The empty cradle
Why the world stopped having children?
In 1960, the average woman on Earth had five children over her lifetime. Today she has about 2.3. In South Korea, that number has fallen to 0.72, less than one child per woman, the lowest fertility rate ever recorded for any nation in human history.
Japan sits at 1.15. China at 1.09. Italy at 1.2. Spain at 1.1.
These aren’t temporary dips caused by a single recession or pandemic. They’re a structural change in human behavior that’s been building for over a century, is now accelerating across every major region, and will reshape economies, governments, and societies in ways most people aren’t yet taking seriously.
More than two-thirds of humanity now lives in a country where the fertility rate has fallen below the replacement level of roughly 2.1 children per woman. The UN projects the global population will peak sometime in the 2080s and then begin a sustained decline. Some demographers think the global fertility rate may have already slipped below replacement.
This is the opposite of an overpopulation story. And it’s one of the most consequential forces shaping the next century.
How we got here: the first transition
To understand what’s happening now, you have to start a long way back.
For most of human history, fertility was high because death was high. In pre-industrial Europe, roughly one in three children died before the age of five. Families needed many births to maintain household labor, ensure support in old age, and simply sustain the community across generations. Having many children wasn’t a lifestyle choice. It was rational economics under conditions of extreme uncertainty.
The first major shift came with industrialization and the decline of child mortality, what demographers call the demographic transition. As medicine improved and food supply stabilized, child survival rates rose sharply. This created an initial population boom: birth rates stayed high while death rates fell. But within a generation or two, families adapted. When parents could expect their children to survive, they had fewer of them. The logic of the large family dissolved.
By the late 19th century, fertility had already begun falling in Western Europe. Demographers expected it to stabilize around replacement level, roughly two children per woman. In most models, that was the endpoint. It didn’t stabilize.
The second shift nobody fully predicted
A deeper transition began in the 1960s and 1970s, driven by forces that classical demographic theory didn’t anticipate.
The arrival of reliable contraception gave women, for the first time in history, genuine control over whether and when they had children. Childbearing had always been tied to biological inevitability as much as social choice. That changed in less than a decade, and fertility rates fell sharply wherever contraception became widely available.
Then women entered the formal labor market in large numbers. As they gained access to education and professional careers, the opportunity cost of leaving work to have children rose dramatically. A woman with a university degree and a professional career gives up significantly more income and career progression by having children than her grandmother did. Economists call this the “foregone income” effect. It’s one of the most consistently documented drivers of lower fertility across countries and time periods.
Then came urbanization. In rural agricultural economies, children contributed economically to the household from an early age, they worked the land, tended animals, eventually took over the farm. In cities, children are purely a financial cost for many years. Housing is expensive and small. Education is long and costly. Children produce nothing economically for over two decades. The move from farm to city quietly transformed children from economic assets into economic liabilities, not emotionally, but financially.
By the early 1970s, these forces were producing fertility rates below replacement across much of Western Europe. Demographers named the phenomenon the Second Demographic Transition: as societies became more individualistic, secular, and educated, people would prioritize self-fulfillment over family formation, and fertility would settle well below replacement.
They were substantially right. What they underestimated was how far below replacement fertility would eventually fall, and how quickly the same dynamics would spread to countries where economists expected them far later.
Why it’s accelerating now
In previous generations, the demographic transition happened after countries became wealthy. Britain, France, Germany, and the United States all saw fertility fall below replacement only after achieving high levels of income, education, and urbanization. Development came first, fertility decline followed.
That sequence is breaking down. Countries like Bangladesh, India, and Vietnam are seeing fertility fall to or below replacement at far earlier stages of economic development. The same forces are arriving faster: smartphones are compressing the education and urbanization process. Social media is spreading norms of delayed family formation globally. The financial pressures that drove fertility down in rich countries are being felt in middle-income countries that lack the resources to offset them.
The housing problem is probably the most underrated driver. Research from the University of Toronto and multiple European institutions has documented a strong relationship between housing costs and fertility. When rents rise relative to incomes, young people delay marriage and household formation. A delay in first birth often means fewer total births, because biological fertility declines with age. Between 1990 and 2020, rents rose 149 percent across the United States while the fertility rate fell from 2.08 to 1.64. Cities with the most expensive housing consistently show the lowest birth rates - Tokyo, London, San Francisco, Seoul.
