Not War, Not Plague — We May Have Just Crossed the Fertility Line That's Never Been Crossed Before
Summary
A September 2026 NBER working paper titled "Terra Incognita" by Jesús Fernández-Villaverde and Patrick Norrick proposes that humanity is, as of 2026, likely below replacement fertility for the first time in recorded history — a threshold not crossed even during wars or pandemics. The paper analyzes fertility data from 236 countries since 1950 using a single-factor model, finding that the global common component peaked in 1978 and that 219 of those countries are now in decline, with not one showing signs of leveling off. Perhaps the most counterintuitive finding is that the sharpest drops are concentrated not in wealthy, highly educated societies but in lower-income countries and among women with less education, upending the conventional narrative about demographic transition. While the UN's WPP 2024 estimates the current global total fertility rate at 2.25 — still technically above the standard replacement threshold of 2.1 — the paper argues that adjusting for high infant mortality in low-income countries raises the effective replacement threshold to approximately 2.20, which is already likely exceeded according to ZME Science's reporting on the study. The authors conjecture that modernity itself, rather than any single policy variable, is the underlying driver, as modern life structurally raises the cost of larger families while reducing the cost of child-free living. This analysis suggests that governments clinging to fertility recovery as a policy goal are pursuing an objective that may be structurally unachievable, and that the more productive frame is designing societies that function well with smaller, aging populations.
Key Points
For the First Time in Human History, We May Be Below Replacement Fertility
NBER Working Paper 35809, "Terra Incognita," published September 28, 2026, advances a genuinely historic claim: as of 2026, humanity is likely below replacement fertility for the first time ever. The authors — Jesús Fernández-Villaverde of Penn and Patrick Norrick of Northwestern — are careful to use the word "likely," and that qualifier matters enormously. This is an unreviewed working paper advancing a modeling-based argument, not a confirmed scientific consensus, and it should be read in that spirit. The tension with official data is real: the UN WPP 2024 estimates the current global TFR at 2.25, while this paper adjusts the effective replacement threshold upward to approximately 2.20 to account for elevated infant mortality in lower-income countries, and concludes we're likely already below that adjusted line — a position based on ZME Science's reporting on the paper's methodology. According to that same reporting, among 37 countries with high-quality birth statistics, 33 recorded fewer births in 2024 than the UN predicted, with an aggregate shortfall of approximately 1.65 million, or about 9.3%. The paper itself notes that this has never happened before — not during wars, not during pandemics — and the magnitude of that historical claim is what makes it impossible to set aside, even accounting for the pre-peer-review status. A single-factor analysis of 236 countries over seven-plus decades is not a trivial piece of work, and it deserves serious engagement regardless of how peer review ultimately goes.
219 Out of 236 Countries Are Declining, and Not One Has Leveled Off
The methodological backbone of this paper is a single-factor model applied to fertility data from 236 countries going back to 1950, and the finding is that a shared global common component to fertility trends peaked in 1978. Every country since then has been on the back side of that peak. Out of 236 countries studied, 219 are currently declining, and — here's the phrase I find most unsettling — "not one leveling off." No country has hit a floor and stabilized. The UN WPP 2024, per INED's summary, confirms that two-thirds of the world's population already lives in countries where the TFR is below 2.1. The fact that this pattern holds across wildly different political systems, cultural traditions, religious frameworks, and income levels makes it very hard to attribute to any single cause. If it were purely a function of wealth, we wouldn't see it in Colombia at 1.06. If it were purely about women's education, we wouldn't see it accelerating among less-educated women in lower-income countries. The universality is what makes this pattern both scientifically interesting and practically sobering. The IHME GBD 2021 study in The Lancet projects that by 2100, 97% of countries will be below replacement — which means the current 219-out-of-236 figure isn't a peak, it's a waypoint on a longer trend still unfolding.
