When the issue is climate change, a prevalent rationalist take goes something like this:
"Climate change would be a top priority if it weren't for technological progress. However, because technological advances will likely help us to either mitigate the harms from climate change or will create much bigger problems on their own, we probably shouldn't prioritize climate change too much."
We could say the same thing about these trends of demographic aging that you highlight. So, I'm curious why you're drawn to this topic and where the normative motivation in your writing is coming from.
In the post, you use normative language like, "This suggests that we need to lower costs along many fronts of both money and time, and also we need to stop telling people to wait until they meet very high bars." (In the context of addressing people's cited reasons for why they haven't had kids – money, insecurity about money, not being able to affords kids or the house to raise them in, and mental health.)
The way I conceptualize it, one can zoom in on different, plausibly-normatively-central elements of the situation:
(1) The perspective of existing people.
1a Nation-scale economic issues from an aging demographic, such as collapse of pension schemes, economic stagnation from the aging workforce, etc.
1b Individual happiness and life satisfaction (e.g., a claim that having children tends to make people happier, also applying to parents 'on the margin,' people who, if we hadn't enouraged them, would have decided against children).
(2) Some axiological perspective that considers the interests of both existing and newly created people/beings.
It seems uncontroversial that both 1a and 1b are important perspectives, but it's not obvious to me whether 1a is a practical priority for us in light of technological progress (cf the parallel to climate change) or how the empirics of 1b shake out (whether parents 'on the margin' are indeed happier). (I'm not saying 1b is necessarily controversial – for all I know, maybe the science already exists and is pretty clear. I'm just saying: I'm not personally informed on the topic even though I have read your series of posts on fertility.)
And then, (2) seems altogether subjective and controversial in the sense that smart people hold different views on whether it's all-things-considered good to encourage people to have lower standards for bringing new people into existence. Also, there are strong reasons (I've written up a thorough case for this here and here) why we shouldn't expect there to be an objective answer on "how to do axiology?."
This series would IMO benefit from a "Why I care about this?" note, because without it, I get the feeling of "Zvi is criticizing things government do/don't do in a way that might underhandedly bias readers into thinking that the implied normative views on population ethics are unquestioningly correct." The way I see it, governments are probably indeed behaving irrationally here given them not being bought into the prevalent rationalist worldview on imminent technological progress (and that's an okay thing to sneer at), but this doesn't mean that we have to go "boo!" to all things associated with not choosing children, and "yeah!" to all things associated with choosing them.
That said, I still found the specific information in these roundups interesting, since this is clearly a large societal trend and it's interesting to think through causes, implications, etc.
There is little sign that the momentum of the situation is changing. Instead, things continue to slowly get worse, as nations in holes continue to keep digging. The longer we wait, the more expensive the ultimate price will be. We will soon find out what the new administration does, which could go any number of ways.
Table of Contents
Not Enough Dakka
South Korea since 2006 has spent just over 1% of GDP on baby making incentives.
It is not doing that much. But then what would you expect?
As Emmett Shear says, wake me when it’s a lot higher. Which they are indeed proposing. I do not understand why people like Tim Carney respond the opposite way.
Yes. Yes I will. It’s not enough money.
Paid sick leave decreases, rather than increases, fertility. This is despite paid sick leave being obviously very helpful when being pregnant, and also when having a child, which makes you sick more often and gives such flexibility higher value in general. Women today want more children than they have. If paid sick leave is decreasing fertility, something is going very wrong.
Why would this have the opposite effect? Could this be a wealth effect, a classic case of giving people what you think they want, rather than what they actually want or need? The paper’s suggestion that this facilitates use of birth control? How does that even interact with ‘sick leave’ and why is this even an issue? Something else?
My best guess is that this essentially forces the women to take more sick leave, but in a way that they see as hurting their careers, or that increases how much they expect their careers to be hurt, or causes employers to try and stop women from having kids, or some combination thereof. So things end up getting worse. But that still feels like a weird ‘just so’ story.
Are Hungary’s pro-fertility policies failing? We’ve looked before, but here we go again.