The cost of raising children has also risen sharply in East Asia, driven by intense educational competition. In South Korea, private tutoring centers called hagwons have become near-mandatory. Monthly attendance in middle and high school runs 300,000 to 500,000 Korean won, with elite facilities in Seoul’s Gangnam district charging double that. The total estimated cost of raising a child from birth through university in South Korea sits somewhere between $225,000 and $375,000, among the highest anywhere in the world.
When one child costs that much, three becomes economically unimaginable for most families.
South Korea as the laboratory
South Korea is a preview. What it’s experiencing today, other countries are on track to experience within one to three decades.
In the 1960s, South Korea’s fertility rate was 6.0. The government, then focused on rapid industrialization, actively ran antinatalist campaigns, smaller families were publicly encouraged. Those campaigns worked. By 1983, fertility had fallen to replacement level. By 2001, it had fallen to 1.3. By 2023, 0.72.
The South Korean government has spent over 280 trillion won, roughly $210 billion, on pronatalist programs since 2006. Cash payments for births. Expanded childcare. Subsidized fertility treatments. Housing incentives. Parental leave policies. The fertility rate during that same period fell another 33 percent.
The failure of those incentives tells you something important about what’s actually driving the decline. Families aren’t declining to have children because a cash payment wasn’t large enough. The entire structure of modern urban life - the housing market, the labor market, the education system, the distribution of caregiving responsibilities between men and women, has become deeply hostile to family formation. You can’t write a check large enough to fix a structural problem.
Sweden and France, which have had comprehensive family support systems for decades, maintain higher fertility rates, typically between 1.5 and 1.8. Neither has reached replacement either. But the gap between 1.8 and 0.72 is partly about policy. Countries that invested in affordable housing, universal childcare, and genuine workplace flexibility decades ago have bad-but-manageable demographic trajectories. Countries that didn’t now face something much harder to reverse.
What happens to an economy when it ages
Demographics are the slowest-moving of all major economic forces. They announce themselves decades in advance and still manage to catch governments off guard.
The core mechanism is simple. GDP growth depends on the number of workers and how productive each worker is. An aging, shrinking population creates pressure on both. Fewer young workers enter the labor force. The ratio of working-age adults to retirees, the dependency ratio, gets worse. Every active worker must support more people who’ve stopped working.
Japan has lived this in real time for thirty years and provides the clearest picture of where declining fertility leads. Japan’s working-age population peaked in the mid-1990s and has been falling since. The IMF estimated in 2018 that Japan’s economic growth would decline by an average of 0.8 percent per year over the following four decades due to demographics alone. Japan’s public debt has exceeded 250 percent of GDP, driven substantially by the costs of supporting an enormous elderly population.
Today, roughly 29 percent of Japan’s population is over 65. By 2050, that’s projected to exceed 40 percent, the highest elderly share of any major country in the world.
China faces the same trajectory in a more compressed and economically dangerous form. Japan got wealthy before it got old. China is getting old before it gets wealthy. The one-child policy, which ran from 1979 to 2015, suppressed fertility for decades and also created what demographers call the 4-2-1 problem: one working-age adult now potentially carries two parents and four grandparents. China’s pension system covers a smaller share of its population than Japan’s did at a comparable demographic stage. The AEI estimates that at current rates, China could lose as many as 600 million inhabitants by end of century.
Across Europe, the politically contentious fights over raising retirement ages in France, Germany, and the Netherlands are directly connected to this arithmetic. When more people live into their eighties and nineties while fewer young workers enter the economy, the pension math breaks. There’s no painless way to adjust.
Pension systems designed in the 1950s and 1960s assumed a particular ratio of workers to retirees. At the start of Social Security in the 1930s, the U.S. had roughly 40 workers per retiree. Today it’s approximately 2.8. Politicians have largely avoided confronting what that means.
Who gains and who doesn’t
Africa is the exception that complicates the picture.
Sub-Saharan Africa still has fertility rates exceeding four children per woman across much of the region, though they’re also declining. The IMF estimates that favorable demographics could add roughly one percentage point to annual GDP growth in countries in what economists call the “demographic dividend” phase, when a large, growing working-age population enters the labor force. Nigeria, Ethiopia, and the Democratic Republic of Congo are projected to see significant population growth over the coming decades.
But that advantage is conditional. Demographic dividends materialize only if economies can generate enough jobs for the young. Education and skills investment must keep pace. If working-age populations grow faster than employment opportunities, the dividend becomes a liability, producing the youth unemployment and instability that have characterized parts of the Middle East and North Africa, where an earlier dividend was largely squandered.
