, 1.10 in Spain, 1.60 in the United States, 1.20 in Japan, 0.75 in Korea.⁵ ⁶ ⁷ The European Union as a bloc recorded 1.34 in 2024 — the lowest in its history.⁸

Total fertility rate by selected economy, 2024. Replacement level is 2.1.

Economy	TFR (2024)
South Korea	0.75
Spain	1.10
Italy	1.18
Japan	1.20
European Union	1.34
Germany	1.35
United States	1.60
World	2.25
Replacement level	2.10

Becker's model anticipated the direction. It did not anticipate the floor.

The newer economics literature — most visibly the 2023 handbook chapter by Doepke, Hannusch, Kindermann, and Tertilt, plus a follow-up strand of NBER work — documents that the old income–fertility gradient has flattened and in several high-income economies inverted.⁹ Within rich countries today, there is no reliable relationship between being richer and having fewer children. What was once a smooth curve has become a cliff. Becker's framework explained the descent; it does not explain why the descent does not stop.

Doepke and co-authors call it a "new era." I would put it more sharply: we have left the territory in which any pre-2000 fertility model is predictive. The drivers now are different — gender norms inside the household, the labor-market penalty on mothers, housing costs, credential inflation extending the age of first birth, and the rising share of couples who face biological obstacles to conception. The macro consequence is the same: the wealth engine is losing its fuel.

The math of national wealth when the denominator shrinks

GDP is, to a first approximation, the product of the number of workers and the output per worker. Demography sits under both.

Consider Japan, the country furthest into the scenario the rest of the world is driving toward. Japan's working-age population peaked at 87.3 million in 1995 and fell to 73.7 million in 2024 — a 16% decline.¹⁰ Between 1990 and 2022, Japan's real GDP grew at an annualised 0.8%, the second lowest among G7 countries. The working-age population fell by 0.46% per year over the same window.¹¹ Japan did not stagnate because its companies stopped innovating; Toyota, Sony, and Keyence remain world-class. It stagnated because the denominator kept shrinking.

China is now following the same curve, faster. Its population has declined for three consecutive years. Its working-age population has contracted at 0.8% per year since 2011, and on current trajectories China's prime-age labor force will fall 31% from its peak over the next thirty years.¹² The IMF projects that adverse demographics alone will subtract roughly 2.7 percentage points from China's annual GDP growth by the late 2020s, relative to the 2016–2018 baseline.¹³ For a country that built its geopolitical weight on a decade of 8–10% growth, losing a third of its labor force while trying to move up the value chain is not a manageable headwind. It is a structural regime change.

The IMF's April 2025 World Economic Outlook estimated that global growth will be roughly 1.1 percentage points slower than its pre-pandemic average if governments do nothing — with demographic drag accounting for most of the gap.¹⁴ Goldman Sachs' Path to 2075 work reaches similar conclusions: global population growth, around 2% per year a half-century ago, will approach zero by 2075; the global population peak has been revised downward from over 11 billion to roughly 10 billion; and global real GDP growth is forecast to drift below 2% per year over the horizon.¹⁵

None of this is speculative. These are base-case projections from the most orthodox institutions in the world, applied to cohorts that have already been born. The workforce of 2045 is in pre-school today. The workforce of 2060 is the cohort that was never conceived.

Why fewer people make a country poorer, not just smaller

There is a comforting objection that circulates in policy circles: if the population shrinks, per capita wealth could still rise. Fewer mouths, more resources, higher productivity. In a frictionless model, perhaps. In the actual economy we have built, no.

Pension and healthcare systems are calibrated to population pyramids, not population rectangles. Pay-as-you-go pension systems, Medicare-style health programs, and long-term-care infrastructure assume that each generation of retirees is supported by a larger generation of workers. When the pyramid inverts, the implicit tax on the working-age shoots up. The European Central Bank's work on ageing projects fiscal pressures of 4–9% of GDP in some member states by mid-century from this effect alone.¹⁶

Innovation scales with the number of minds at work on hard problems. The stock of researchers, founders, and skilled engineers is roughly proportional to the size of the educated working-age population. A shrinking cohort means fewer startups, fewer patents, fewer discoveries — and because innovation compounds, the loss is not linear, it is exponential across decades.

Capital formation depends on savers in their peak earning years. A population dominated by retirees dis-saves. Investment rates fall, risk capital dries up, long-horizon bets become harder to finance. Japan's "lost decades" were not primarily about bad policy; they were about a savings pool that was aging out of productive deployment.