Technically if you add up all family support benefits of all kinds, including pre-existing ones, I do see claims this adds to 5%, versus an OECD average of 2.1%. The majority of that 5% was from pre-existing supports, the change was only 1%-2% of GDP. That’s not nothing, but a long cry from 5%. America is unusually low here, we mostly do our transfers using other methods.
More Births responds that they are very much running ahead of the obvious comparable countries. Depending on when you start, they are doing relatively well, and certainly doing far better than many other places.
Unfortunately Hungary in 2024 then had a 17% decline in June compared to 2023. How you view that depends on how much you factor in where they started.
The FT has an article portraying the whole thing as a clear failure despite generous subsidies. In their central example a family got 80k Euros, and had as many kids as was medically safe for them to have. Their explanation is that the boost in fertility was merely a shift in getting a generation to have their kids earlier.
I don’t buy this for several reasons, including that if you attempt to have kids earlier you are going to end up with more kids, if only because you might change your mind and might have medical issues if you wait, or you might then have time to decide to have more.
One strong argument here is that the subsidies are structured badly. If you give people tax breaks, then that helps once you are already well off, but the poor who would be most sensitive to subsidies get left out. And if you are then rewarding earning more by being regressive, that cuts against prioritizing having a big family.
I would also point out that everything we know says you want to prioritize getting parents cash money quickly. That impacts behavior far more than long term subsidies. Giving people tax breaks after several children is exactly the kind of move that is not going to get much impact on fertility per dollar spent.
We do know that family policy can boost marriage rates.
As someone who had to decide whether to get married, I can verify this absolutely makes a huge difference. The incentives here can be very, very large. If we had gotten married earlier, it would have plausibly cost six figures in lost financial aid.
Also, it seems quite obvious to me that if you boost the marriage rate, you also boost the birth rate. As in, yes, being counterfactually married should quite obviously lead to decisions to have more children. So should giving married people better financial conditions relative to the unmarried, over and above changing people’s marriage decisions, although the size of that mechanism is reasonably disputed. How could these things fail to be true?
Aside from Poland those are not adjacent countries, and some similar countries did well without similar subsidies (on a world-relative basis) but none of this seems suggestive of subsidies not working.
France offers proposal for free fertility tests at ages 18-25, women’s groups are ‘outraged’ because they focused on physical fertility rather than making kids affordable. How dare they provide information about reproductive health. Yes, affordability is the bigger issue. I still will never understand the attitude of ‘this is a good thing and in no way interferes with other good things but is not the best possible thing, so we are going to be outraged you proposed it.’
Except yes, Macron’s plan also includes financial support for new parents. Alas, no number is mentioned here. I am confident the number is far too low, because Macron would never dare, and also because if the number was high enough I would have heard.
The plan also emphasizes the importance of paternal involvement in a child’s life, including a proposed ‘duty to visit’ for fathers. I worry this would actively discourage fertility far more than it encourages it. I also am not convinced, if a father needs a law to be there, that you want him around.
Surrogacy, it says, is excluded due to ‘ethical concerns.’ Still with this nonsense, although that’s a lot less bad than Italy going nuts and banning surrogacy outright.
A new post puts America’s socially optimal fertility rate at 2.4, and estimates we should place a value of $1.17 million on each additional birth, and to do this should be willing to spend $290k per birth. They suggest greatly increasing the child tax credit.
Remember that my estimates of the effective cost to induce a new birth are consistently in the realm of $300k in marginal spending, roughly a quarter of this social value and equal to the paper’s proposed willingness to pay. There are better ways to do this than writing checks, but writing checks works, and it works better the more you frontload the payments, and the more you pay outright in cash.
Embryo Selection
The correct value of effective polygenic screening is highly unlike zero. If you think there is nothing wrong with it, and it works, you substantially improve your offspring’s outcomes on a variety of metrics, as per your definition of improve. Yes, IVF is highly annoying and expensive, but the upside is huge. 30% of people recognize this, and 55% would at least consider it.
General approval is very, very good. 67%-11% say benefits outweigh costs, 77%-12% would have it be legal. Good show, everyone.
The range of approvals for different outcomes mostly matches what you would expect.