The Real Surprise: The Drop Is Sharpest Among Poorer, Less-Educated Women
The conventional narrative about demographic transition runs roughly like this: as countries develop economically and women gain access to education and careers, fertility falls. Rich country, educated women, low birth rates. It's a tidy story. The NBER paper calls what it found "the real surprise," and it's this: the steepest fertility declines are not happening among wealthy, highly educated populations — they're concentrated in lower-income countries and among women with less formal education. AEI's analysis shows Colombia's 2024 TFR at 1.06, Iran at 1.44, Turkey at 1.48 — countries that were not supposed to be in this conversation yet based on the conventional model. The NBER framework identifies Thailand among countries that have experienced similarly dramatic declines. This matters enormously for policy thinking: if declining fertility were primarily about women choosing careers over children in rich, educated societies, you'd expect it to be containable and geographically bounded. But if it's spreading fastest in populations that don't fit that profile, then the driving mechanism must be something more fundamental than education or income alone. The IHME projections extend this logic: by 2100, Sub-Saharan Africa, currently at a TFR of 4.3, is projected to account for its share of global births shifting dramatically — suggesting the wave of fertility decline is simply lagging there, not absent.
Modernity Itself Is the Culprit — Not Any Single Policy Variable
One of the paper's most provocative contributions is what the authors offer as a conjecture rather than a finding: that "modernity itself" is the fundamental driver of the fertility decline. They go through the standard candidate explanations — access to contraception, women's education levels, urbanization, rising incomes — and conclude that none of them individually accounts for the specific global pattern the data shows. The deeper driver, they suggest, is the structural logic of modern life: it makes having a third child expensive in ways that can't be easily offset by policy incentives, while making child-free or small-family living increasingly comfortable and socially accepted across cultures. The paper captures this with the memorable phrase "children come in integers" — you can't have 0.7 of a child, so moving the cohort TFR from 1.8 to 1.3 doesn't require any dramatic cultural shift, just a modest increase in the share of couples stopping at one instead of two, or delaying until they don't. The policy implication is uncomfortable: if modernity is the driver, then any society that has undergone modernization is susceptible, and no set of incentive programs can fully counteract the underlying structural pull toward smaller families. This conjecture is unverified and will be tested by subsequent research, but it has the advantage of explaining the data as it actually exists — including the counterintuitive finding about lower-income countries — better than any of the conventional single-cause explanations.
Decades of Pro-Natalist Policy Have Revealed Their Structural Ceiling
South Korea has run one of the world's most aggressive and well-resourced pro-natalist campaigns for decades: birth bonuses, subsidized childcare, expanded parental leave, housing incentives — essentially every tool in the policy toolkit has been deployed at scale. The TFR still dropped to a historic low of 0.72 in 2023. The 2025 recovery to 0.80, with 254,341 births and a 6.7% year-over-year increase, is genuinely positive — two consecutive years of improvement — but the gap between 0.80 and the 2.1 replacement level is 1.3 births per woman, which is not a rounding error. The Hungary case, analyzed by NIUSSP researchers, is instructive in a different way: Hungary committed family spending at approximately 6% of GDP, among the world's highest, and did record a TFR increase from 1.25 to 1.59. But the actual count of births barely moved — from around 90,000 to roughly 92,000 — still far short of the approximately 127,000 that population replacement would require. The NIUSSP research concludes that the TFR gain was largely a tempo effect: statistical improvement from shifting when births happen, not how many total births occur. Even Nordic countries — the gold standard for family-supportive policy and gender equality — continue to see fertility decline, meaning this structural ceiling isn't a failure of implementation but a limit on what policy can achieve in a modern economy. The NBER paper's statement that "nothing in an economy pushes fertility back to 2.1" reads not as a policy indictment but as an honest description of what the data, cross-checked against the world's most ambitious policy experiments, actually shows.
Positive & Negative Analysis
Positive Aspects
- Declining Fertility Reflects — and Reinforces — Expanding Rights for Women
One dimension of this story that consistently gets lost in demographic anxiety is what's actually driving the numbers at a structural level. The conditions that produce low fertility are largely the same conditions that produce meaningful freedom of choice for women: access to education, economic independence, legal equality, and the social space to decide how to structure one's own life. Historically, societies with very high fertility rates were societies where women's options were severely constrained — where marriage and repeated pregnancy were the default path, not one option among many alternatives. The fall in fertility in lower-income countries, which the NBER paper identifies as the real surprise, can also be read as evidence that women there are gaining increasing agency over their own lives and futures. From Colombia at 1.06 to Iran at 1.44, these numbers reflect, at least in part, women making different choices than prior generations were structurally permitted to make. Framing the entire conversation around the fertility number risks obscuring this underlying structural shift, and policy approaches that treat women's bodies as population-production instruments — as some aggressive pro-natalist programs implicitly do — risk reversing the progress that partially underlies the trend itself.