Housing, infrastructure, and cities themselves were financed against the assumption of growth. Shrinking cities are not cheaper versions of growing cities. They are fiscally distressed versions of themselves, with the same fixed costs and a narrower tax base. Rural Japan, eastern Germany, and interior Spain already illustrate the pattern.

Becker's original framework was about household-level choices. The macro consequence he did not fully price is that the wealth of nations was never just the sum of individual productivity — it was a compounding system that required a growing substrate of people to compound against. Take the substrate away, and the system does not coast. It decays.

The pronatalist policy experiment has, broadly, failed

The most uncomfortable finding in the recent literature — and the one policy audiences resist most strongly — is that the pronatalist toolkit built over the last three decades is not moving the needle in anything close to proportion to its cost.

Hungary under Viktor Orbán is the most ambitious experiment in the OECD. Since 2010, Hungary has deployed zero-interest "baby-expecting" loans, housing grants for married couples, personal income tax exemptions for mothers of four or more, and debt forgiveness scaled to children. The country now spends roughly 5% of GDP on family subsidies — the highest share in the developed world.¹⁷ In 2024, Hungary's TFR registered 1.39, with 127,500 deaths against 77,500 births — a natural decrease of 50,000 in a country of under 10 million.¹⁸ Whatever early bounce the policies produced was largely a tempo effect — women having children slightly earlier rather than more of them — and the underlying cohort fertility has continued to erode.

South Korea has spent an estimated $270 billion on fertility and family policy since 2006. The TFR over that period fell from 1.12 to 0.75.¹⁹

France, long treated as the counter-example, is no longer one. The French TFR has slid below 1.7 and continues to fall despite the most institutionalized family-benefit system in Europe.

Anne Gauthier's 2025 review of family policies across low-fertility countries, published in Population and Development Review, concludes that while specific benefits produce measurable tempo effects, no combination of cash transfers, parental leave, and childcare subsidies has been shown to durably reverse ultra-low fertility in a developed economy.²⁰

Six economies, 1990–2024: regardless of policy regime — Hungary's 5%-of-GDP subsidies, Korea's quarter-trillion-dollar spend, France's institutionalized benefits — every trajectory declines.

The intuition is simple once you say it out loud. Cash transfers and tax credits cannot make two partners agree to have a child they have already decided, for reasons of career, housing, health, or time, not to have. They cannot bend the biological clock of a generation marrying at 33 and trying to conceive at 35. They cannot buy a couple the reproductive function that an unmanaged chronic condition or an untreated fertility disorder has already compromised.

The policy world has been trying to bribe a demographic shift that is partly cultural, partly economic, and — this is the part everyone under-counts — partly medical.

The infrastructure problem hiding inside the fertility problem

Here is the piece of the story that almost never appears in the Economist or the IMF working papers.

In high-income countries, roughly one in six adults will experience infertility at some point in their reproductive lives — a figure the WHO updated in 2023 based on pooled prevalence data.²¹ The share of first births to women over 35 has doubled in most OECD countries since 2000. IVF cycles globally passed 3 million per year. Egg-freezing volumes have grown at over 30% annually for a decade in the major markets.

This is not a niche medical story. It is a structural input to the macro story. A rising share of the fertility gap between what couples say they want (typically 2.1–2.4 children) and what they end up having (1.3–1.6) is biologically mediated — driven by delayed family formation colliding with age-related fertility decline, by undiagnosed conditions like endometriosis and PCOS, by male-factor infertility that is rarely measured, and by access barriers to reproductive medicine that make assisted conception a luxury good in most of the world.

Put differently: if you treat fertility as purely a cultural and policy problem, you are leaving an enormous lever on the table. The lever is infrastructure — the scientific, medical, emotional, and data infrastructure that determines whether the children people actually want to have can actually be conceived and carried.

From hierarchy to intelligence

Jack Dorsey and Roelof Botha published an essay in March 2026 called From Hierarchy to Intelligence.²² Its argument, stripped to one line, is that the twentieth-century firm was organized as a hierarchy because hierarchy was the cheapest available coordination technology — humans relaying information up and down layers of managers. That constraint no longer holds. "The intelligence lives in the system," they write, "and the people are on the edge, where the action is."

The essay is about companies. The frame is about everything.