Everyone hates physical diseases like cancer and heart disease. Preventing some mental diseases are mostly unobjectionable (and yes the implications of that sentence are as crazy as they sound). If it is a ‘condition,’ people are mostly fine preventing it. Obesity is the most objectionable, but seems like a very clear place to have a preference, given its impact on health and other life outcomes. Whether or not you think obesity is a person’s choice given their genes, you should want to be able to select against it.
Traits, on the other hand, give people the willies. Eliminating bad things is different from looking for good things in people’s minds. A lot of this is framing. Note how much worse ‘BMI’ does than ‘obesity.’ Even more so ‘life satisfaction’ versus ‘depression.’ There’s a strong anti-vanity streak here, given the opposition to making your child not bald. And a highly reasonable big jump at the top on skin color, while noticing that you do also choose the parents.
To some extent I sympathize. You don’t want people to race to give their children genetic positional goods, forcing others to follow or be left behind, with no social gains. But you do want to give them absolute goods that make people healthier, happier, smarter, more productive and so on, including well above the median.
Thus my number one disliked trait selection would be height. Height is mostly a positional good. We should save our selection pressure for positive sum games. Personality traits should be handled with caution. Mostly I would want to invest available trait stats into intelligence and constitution, but if you could also offer me strength, dexterity or wisdom, or the positive sum forms of charisma, I’m definitely listening.
Costs
Reddit is asked why more people are not wanting kids. The answer comes back loud and clear, and that answer is:
Also that people have been taught not to have kids until the money is sorted out, whereas in the past people would more often muddle through. Expected time and attention spent on kids also gets mentioned, both kids being treated as needing vastly more supervision and there being less others around to help with that.
Another one mentioned a few times is mental illness. A lot of people are now diagnosed with mental illness, which is some combination of increased diagnosis and viewing things differently, and also higher rates of mental illness. That leaves a lot of people not wanting to pass that on to their kids, or terrified they can’t be good parents.
This suggests that we need to lower costs along many fronts of both money and time, and also we need to stop telling people to wait until they meet very high bars.
Proving that Dakka Works
How much does having children lower lifetime earnings?
Maxwell Tabarrok cites a new Danish study of women who attempted IVF, and concludes it has strong evidence that having children does not reduce long term earnings. There is a correlation, but he concludes it is not causal.
That seems weird. Children are a huge time sink and you are forced to take time off. How could that not matter? Maxwell says women largely time their kids to correspond to counterfactual earnings peaks, which says a lot about how much money is driving lower fertility. And the counterforce to less time is higher motivation and justification.
If you do not have kids, it is very easy to satisfice on money, to choose more rewarding or less stressful jobs or those with less hours, and end up earning less, because you can. Similarly, when you negotiate salaries and such, saying you have to support a family is a strong argument, as I have witnessed many times. So it is not so crazy to me that these effects might roughly cancel out.
I would go a step farther than Maxwell does in the conclusion. If the result is correct, then it shows that financial considerations are greatly warping fertility choices. If that is true, then well-structured payments and other incentives can greatly change those choices.
You cannot have a world where women are carefully timing kids to not interfere with their earning potential, and also have a $70k baby bonus (as proposed in South Korea) not make a huge difference. Even better, you could vary or condition the bonus based on timing.
One can also look to this paper on the willingness to pay for IVF. They see no long-term ‘protective’ effects (of having no child) on earnings.
IVF
How much are people willing to pay for IVF if they are infertile? This varies really quite a lot. The majority of the time the answer is $0, or actually far less than $0. Other times, the answer is almost anything if they think it will work. People very highly value their fertility preferences. I do not think that marginal willingness to pay is a good measure of overall welfare gains in this spot.
A lot of people do end up in the middle as well, if only due to of inability to pay.
This offers us another insight. If at least 40% of response to insurance on IVF is liquidity effects, then it would stand to reason that 40% of the response to child subsidies would also depend on it addressing liquidity effects.
In other words: If I offer a $10k subsidy payable over time, versus a $6k subsidy payable on birth (and perhaps even partially before?), we should expect those to have similar fertility impacts. You really, really want to do cash on delivery.