- Smaller Future Populations Could Ease Long-Term Environmental Pressure
Eight billion people generate an enormous ecological footprint in carbon emissions, resource extraction, habitat loss, and waste production. The prospect of a smaller future global population isn't just a demographic abstraction; it has real-world implications for the planet's capacity to recover and stabilize over the long run. A structurally declining population means structurally declining aggregate demand for fossil fuels, land use, water, and raw materials over time — not in the near term, but meaningfully over the second half of the century and beyond. The IHME GBD 2021 study projects a global TFR of approximately 1.6 by 2100, which on a long enough timeline implies significant population reduction from the projected peak. Fewer people dividing finite resources also creates structural conditions for improving quality of life on a per-capita basis: more educational investment per child, less pressure on healthcare infrastructure per patient, more housing per household. The transition itself is painful and carries real near-term costs, but the destination — a smaller, more sustainable population with higher per-capita resource allocation — is not obviously worse than the alternative trajectory of continued growth against planetary limits.
- Labor Scarcity Drives Automation Investment and Productivity Innovation
The economic pressure created by population decline generates powerful incentives for technological investment that can, in the right conditions, substantially offset the direct impact of a smaller workforce. Japan and South Korea already rank among the world's highest in manufacturing robot density — a direct response to decades of labor market tightening driven by demographic contraction. AI and automation don't just feature in the tech sector; they become a systemic economic response to the structural reality of fewer workers at every level of the economy. When companies cannot rely on a steady supply of labor, they invest in capital that multiplies the productivity of the workers they have. This can maintain or even increase per-capita output even as the total workforce shrinks. I think of this as a "forced innovation" dynamic: it's not comfortable, and it doesn't happen automatically without deliberate investment, but it does create real incentives that wouldn't exist if population were growing. The potential for AI-assisted services, automated manufacturing, and productivity-enhancing infrastructure to compensate for demographic decline is real, even if realizing it requires sustained policy support and private investment commitment.
- Population Decline Forces Long-Overdue Structural Redesign of Social Systems
There's a version of this demographic story that isn't a crisis but a renovation opportunity — a compelled reckoning with the fact that many core social systems were designed for demographic realities that no longer apply. Pay-as-you-go pension systems made sense when five workers supported one retiree; they need fundamental rethinking when that ratio approaches two to one. Education systems built for large youth cohorts need to shift toward lifelong learning models. Urban planning that assumed continuous residential growth needs to engage with intelligent density management and adaptive reuse of existing built environment. These are painful transitions, but they are also genuine opportunities to build systems actually fit for the world that's coming rather than the world that's receding. Societies that embrace this redesign early — that stop trying to inflate their populations back to an earlier configuration and instead optimize for smaller, older, more technologically capable demographics — will likely end up with better-functioning institutions than those that spend another generation defending systems built for demographics that no longer exist. The NBER paper's finding that no economic mechanism pushes fertility back to 2.1 is, from this angle, clarifying rather than depressing: it closes one unproductive door and points toward others that are more genuinely open.
Concerns
- Welfare Systems Face Structural Solvency Crises Without Comprehensive Reform
The most immediate and concrete risk of sustained below-replacement fertility is what it does to the financial architecture of social insurance systems. Pay-as-you-go pension systems — which fund current retirees through contributions from current workers — are deeply sensitive to the ratio of contributors to beneficiaries. As that ratio deteriorates with demographic decline, maintaining current benefit levels requires either significantly higher contribution rates from a shrinking pool of workers, cuts to benefits, or some combination of both. This is not a theoretical future problem: it is already a live fiscal and political crisis in multiple countries. South Korea's national pension fund solvency has been a recurring domestic political flashpoint, and the underlying trajectory — fewer workers, longer-lived retirees — is exactly what below-replacement fertility produces at scale. The UN WPP 2024 data, per INED's analysis, shows that two-thirds of the world's population already lives in countries below 2.1 TFR, meaning this is a near-global structural challenge, not a localized edge case. Healthcare systems face parallel pressures: older populations generate substantially higher per-capita healthcare demand at the same time the working-age taxpayer base that finances those systems is shrinking. Without systematic and politically difficult reform, the fiscal math of these welfare systems becomes increasingly untenable across the developed world.