The twentieth-century approach to fertility was also hierarchical. A national government set a policy — a cash transfer, a tax credit, a parental leave rule — and pushed it down through ministries, clinics, and media campaigns. Desired behavior at the edge would, in theory, follow the incentive at the top. The policy world is still operating this way. It is also, measurably, not working. Hungary at 5% of GDP. Korea at a quarter-trillion dollars. France at six decades of institutional memory. All still falling.

What is changing — quietly, mostly outside the policy conversation — is that the coordination technology itself is now available. An AI system can maintain a continuously updated model of a patient's cycle, diagnoses, emotional state, prior outcomes, genetic risk, and treatment options, and coordinate care across what used to be disconnected specialties. It can absorb the cognitive load that currently drives clinicians to 30% burnout rates and patients to abandon treatment halfway through. It can bring the best of reproductive medicine — the version that exists today only in a handful of elite clinics — to the edge, where the patient is.

This is what "fertility intelligence" means as a term of art, and why I called this newsletter that. The wealth of nations in the twenty-first century will not be rebuilt by hierarchical policy. It will be rebuilt by intelligence pushed to the edge of the reproductive journey — decision support in the exam room, companion support in the 3 a.m. wait for a beta-hCG result, longitudinal data loops from the patient back to the science.

That is the thesis behind what we are building at Fertiligent.AI. The question we are holding is not only "how do we convince people to want children." It is: for the roughly 80–120 million people of reproductive age in the OECD who already want them and are struggling, what does the reproductive stack actually look like in 2026, and how do we make it work at population scale — evidence-based on one side, empathy-driven on the other?

That is an engineering and a clinical question, not a slogan question. AI-guided cycle optimization, automated embryology, expanded access to diagnostics, integrated endocrine and metabolic care, preservation options that are routine instead of elite, and a companion layer that holds the patient's emotional reality across a multi-year process — this is the infrastructure of a fertility-intelligent economy. It is also, incidentally, the single policy lever that has been under-invested in by roughly two orders of magnitude relative to the cash-transfer budgets of the last thirty years.

What CEOs and policymakers should actually do

If you accept the argument so far, a short list follows.

Stop measuring the problem the wrong way. Period TFR is the headline number and the wrong one to track. Cohort completed fertility, the gap between desired and actual children, and reproductive-age biological markers are the honest metrics. Hungary's "success" under TFR evaporates under cohort metrics. Korea's decline is even worse than the TFR suggests.
Stop treating fertility policy as a values debate. The evidence from a dozen countries is that the desired-vs-actual gap is large — on the order of 0.5 to 0.8 children per woman — across both religious and secular populations, both conservative and progressive governments. This is an infrastructure-and-access gap, not a culture war.
Treat reproductive medicine as core public-health infrastructure. In most OECD countries, the wait time for a fertility workup is longer than the biological clock will tolerate, and the out-of-pocket cost of a single IVF cycle is a multiple of monthly household income. Treating fertility care as a boutique product rather than primary healthcare is, on a 30-year horizon, one of the most expensive policy errors a developed economy can make.
Take immigration seriously, with humility. Goldman's demographic math assumes substantial cross-border labor flows to cushion the rich-country decline. Politically, most of those assumptions look fragile. The countries that will get through the next fifty years with their wealth engines intact will be the ones that combine pro-fertility infrastructure with functioning, skill-aligned immigration systems — not one or the other.
Price demography into every long-duration decision. Corporate strategy, pension design, sovereign debt, housing stock, defense planning, AI deployment — every one of these has a demographic term in the denominator that most models still treat as noise. It is not noise. It is the signal.

The close

Adam Smith wrote that the wealth of nations rested on labor, specialization, and trade. He was describing an economy in which the labor pool expanded with every generation. The architecture of modern capitalism — pensions, public debt, innovation ecosystems, housing markets, healthcare systems — is a monument to that assumption.

The assumption has quietly broken. In 97% of the world's countries, the Lancet's Global Burden of Disease team projects, fertility will be below replacement by 2100.²³ The global TFR itself will settle somewhere around 1.59 by 2100, under half the 1990 level.²⁴ The workforce of the wealthiest economies is already shrinking. The countries still above replacement today are, almost without exception, not the countries that will be exporting capital or setting the technological frontier in 2075.

