Humanity has a variety of short-term problems and long-term problems. The latter are typically harder to address, because the harms manifest so slowly. By the time things get severe enough to cause a popular outcry, the underlying causes may have become entrenched. Unlike the proverbial frog in a boiling pot of water, it’s hard for humanity to just jump out once things get hot enough.
The long-term problem that people usually talk about is climate change, and that’s certainly worth worrying about. Fortunately, with the advent of cheap green energy, we now have the means to address that problem. But there’s another challenge that’s slowly creeping up on the human race, which we don’t have a good answer to: population aging and decline.
The U.S. used to be an outlier in terms of fertility rates. But since 2008, its rate has fallen by almost a quarter — due largely to shrinking fertility among Hispanic Americans — and now stands well below the rate of 2.1 required for long-term population stability:
Because of this, the optimistic population projections of earlier decades have abruptly evaporated:
But after the next two or three decades, where will the immigrants come from? The whole world is experiencing the same trend. Over the last decade, Asia — where the bulk of the human population currently lives — has seen a big decline in fertility, while Latin America has seen continued declines. Both are now below the replacement rate:
Religion has proven no barrier to this. The most devoutly Islamic countries of the Middle East, North Africa, and Central Asia have all seen huge fertility declines.
Some forecasters even predict that the continent’s population will peak in less than four decades. Even if the rest of the world is willing to take in very large numbers of African immigrants (doubtful given recent political shifts), Africa is not going to ride to humanity’s demographic rescue for very long.
And to make things worse, the trend toward lower fertility rates doesn’t seem to have any bottom. In past decades, many demographers assumed that family size would fall to around 1.4 to 1.6 children and then stay there indefinitely, or even recover a bit. But in recent years, fertility in China and South Korea has crashed to levels once thought to be unimaginably low:
As things stand today, no one on the entire planet knows how to stop this trend.
In response to the looming global shrinkage of the human race, some people choose to simply shrug their shoulders and say things like “I’m fine with a less crowded country”. But this is like responding to climate change by saying you enjoy warm weather. There are plenty of reasons why an aging, shrinking population is a threat to the living standards that we currently, and to ignore these is simply to bury your head in the sand.
Why a shrinking, aging population is a threat to our way of life
In fact, I wrote a post about the economics of population aging a little over a year ago:
There are basically three economic harms from population aging and shrinkage:
A higher dependency rate — every working-age adult has to toil harder and consume less in order to support a growing number of people who are too old to work
Lower productivity growth — there are reasons to think that an older, shrinking population is less capable of generating continuous improvements in living standards
Financial and macroeconomic disruption — a shrinking population disrupts systems like Social Security, creates negative expected returns in asset markets, and puts downward pressure on interest rates
The first of these is the simplest to understand. In the 1990s and 2000s, there were more than 5 working-age Americans (age 15-64) for every elderly American (64+). By 2021, there were fewer than 4. That means that the economic burden of supporting each elderly American is now shared among only 4 people instead of 5.
And in other rich countries, it’s even worse. In France, there are only 3 working-age people for every elderly person. In Japan, there are only two:
If you shrug at population aging, it means that you think it’s fine for either A) working-age people to toil for most of their waking lives and pay enormous taxes, or B) old people to live in poverty. I do not think either of those sounds fine. That does not sound like an enticing, optimistic future to me.
The usual counter to this is productivity growth — if each worker produces more output, then aging doesn’t need to be so much of a burden on the young. This is not a great argument to begin with, because for any given amount of productivity growth, population aging means either more work and/or lower living standards. It’s not like we can just snap our fingers and make productivity growth increase to compensate for aging — we’re already trying to increase it as fast as we can.
But on top of that, there’s a good amount of evidence that population aging reduces productivity growth! Here I’ll just quote from my post last year:
Researchers aren’t in complete agreement, but the general preponderance of evidence seems to suggest that aging exerts a persistent drag on rich economies — not just through a greater proportion of retirees, but via lower productivity growth. Here’s a graph of what that might look like, from a paper that I think does a particularly good job of isolating the causal impact of aging, by looking at the impact of demographic factors that were “baked into the cake” long ago:
These numbers might be a little confusing, so let’s unpack what this means. In 2000, ~16% of Americans were over 60; by the late 2020s this is projected to be around 24%. That’s about a 50% increase in the percentage (sorry, I know that “percent of a percent” is a bit confusing). So according to these estimates, that means the U.S.’s total labor productivity increase over that time frame would be expected to be about 18.5 percentage points lower due to aging. Given that it’s just the early 2020s now, and labor productivity has already increased by about half since the year 2000, an 18.5 percentage point drop is substantial but hardly crippling.