IVF also has other disadvantages. It is highly uncertain, and people with moderate willingness to pay are going to be risk averse on that, although this could conflict with the liquidity issue. IVF is physically highly uncomfortable, if it was a trivial procedure willingness to pay would likely go up. There is also certainly some ‘it is unnatural’ tax, and the risk of dealing with multiple babies at once is not fun either. IVF is wonderful, but you’d pay even more to get a natural conception, if IVF is not also being used to do any form of embryo selection.
But what is the lower bound being offered here? 22% of annual income for a 30% chance of having an a child is 73% of annual income per child.
If nothing else, this seems like overwhelmingly strong evidence that IVF should be fully covered by insurance or by the state for all infertile couples, in all areas with below replacement fertility. It is the lowest hanging of fruits.
IVF for embryo selection beyond avoiding particular health concerns alas remains remarkably unpopular. The term ‘ruining it for everyone’ seems relevant, and now we have to deal with the consequences. The good news is that if you stick to health concerns, people are mostly sane about this, with 72% approval (versus 11% disapproval) for screening in general and similar for doing it for health. And 82% said they’d be at least somewhat interested conditional on already using IVF.
Genetics
Lyman Stone argues extensively that we should not expect genetic selection to get us out of our fertility problems any time soon. I think that at the limit ‘life finds a way’ applies no matter what your simulations and correlations tell you, but we should not rely on anything like this as a practical solution.
Japan to become the second country to allow gene editing before birth.
Cultural Trends
American women are very not concerned (14%) about overall fertility. Men are more concerned (30%) but not enough to do anything about it, and younger people (and more liberal people) are even less worried than that. Until that changes dramatically, we will never be able to try solutions capable of working. Samo Burja reports similar attitudes across genders in his anecdata, and points out the preferences cut across many proposed explanatory factors.
The culture is all too eager to tell us that children, or even marriage, will make us miserable, when it is not true even if you discount the long term. The latest example was this, where there was a widely distributed claim in a new book that said married women are miserable, because they report being unhappy when their spouse isn’t around… but that actually meant ‘spouse absent’ meaning no longer living with them, not ‘stepped out of the room.’ So married people are indeed happier, so long as they actually live together, which is highly recommended standard practice. Whoops.
The percentage of births to unwed mothers is very high, around 40%, but has peaked.
When the parents are together, they are staying together more often, as well.
We’re not that close to 1960-level numbers, but that’s a dramatic fall in the divorce rate. The decline in unwed births is smaller, but noticeable and looks steady.
Robin Hanson notes that many recent cultural trends among wealthy nations have primarily only happened in Western countries, moving away from mostly static Asia and Africa. Elsewhere, wealth did not predict the changes, but still did predict fertility drops. This matches previous observations that East Asia now has the worst of both worlds, where women and families have to deal with modern challenges, expectations and demands and also older ones, but women can also opt out entirely. So they do.
Kelsey Piper asks, why can’t we be normal about all this?
After Dobbs, vasectomies and tubal ligation procedures are up. Tubal ligations are up over 400 per 100k individuals with healthcare appointments on a monthly basis. That sounds like a lot, and is more than double from 2019. Vasectomies are over 100 per 100k appointments.
At least some corners of the internet are supportive, I guess?
How would you design a city so that more people would fall in love? In all seriousness, I would start with YIMBY. Build, baby, build, so housing costs are affordable. That gives people the opportunity. Next up better mass transit and fully walkable, not being able to see someone logistically is a huge barrier and everything gets more pleasant. Then yes, absolutely, you create a bunch of parks and benches and monuments and museums and so on, and go from there.
These are large effects.
I would double down on Robin Hanson’s warning to beware the philosophy degree.
If your philosophy degree greatly reduces your fertility, what use is your philosophy?
How far gone is our philosophy?
Rather far gone, given that this is how Politico’s Gaby Del Valle framed a conference about the idea that maybe children are good, actually:
Gaby, it seems, cannot imagine any reason one might think that children are good or that the country would be better off with more of them. They couldn’t mean what they say about demographic collapse and our dependence on growth. They couldn’t be genuine in their values. It must be a political takeover, or racism.