Several advanced economies are trying to manage declining populations through immigration. Germany absorbed over a million refugees and migrants in 2015 alone. Canada has built high immigration into its explicit demographic strategy. The U.S., despite periodic political controversy, has been sustained by immigration in ways that have kept its population trajectory better than most peers.
But immigration can only move workers around the globe, not create them. If fertility falls below replacement nearly everywhere, as current trends suggest it might, the pool of potential migrants shrinks alongside the populations sending them.
What governments keep getting wrong
The instinct is to write checks.
Baby bonuses, child tax credits, and one-time payments for newborns have been tried across dozens of countries. The evidence that these cash transfers meaningfully shift fertility rates is weak. Families don’t decide to have an additional child because of a one-time payment that doesn’t cover even a small fraction of the total cost. What research consistently shows is that the costs driving the decline are structural: housing, childcare, career penalties for mothers, the expectation that raising children well requires enormous ongoing investment of money and time.
The gender dimension gets consistently underweighted. In countries where gender equality in the labor market has advanced faster than gender equality at home, women face a stark set of tradeoffs. Full-time professional and primary caregiver are both expected. Many women rationally choose career over children, or have fewer children than they’d prefer. Research from Japan, South Korea, and southern Europe, all with very low fertility and significant gender inequality in domestic labor, supports this connection consistently. Countries where men take more caregiving responsibility have higher fertility.
Countries most in denial about their demographic problem tend to be the ones with the strongest cultural and institutional resistance to changing gender norms. Offering cash while leaving intact the social expectations that make having children expensive and career-damaging for women is spending money to avoid the harder conversation.
What comes next
The slow-motion nature of demographic change creates a false sense that something can always be done later. It can’t. The decisions made in the next ten to twenty years will determine trajectories that play out over the next century.
A few things seem reasonably predictable. Fiscal pressure on pension and healthcare systems in aging economies will intensify. Governments will face choices that aren’t comfortable: raise taxes, cut benefits, raise retirement ages, or run larger debts. Political systems will strain as elections become increasingly dominated by older voters whose preferences systematically differ from the younger workers bearing the costs. Countries that delay these conversations face sharper adjustments.
Labor shortages in advanced economies will persist and deepen. Construction, healthcare, agriculture, and elder care itself will compete harder for a smaller pool of workers. That will push wages up in some sectors and accelerate automation in others.
AI and automation may partially offset the effects of a shrinking workforce. If productivity per worker rises fast enough, an economy can maintain output even as the number of workers falls. Japan has invested heavily in robotics as an explicit demographic response. But productivity gains from technology are historically uneven in timing and distribution, and they may not arrive fast enough, or broadly enough, to compensate for the pace of decline in the most affected countries.
The deeper question is whether any government will address structural causes rather than symptoms. The countries with even partial success, Sweden, France, and to some extent Germany, did so through investments made over decades in affordable housing, childcare, gender equality in the workplace, and family-friendly labor markets. These are expensive and politically difficult. They require planning for outcomes that will only be visible thirty years later.
Most governments don’t plan thirty years out. That gap between the timescale of demographic problems and the timescale of democratic politics is probably the hardest part of this challenge.
Closing thoughts
Here’s a thought experiment worth sitting with.
In 1960, the average South Korean woman had six children. In 2023, she had 0.72. That change happened within a single lifetime. A woman born in Seoul in 1960 into a large family has lived to see her granddaughters, statistically speaking, have no children at all.
Nothing in human history produced that kind of demographic shift that fast. Not wars, not famines, not plagues. It happened through millions of individual rational choices: to pursue education, to prioritize careers, to wait until housing is more affordable, to do the math on what raising a child in a competitive city actually costs, and to conclude that one is enough, or zero is more honest.
Most of those choices are individually reasonable. The person who decides against a third child because they can’t afford the apartment upgrade, the woman who delays her first child until her career is established only to find that biological windows have narrowed, they’re responding rationally to the world as it is. The problem is that millions of rational individual choices can aggregate into a collective outcome nobody designed and many wouldn’t have chosen.
The world built over the past century, its pension systems, cities, labor markets, and growth models, assumes a particular demographic shape: more young than old, more workers than retirees, a population that grows or stays stable. That assumption is dissolving quietly and quickly.
Understanding how we got here is the first step. Figuring out what kind of world to build for the population that’s actually coming is the harder one.
Let us know what you think of this post in the comments section! We hope this piece helped illuminate the deeper forces behind a system many of us interact with, but rarely stop to examine. If it resonated, feel free to share it with others who enjoy thoughtful, long-form analysis.