- Long-Term Economic Growth Faces Structural Deceleration
Population growth has historically been one of the reliable engines of economic expansion, through both labor supply and consumer demand. Demographic decline works against both simultaneously. Fewer workers means lower potential output absent significant productivity offsets. Fewer young households means structurally lower demand in key growth sectors — residential construction, education, entry-level consumer goods. The NBER paper's conclusion explicitly names slow growth as a key economic consequence of the demographic trend it identifies. If the paper's model correctly implies that the global population peak could arrive much earlier than the UN projects, then much of the long-range economic planning built on UN demographic assumptions is calibrated to a larger future economy than will actually materialize. That misalignment between projection and reality — in pension funding, infrastructure investment, sovereign debt issuance — creates systemic risk that compounds over time. The IHME's Lancet projections of a global TFR near 1.6 by 2100 reinforce the view that this isn't a temporary disruption but a structural shift that will define economic conditions for generations to come. Technology and productivity growth can partially compensate, but industries directly tied to population size — real estate, domestic consumer goods, certain educational services — face irreversible demand contraction that no amount of automation resolves.
- Global Demographic Imbalance Will Intensify Immigration Politics and Geopolitical Tension
The extreme divergence between Sub-Saharan Africa's current TFR of 4.3 and the sub-replacement levels of East Asia, Southern Europe, and much of Latin America creates the conditions for significant geopolitical friction around migration in the decades ahead. The IHME GBD 2021 study projects Sub-Saharan Africa's share of global births rising to 54% by 2100 from 29% in 2021. The temptation for aging, labor-scarce wealthy nations to view this as a convenient demographic complement — abundant youth supply in one region, scarcity in another — ignores both the declining fertility trends in traditional emigration source countries and the serious ethical problems with framing African young people as a labor reserve for wealthier societies. Competition for globally mobile workers will intensify as source country demographics tighten — Colombia is already at 1.06 — and that competition increasingly intersects with national security concerns, sovereignty narratives, and cultural anxieties in ways that make rational immigration policy genuinely harder to achieve. At the global level, immigration is zero-sum for population balance: one country's immigration gain is another's demographic loss. The political dynamics around this will get significantly harder before they get easier, and current policy frameworks in most countries are not well-equipped to handle the pressures that are coming.
- Social Vitality and Cultural Dynamism May Erode With Rapid Aging
Beyond the fiscal and economic dimensions, sustained below-replacement fertility combined with rapid aging raises subtler but real concerns about the vitality and adaptive capacity of societies over time. Entrepreneurship, cultural innovation, political reform movements, and social experimentation have historically been disproportionately driven by people in their twenties, thirties, and early forties. As the absolute number of people in those age brackets shrinks — and as the political weight of older cohorts increases proportionally — societies may find it structurally harder to take risks, embrace disruption, and sustain the kind of experimental energy that produces long-run adaptability and institutional renewal. Consumer markets shaped by aging demographics also tend to be more conservative and risk-averse, which can reduce incentives for companies to invest in genuinely new product categories and services. South Korea's average age at first birth of 33.2 in 2025 is itself a structural signal: as childbearing shifts later, the next generation's birth cohort will be smaller, and the population will continue aging in ways that compound over decades. The IHME's projected TFR of approximately 1.37 for Western Europe by 2100 implies a level of demographic aging in those societies that represents a genuinely unprecedented shift in age structure, one whose full institutional and cultural consequences are difficult to model but should not be assumed benign.
Outlook
Let me think through what actually happens next — across the near term, medium term, and long term — starting with the academic dimension, because this paper's first gauntlet is the peer review process.
The NBER working paper dropped on September 28, 2026, and it hasn't yet been through peer review. That makes the months ahead particularly interesting from a scholarly standpoint. The paper's single-factor model and its adjusted replacement threshold of approximately 2.20 will attract rigorous critical scrutiny from demographers worldwide. I expect the debate to center on two technical fault lines: first, whether the single-factor modeling approach adequately captures the regional heterogeneity that clearly exists across the data; and second, whether the upward adjustment to the replacement rate threshold is methodologically defensible given the existing uncertainty in infant mortality data for lower-income countries. The gap between the UN WPP 2024's estimate of 2.25 and this paper's conclusion that we're likely already below the effective replacement threshold is large enough that it demands explanation. Subsequent research will either close that gap or clarify its sources. ZME Science's reporting notes that among 37 countries with high-quality birth statistics, 33 recorded fewer births in 2024 than UN models predicted — a shortfall of about 1.65 million births, or 9.3% — which is the kind of empirical signal that will pull other researchers into this space quickly.