They are not the only ones. They are the only ones at the conference that markets to exactly that kind of people. I am not going to that sort of conference. But yes, the fact that this is the way they had to fill out the conference is a sign of the times.
A famous finding is that the high school ‘baby simulators,’ designed to each kids about the perils of teen pregnancy, actively backfired if what you care about is reducing teen pregnancies.
In general, if some choice is happening 1% of the time, and you want that to be even lower, what do you do? It stands to reason that drawing lots of attention to it, giving people ‘the facts’ and making people really think hard about it might not be your best plan. This is especially true if there is lots of existing misconception and hyperbole working in your favor. Many programs to get kids to not do things actively backfire.
A funny suggestion at the link was making people do this once a year. Presumably that would have very different impacts, in addition to being deeply silly. But yes, my presumption is that any form of drawing attention and thought to the question would increase fertility.
The Guardian being The Guardian, they really are the king of the terrible take.
Exactly. The whole point is to make it easier to have kids. By responding with ‘are they really model parents?’ you are exactly proving why this is so important. If we only let ‘model parents’ have kids there won’t be many kids. If parents think they have to act as model parents all the time, they will be miserable and often opt out or quit early. You do need to pass a minimum bar, but past that the important thing is to show up, stick around and have the kids at all.
Rob Henderson notes that fertility collapse is among poor women, whereas college educated women’s fertility is largely stable in America. He has a theory.
Denial
Yeah, no. This is a completely insane baseline estimate. I do not especially worry about the solvency of the trust fund under baseline assumptions because I also see the assumptions about AI and various other things. The worry is if this could be people’s excuse for not panicking. That would be a problem.
Urbanization
Razib Khan tells the story of declining fertility as the story of urbanization. No question this is a key part of the story. Fertility and urbanization have a national correlation of -0.48. Within nations, cities have always, going back to ancient times, been much lower fertility than rural areas, with cities usually below replacement. A lot of that was always due to poor health and plague, an effect that used to be far larger than today, but that is only one reason of many. And urbanization is way up.
The Marriage Penalty
Your periodic reminder that we tax marriage, which also means we tax fertility. We do it less than we used to, but we still very much do it.
When you tax something, you get less of it.
They offer a variety of proposals. At core this is a basic set of arithmetic problems. It is not difficult to adjust the numbers such that it is almost always beneficial or at least neutral to be married, especially when there are children involved.
Indeed, if we cannot do better, there is a very obvious solution. Raise base rates as needed to compensate, and then allow married couples to file as if they were unmarried, if they calculate that this is cleaper. End of penalty.
The Biological Clock
I do not know how much of the fertility drop is ‘women and also men do not appreciate that there is a biological clock and they only have so long to have kids.’
I do know the answer is ‘quite a lot.’ So you get things like this:
I remember health class. This is likely the most important one thing to include. Everyone needs to know what the timeline looks like.
If they choose to ignore it, that needs to be an informed choice.
Technology Advances
The ultimate IN MICE.
It does seem at least somewhat reasonable to say ‘either the technology to do this will exist 10 years from now when I need it, or we probably have much bigger problems.’ It still seems like a relatively cheap action to prepare in case that’s not true.
Big Families
What motivates educated women who have five or more kids? Catherine Ruth Pakaluk writes a book in which she asks fifty of them, Hanna’s Children. Mostly they were motivated by the belief that children are the best and most valuable thing. They knew that having lots of kids was difficult and expensive and terrifying and required sacrifice, and they did it anyway and made it work. The group was also largely religious.
The reviewer here noted that the marginal cost of additional children seems to decline. That has been my experience as well. There are dire warnings that two is more than twice as hard as one, or three will be so much harder than two. It definitely brings additional challenges, but my experience is that this is not so, there are decreasing marginal costs all around. They complement each other, and I think are clearly better off for having each other, and many of the costs in both money and time are fixed or scale highly sub-linearly.
Au Pairs
Au Pair programs are the definition of win-win.
A student gets a place to stay and a chance to study in America, and some walking around money. A family gets badly needed childcare. It is completely voluntary. The economic benefits are obvious. Everyone wins. The possibility of an Au Pair substantially enhances options, and thus fertility.