The aggregate data show a clear relationship between an older workforce and lower productivity at the state-industry level, in both cross-section and panel models. The results are confirmed using employer-employee linked data from private sector human resource company ADP that show having older coworkers reduces an individual’s wages…The results suggest that between a quarter and full percentage point of the slowing in annual productivity growth in recent years may be due to aging[.]
We don’t exactly know why this happens. One theory is that when companies become crowded with older managers, they’re less able to pivot to new business models, embrace new technologies, and perceive new trends in demand. Entrepreneurship tends to decline in people’s 50s and 60s as well. The need to support more old people might make young people take fewer risks. Old workers might simply be less productive at certain tasks (the data on that is unclear). But whatever the reason, the result seems pretty consistent — aging is a drag on productivity.
A lot of people respond to concerns about aging by saying that we’ll just automate our way out of it — replacing humans with robots. But this is just another way of saying “we’ll increase productivity to compensate”. All of these empirical results about aging slowing down productivity already include all the automation that aging societies do in order to bolster their decreasing labor forces.
So far, I’ve been talking about the effects of aging on productivity. There’s also the possibility that a shrinking population will make it harder to improve technology. This is because a shrinking population means a shrinking pool of people to do research. Chad Jones, probably the top growth theorist in the econ profession today, made this argument in a 2022 paper:
We show below—first in models with exogenous population growth and then later in a model with endogenous fertility—that negative population growth can be particularly harmful. When population growth is negative, both endogenous and semi-endogenous growth models produce what we call the Empty Planet result: knowledge and living standards stagnate for a population that gradually vanishes…
The “idea value of people” is tied to the flow of new ideas that are created at each point in time and is, at least partially, a positive externality that would lead optimal population growth to exceed the equilibrium rate in many models. With negative population growth, however, the flow of new ideas goes to zero. It is this force that allows the optimal allocation to be trapped in an Empty Planet[.]
If you want to see the math of how this result emerges, check out the paper, which also has some other interesting ideas. But the basic principle — that fewer people means a smaller pool of researchers — is pretty easy to understand.
In fact, there are probably other reasons to suspect that a shrinking population will lead to slower — or even negative — growth in living standards. The theory of economic agglomeration shows how urbanization drives growth by allowing producers, consumer, and workers to all live near each other. With a shrinking population, this process goes into reverse — cities will shrink, but also the network of cities that trades with each other will shrink, making humanity poorer.
Yet another potential mechanism, which I alluded to above, has to do with finance. A shrinking population drives down rates of return for companies — fewer customers means less profit growth in the future. That pretty obviously discourages investment. Who wants to build more factories and offices and develop new products to serve a shrinking market? And if investment falls, then investment in research and development will fall, slowing the rate at which new ideas are found.
In fact, if the risk-adjusted expected rate of return goes below zero, companies have no incentive to invest at all. Why would you invest in projects that will lose money, when you can just hold cash instead? At that point, companies just stop investing at all, and you get the dreaded secular stagnation.
So there are lots of mechanisms by which an aging, shrinking population will make humanity poorer. Some of these are theoretical, and some rely on empirical estimates, but some are just simple and incontrovertible. JD Vance might have been insensitive with his comments about “childless cat ladies”, but when he calls low fertility a “catastrophic problem”, he’s absolutely 100% right.
When I talk about these problems to technologists, they sometimes declare that AI will save us — or even that since AI is about to replace all the human workers, a smaller population is a good thing. Indeed, it’s theoretically possible that AI will be able to substitute for human researchers, ushering in a golden age of discovery despite population decline.
But it’s also very possible that this will not happen — that the inherent limitations of AI technology, or limitations on energy, compute, and data, will prevent the golden age of virtual researchers from riding to the rescue of a shrinking planet. It’s possible that AI will always need human researchers to complement it. And if that’s the case, then we have no choice to but to address the problem of low fertility, or else face the economic consequences.
The problem is, we currently have no real idea how to address the problem.
Wanted: policies that increase the fertility rate
Here’s the thing about exponential functions: They either explode to infinity or decay to zero.
Human population growth is an exponential. This is because humans are the thing that produces other humans — the more you have, the more you produce. If every human, on average, produces 1.000001 other humans, the population will eventually explode to infinity; if every human, on average, produces 0.999999 other humans, the population will eventually decay to zero. Here are pictures of the two possible futures:
Graphs by GPT-4o
So how do we create a world where the human population neither explodes into overpopulation or trails off into an empty planet? The only way is active stabilization. We need ways to nudge fertility a little bit above the replacement rate when it’s too low, or knock it down a little bit below the replacement rate when it’s too high. We don’t want to send fertility back to 5 or 7, but we need some way to be able to return it to 2.1 if we want to.