At the national policy level, I think this paper's ripple effects will be felt in ways that aren't always visible as direct responses to the paper itself. South Korea is the obvious focal point. The country recorded a TFR of 0.80 in 2025, with 254,341 births and two consecutive years of modest recovery — a positive signal in a country that hit 0.72 in 2023. But the structural context is sobering: 37.3% of mothers were 35 or older in 2025, and the average age at first birth was 33.2 years. When the modal childbearing age is 33, population-level arithmetic works very differently than when it's 27. Even a sustained TFR recovery to higher levels still leaves a massive gap relative to replacement. The NBER paper's core argument — that nothing in an economy pushes fertility back to 2.1 — lands differently in countries that have been pouring resources into pro-natalist programs without sustainable results. It gives political cover, and perhaps intellectual permission, to redirect spending from fertility recovery toward adaptation infrastructure. I think we'll see some countries begin to formally articulate that pivot in their demographic policy documents.
The medium-term picture — looking across the next three to seven years — centers on whether a meaningful policy consensus shifts from fertility recovery to what I'd call "demographic adaptation." This isn't just about Korea. Colombia's TFR at 1.06, Iran at 1.44, Turkey at 1.48 — these are middle-income countries that spent decades being categorized as high-fertility developing nations and are now below levels that Northern European countries would have found alarming a generation ago. The speed of this transition is genuinely new. The IHME GBD 2021 study, published in The Lancet, already projected that 155 of 204 countries would be below replacement by 2050, and 198 of 204 — that's 97% — by 2100. If the NBER working paper's implied timeline is correct, that future is arriving ahead of schedule. I think we're likely to see a wave of policy strategy documents that stop using TFR recovery targets and start using different metrics: healthy aging rates, effective labor force participation ratios, per-capita productivity growth, and immigration integration efficiency. That's not surrender — that's good management.
The economic structure will also shift visibly over this medium-term window. Labor scarcity is already a chronic constraint in a number of advanced economies, and that pressure will intensify. What's underappreciated is that this isn't just a labor supply problem — it's also a consumer demand problem. Fewer young people means structurally smaller markets for housing, education, children's goods, and entry-level services. Industries that have been sized for a growing or stable population will need to right-size or pivot. At the same time, silver economy sectors — elder care, senior-oriented healthcare technology, retirement housing, cognitive health services — will grow faster than nearly any other segment. I believe the senior economy will become a central global investment theme in a way it hasn't been before, simply because the demographics make it structurally inevitable. Real estate markets will face a bifurcation: large family-sized housing in areas without strong senior migration patterns will lose demand, while well-serviced urban units remain under pressure.
Immigration policy will also become significantly more contested over this period. At the national level, immigration is still the most practical short-term lever for labor market balance. But there's a paradox that hasn't fully entered public discourse yet: the traditional source countries for migration are themselves aging. Colombia's TFR at 1.06 means there's no demographic surplus of young people waiting to emigrate. The competition for globally mobile workers will intensify, and that competition will increasingly intersect with security and sovereignty concerns in ways that make it a much harder political conversation than a purely economic one. I think immigration will become one of the defining geopolitical friction points ahead — not because of ideology alone, but because the underlying demographic math makes globally mobile young workers a genuinely scarce resource that nations will compete to attract.
Looking further out — to the horizon of five to ten years — the key question is how the expert forecast consensus about population peak timing evolves. The UN WPP 2024 projects a global population peak in the mid-2080s at roughly 10.3 billion. According to ZME Science's reporting on the NBER working paper, the model implies a significantly earlier peak — closer to 2056 and at a level closer to 9 billion — a difference of roughly three decades. That is not a rounding error; it is a fundamentally different picture of the demographic century. Even the UN's own projections show global TFR falling to approximately 2.07 by 2050, which is below the standard 2.1 threshold — meaning the direction of travel is not seriously disputed. What's disputed is how fast. If the actual trajectory tracks closer to the NBER model's implied path, then virtually every long-range planning assumption in pension finance, infrastructure investment, and national debt management is built on a demographic foundation that overestimates future population size.