This is not a story about decline for its own sake. It is a story about what happens when the compounding engine of the modern economy runs low on its most basic input — people — and the policy apparatus built to manage it turns out to be pointed at the wrong problem.

We can build an economy fit for this century. Not by exhortation, and not by subsidy. By treating fertility as what it actually is: a piece of civilizational infrastructure, currently under-engineered, and the single most leveraged input into the wealth of nations. Move the intelligence to the edge — into the exam room, into the clinic workflow, into the patient's pocket at midnight — and a surprising amount of the gap between desired and actual fertility becomes addressable.

That is the work. It is also, I think, the opportunity of a generation.

— Sergei

Sources and citations

Georgetown Journal of International Affairs, "The Necessary Paradigm Shift for South Korea's Ultra-Low Fertility," September 2024; Statistics Korea provisional data, 2024.
Statistics Korea, Vital Statistics 2024; CNBC, "Miracle under threat: South Korea's birth rate collapse," September 2025.
Gary S. Becker, "An Economic Analysis of Fertility," in Demographic and Economic Change in Developed Countries, NBER 1960; Sveriges Riksbank Prize in Economic Sciences, 1992 press release.
United Nations Department of Economic and Social Affairs, World Population Prospects 2024 — Summary of Results (August 2024).
Eurostat, "Fertility statistics," 2024 data release; EUnews, "EU fertility falls to record low of 1.34 children per woman in 2024," March 2026.
U.S. CDC / NCHS, Births: Provisional Data for 2024, Vital Statistics Rapid Release; PBS NewsHour, "The U.S. fertility rate reached a new low in 2024," 2025.
Japan Ministry of Internal Affairs and Communications population estimates, 2024; Fortune, "Japan's population falls by half million as birth rate stays low," April 2025.
EUnews, "EU fertility drops to record low of 1.34 in 2024," March 2026.
Matthias Doepke, Anne Hannusch, Fabian Kindermann, Michèle Tertilt, "The Economics of Fertility: A New Era," Handbook of Family Economics / IZA DP 15224, 2022; IMF Finance & Development, "The New Economics of Fertility," 2022; CEPR VoxEU, "A new era in the economics of fertility."
Japan Statistics Bureau; OECD Employment Outlook 2025 — Japan.
Lundgreen's Investor Insights, "Fueling Japan's growth recovery amid demographic crises," 2025, based on Japanese Cabinet Office and World Bank data.
RAND Corporation, Fertility Decline in China and Its National, Military, and Economic Implications, RR-A3372-1, 2024; Washington Post, "China's shrinking population could crimp global ambitions," 2025.
IMF World Economic Outlook, April 2025, Chapter 2, "The Rise of the Silver Economy: Global Implications of Population Aging."
Ibid.; IMF Finance & Development, "Sustaining Growth in an Aging World," June 2025.
Goldman Sachs Research, The Path to 2075 — Slower Global Growth, But Convergence Remains Intact (Daly et al., 2022, updated 2025); The Path to 2075 — The Positive Story of Global Aging (Daly/Njie/Allen, May 2025).
European Central Bank, The Macroeconomic and Fiscal Impact of Population Ageing, Occasional Paper Series No. 296.
American Enterprise Institute / Center on Strategy and Modernization, "Hungary's Fertility Outcomes Highlight Pro-Natal Policy Limitations," 2024; Wikipedia, "Family policy in Hungary," sourced from Hungarian Central Statistical Office.
Hungarian Central Statistical Office (KSH) 2024 provisional data, via AEI analysis.
Korean Ministry of Economy and Finance budget data, compounded 2006–2024; Georgetown Journal of International Affairs, 2024.
Anne H. Gauthier, "Family Policies in Low Fertility Countries: Evidence and Reflections," Population and Development Review, 2025.
World Health Organization, Infertility Prevalence Estimates, 1990–2021, April 2023.
Jack Dorsey and Roelof Botha, "From Hierarchy to Intelligence," Block / Sequoia Capital, March 31, 2026.
Global Burden of Disease Study 2021 Fertility and Forecasting Collaborators, "Global fertility in 204 countries and territories, 1950–2021, with forecasts to 2100," The Lancet, March 2024.
Ibid.; reference scenario projection of global TFR 1.59 (95% UI 1.25–1.96) in 2100.

See it in action: Try Eva, the patient companion, or talk to our team about bringing fertility intelligence to your clinic.

Related:

Sergei Gorlovetsky, CEO, Fertiligent