Right now, we know lots of policies that can reduce the fertility rate — in fact, those policies have been used to great effect around the world, which is one reason population forecasts have fallen so much. We’re good on that front. But what we don’t know are policies that reliably and durably increase the fertility rate by an appreciable amount. We have downward nudges, but we don’t have upward nudges yet.
There are two main classes of “pronatal” policies that people talk about — 1) paying people to have more kids, and 2) trying to change culture to encourage people to have more kids. Unfortunately, the first of these has only had minor effects so far, while nobody really knows how to do the second.
There’s actually a great economic argument for paying people to have more kids. Child rearing is economically valuable labor, but as things stand, it’s totally uncompensated — people spend tens of thousands of hours of their life, and hundreds of thousands of dollars, basically because they feel like it. But since population growth boosts productivity, parents are adding to the future of the economy in ways they’re not getting paid for. So the thinking is: Just pay them!
In fact, there is evidence that paying people to have kids does get them to have more kids — but only a little bit. Lyman Stone did a review of the literature, and plotted the numbers that studies have found:
These numbers on the y-axis are elasticities. They’re answers to the question “If you give people child subsidies of X% of their income, in present-value terms, by what percent do their birth rates increase?”. And the average number is about 0.25X%, meaning that if you want to boost fertility by 25% — about the amount it has fallen in the U.S. since 2008 — you would need to pay them the equivalent of one extra year of their income today. That might not sound like a lot, but it is a lot. Here’s what Stone wrote:
In 2020, the total fertility rate is probably around 1.71 children per woman. Thus, to reach 2.07, we would need a 21% increase in birth rates. To accomplish this, we would need the present value of child benefits to increase by somewhere between 52% and 400% of household income. For the median woman, this would mean providing a child benefit for the first 18 years of a child’s life worth approximately $5,300 per year in addition to currently-provided benefits, with the range running from $2,800 more per year to $23,000 more per year. This would require a budgetary commitment many times greater than that envisioned by the Bennet-Romney [child tax credit] proposal.
For countries with even lower fertility rates, like South Korea, the payment would be far larger.
That could well be worth it in economic terms. Duncan McClements and Jason Housenloy make a simple model of the positive externalities of promoting fertility, and they find that the breakeven point of pronatal policy effectiveness is pretty close to where the literature says it is. They conclude that it’s worth paying parents up to $290,000 each to have more kids.
Imagine if every mom and every dad got a check $290,000 as soon as they had each child! That would probably get a lot of people to reorient their lifestyles around having kids. But the problem is that politically, this is probably too large a sum to be even remotely feasible. Remember that Biden’s expanded child tax credit, which was $3000 to $3600 per child per year, was scrapped after it proved politically unpopular.
And this doesn’t even take into account the labor disincentive of a policy like that. A very large and careful recent study found that just getting $1000 a month caused a detectable number of people to stop working. A payment several times that large would incentivize even more people to stay out of the labor force. The cost this would impose on the economy would be in addition to the fiscal cost of the taxes needed to pay for the child benefit. It seems unlikely that the American electorate — or almost any electorate — would countenance much higher taxes on working people in order to pay people with children to stay home and not work, even if this did create a net economic benefit for future generations.
And that’s even assuming that the optimistic estimates are right. It’s possible that the research literature is overestimating the effect of pronatal child subsidies. South Korea has reportedly spent $200 billion over the last 16 years to try to raise birth rates, only to see its fertility crash to the lowest level any country has ever recorded.
There is one interesting alternative to cash benefits. Providing free child care does seem to move the needle a little bit on fertility. This is from Flowers et al. (2024):
We build upon previous work by investigating the regulation of childcare and fertility gaps across the U.S. states. Our results consistently show fewer childcare regulations are associated with smaller fertility gaps. This suggests that women are better able to achieve their fertility goals in policy environments that allow for more flexibility in childcare options and lower costs.
An expanded version of this would be to make everything about raising a child free — free medical care, free babysitting, free baby supplies, and so on. If child-rearing could be transformed into a nearly effortless activity — if all the hard parts could be outsourced to industry, leaving the fun parts for parents to enjoy — it might make young people feel that having kids is less of a scary lifestyle transition. This definitely deserves more investigation.
Anyway, the other main idea people have had for pronatal policy is to change our culture. There are certainly pockets of high fertility in developed countries — for example, strict religious communities like Ultra-Orthodox Jews and Mormons. While few would propose reorganizing our society along the lines of Ultra-Orthodox Judaism or Mormonism, it’s possible that if we nudged our culture a little bit in the direction of family-friendliness, or religiosity, or toward norms of child-rearing in general, we could achieve a much more modest version of the same effect.