The IHME's GBD 2021 projections for 2100 paint a picture that would have seemed almost unimaginably dramatic even ten years ago: global TFR at approximately 1.6, 97% of countries below replacement, and Sub-Saharan Africa representing 54% of global births while lower-income and lower-middle-income countries collectively represent 77%. If those projections hold, we are not looking at a temporary demographic transition — we are looking at a fundamental restructuring of where and to whom the world's next generations are born. The projected TFR for Western Europe by 2100 in this study is approximately 1.37, implying a degree of demographic aging in those societies that would be structurally unprecedented. The societies that understand this structural shift earliest and build for it most deliberately will have a significant long-run advantage over those that spend another decade in policy denial.
Let me frame the scenarios explicitly. In the optimistic scenario, a combination of genuinely comprehensive family support policies, intelligent immigration integration, and productivity gains from AI and automation means that population decline in advanced economies is slower and more manageable than feared. Fertility rebounds modestly in the most policy-responsive countries. Social systems are reformed in time to remain functional. The global population peak arrives roughly when the UN projects, the trajectory down is gentle, and the transition is navigated without systemic shocks. In the baseline scenario, current trends persist broadly. The peak arrives somewhere between the UN and NBER model timelines. Most countries gradually accept that fertility recovery to 2.1 is not achievable and shift toward adaptation planning — a real but manageable renegotiation of social contracts around pension and healthcare provision.
In the pessimistic scenario, the NBER model's implied timeline is closer to reality: the population peak arrives in the 2050s, the transition is sharper than welfare systems were designed for, and the politics of aging societies — where older voters structurally dominate elections — make it difficult to implement the forward-looking reforms that adaptation requires. Pension systems in multiple countries face actuarial crises. Youth-to-retiree ratios become politically untenable. Geopolitical tensions around immigration and labor competition escalate into genuine friction. I think the most likely reality lies somewhere between the baseline and pessimistic scenarios. The direction — population decline — is not seriously contested. The variable is how quickly institutions adapt, and history doesn't inspire unconditional optimism about institutional agility in the face of slow-moving structural problems.
One final thought: I could be wrong about the speed of all this, and I want to be explicit about that. Demographic forecasting at multi-decade horizons is inherently uncertain, and population projections have frequently surprised in both directions. Breakthroughs in reproductive medicine could shift the biological constraints on later-in-life fertility. Cultural transformations could occur that I genuinely cannot anticipate. The best thing I can say with confidence is this: the data as we currently have it, from 236 countries over more than seven decades, is pointing in one direction with very few exceptions. If you are thinking about your career, your retirement planning, your city, your industry — the assumption to build around is not "things will return to normal." The productive assumption is: how do we make a smaller, older world work well? The title "Terra Incognita" is apt. The map runs out here. What we draw next is up to us.
Sources / References
- Terra Incognita: The Economics of a Shrinking World (NBER Working Paper 35809) — NBER (National Bureau of Economic Research) — Terra Incognita: The Economics of a Shrinking World (NBER Working Paper 35809)
- South Korea 2025 Birth Statistics — Korea Policy Briefing (Government of the Republic of Korea) — South Korea 2025 Birth Statistics
- UN World Population Prospects 2024 — INED Commentary — INED (Institut National d'Études Démographiques) — UN World Population Prospects 2024 — INED Commentary
- Global Fertility Rate Expected to Drop Below Replacement Level — EurekAlert / Lancet IHME GBD 2021 — Global Fertility Rate Expected to Drop Below Replacement Level
- Evaluating Pronatalist Policies with TFR Brings Misleading Conclusions — NIUSSP (International Union for the Scientific Study of Population) — Evaluating Pronatalist Policies with TFR Brings Misleading Conclusions
- The Global Fertility Crisis Is Worse Than You Think — AEI (American Enterprise Institute) — The Global Fertility Crisis Is Worse Than You Think
- Below Replacement Fertility: World Population May Peak Earlier Than Expected — ZME Science — Below Replacement Fertility: World Population May Peak Earlier Than Expected