Unfortunately, no one has really figured out how to produce such a cultural nudge. A number of Asian countries have organized matchmaking events in an attempt to encourage young people to get married; this has failed spectacularly. Meanwhile, right-wing folks in the U.S. often decry low fertility or even rail against childless people, but it’s not clear how this behavior will actually get anyone to have more kids.
In fact, even if countries encouraged the rise of strict religion, it’s not clear this could overcome whatever tide of modernity is driving lower fertility rates. Saudi Arabia, Iran, and Afghanistan famously encouraged and enforced strict Islam for many decades, and here are the results:
Meanwhile, Mormons in Utah have been experiencing falling fertility lately as well, going from 3.1 in 2010 to 2.3 — just over replacement level — a decade later. And in Israel, fertility among Orthodox Jews is now declining.
Women in Nagi, a town of about 5,700 people, on average bear more than two children. That figure makes it a standout in a country where the average is closer to one than two…In 2019, [Nagi’s fertility rate] hit 2.95—the average number of babies a woman would bear if conditions that year lasted permanently. The number slipped a bit the next two years but still was 2.68 in 2021, the latest year for which data are available…
[V]isitors learn that parents pay no more than $420 a month for daycare for their first child, half that price for their second child and no charge for a third. Parents get the equivalent of $1,000 a year for each child in high school. Caregivers also get help from elderly women who look after children for a nominal sum.
Pronatal policies are obviously part of the story here, but there’s also the possibility that Nagi has produced a social contagion of high fertility. Plenty of research papers find that having kids is socially contagious — when people’s friends have kids, they become more likely to have kids too, even controlling for a vast array of other factors. Maybe the solution to low fertility is to use concentrated local pronatal policies to create communities where having lots of kids is the norm.
This is a promising approach, but one problem here is that people don’t just choose how many kids to have — they also choose where to live. If you create communities where everyone has lots of kids, it’s likely that people who want to have lots of kids will move there from other places. This will reduce the number of kids in other places, potentially accelerating fertility decline outside of the targeted communities. (In fact, it’s possible that part of Nagi’s high fertility rate is due to a selection effect; the town is noted for attracting young couples who are already eager to have children.)
A final reason to doubt the power of cultural approaches is that women in the U.S. already want more children than they’re having. Since around 2008, the number of children that American women say they want has actually risen slightly, to around 3, while the number of children per woman has fallen to 1.6:
If we take women at their word — and if we assume American men want the same — then a culture that encourages people to want more children might not succeed in actually changing behavior.
All this is very speculative, of course. There are certainly plenty of cultural interventions left to try — most of which we haven’t even thought of yet. Some might work. But as of today, in 2024, there’s no known lever we can push on to change the culture around child-rearing.
But I think we need to step up our efforts to find pronatal policies that work. My worry here isn’t just economic. If population aging and shrinkage becomes severe enough, I worry that some countries will resort to extremely draconian solutions to increase fertility.
We’re all familiar with The Handmaid’s Tale, a dystopian sci-fi story in which women are subjugated in order to increase birth rates. I doubt that will happen; it’s more likely that societies would just collapse if they tried something that extreme. But I can think of more realistic, yet still cruel and dystopian, policies to force birth rates up. The most obvious would be to exclude childless people from old-age pensions and medical care.
In the past, when fertility rates were high, children served an economic purpose — they were farm labor, and they were also people’s old-age pension. If parents lived past the point where they were physically able to work, their children were expected to support them. In order to make sure you had at least a few kids who survived long enough to support you, you had to have a large family.
Denying old-age benefits to the childless would be an obvious way to try to reproduce this premodern pattern. This would, of course, result in horrific widespread old-age poverty for those who didn’t comply. I would never want to see my country, or any country, implement such a policy. But at the same time, I predict that some authoritarian states — China, perhaps, or Russia, or North Korea — will eventually turn to ideas like this if no one ever finds a way to raise fertility voluntarily.
A world where free societies dwindle and fall into poverty, and totalitarian societies only survive by violating the bodily autonomy of their citizens, is not a future I want to live in. That’s why I think free societies need to step up their efforts to find policies that help people to have more children. Perhaps these will be some combination of economic incentives — free child care, pronatal subsidies, cheaper housing, and so on. Perhaps they’ll include policies to change cultural norms around child-rearing and family. Maybe they’ll even include technological solutions — AI nannies that offload much of the burden of parenting.
But whatever those solutions end up being, we need to step up the search.