So, of course:
Raising required compensation would dramatically nosedive participation. These people really do not get how supply or demand curves work. Fertility would suffer.
The comments have several people talking about ‘slave labor.’ No one knows what words mean anymore. That includes both ‘slave’ and also ‘labor.’
Childcare Regulations
Yes, it can be this simple, says new paper.
Your childcare regulations must be really harmful if parents respond by having noticeably fewer children. That is as clear a message as you can get. Listen.
The potential changes are big. They estimate that if you shifted from the highest level of regulation (Connecticut) to the lowest (Louisiana), the total fertility rate (TFR) would rise from 1.51 to 1.7, or 13% (!). If every state moved to Louisiana’s level, we would see roughly 38% of that improvement, or a 5% rise in fertility.
I talked Claude through the calculations and I am choosing to skip several adjustments so someone should do it more formally, but when I estimate the actual all-in cost this imposes on parents, I get that each 1 point increase on the 0-10 point scale increases costs by between $6k (low estimate) and $13k (high estimate). At 2.74 points of average improvement per state, using a middle estimate of $9k per point, we get about $24,600 per child that needs such care, for a 5% fertility increase.
Yes, I am fully ignoring the positive benefits to children and parents, because I do not think that has substantial impact on either quality or fertility decisions.
Thus, we can approximate that paying parents $24,600 per child over five years would increase fertility by 5%.
Writing this check would be an expensive way to raise fertility, costing almost $500,000 per additional birth. Note that many existing programs try to write such checks anyway, or do it selectively, to offset their harms.
Also note that we above found that baby bonus payments get at least 40% of their effectiveness from boosting liquidity. By spreading payments out over five years, we lose that benefit. So if we instead were to write the $24,600 check, we should expect to get an 8.3% increase in fertility, and decrease the cost per birth to about $300,000.
Remember the calculation on car seats as contraception? There parents faced an up-front cost, and I calculated that this implied the marginal cost per additional birth from a flat child subsidy program would be about $270,000 (or ~$286k in 2024 dollars).
Two years later, we have an estimate from a different program, and we got $300k.
Those are stunningly similar numbers.
We can now be reasonably confident that this is roughly what such programs would cost if implemented at modest size.
For transfers at birth to new mothers in America, for every $300k we spend, we should expect to get roughly one additional birth.
We can also gain this result from anything that reduces effective costs to parents. Car seat requirements and child care regulations are two good places to start. There are many others.
This would likely to be much tougher in places without a sufficient fertility gap. America has the large advantage that women actively want more kids. All we have to do is enable that.
Another fun note from the same paper is that Lyman Stone notes basically no person-level control variables matter for the fertility gap between desired kids and realized kids.
The Numbers
They are not good.
The amount of variance here is bizarre. Shouldn’t these curves be smooth via the law of large numbers? What is moving them around so much? I am actually asking.
The Housing Theory of Everything
More Births argues that building vertically is inherently disastrous for fertility. The higher your building, the lower your birth rate.
I do not think this need to be true, also a lot of this is correlational or selection effects.
Mostly I think this is a confusion between size of the building and cost of the space.
South Korea’s high rises do not allocate the space you need for a family, especially when you lack easy access to outside space, and kids are not allowed to roam freely.
Meanwhile, the drops in fertility reflect places where zoning changes, not places where the physical buildings change in their impact.
You don’t see a change from 2-plexes to 4-plexes, because those are still in the same types of areas. Then the 5-19 group is again similar, representing the ability to build modest apartment buildings. And then a jump at 20+ or so, which start to only make sense where space is at a premium.
When you build a large apartment building, you lower the cost of housing everywhere, which is good for fertility in any given location. But the particular location is likely to be expensive, and thus locally have lower fertility, again partly via selection.
The only way out is through. If South Korea had twice as many high rises, allowing all units to be larger at lower prices, then the fertility penalty would stop.
New Yorkers leave to raise families partly to get green space and the illusion of safety (and some real safety, although that is mainly from people thinking you are acting unsafely and calling authorities).
But mostly they leave because the rent is high, and taxes are high, and the private schools are expensive.
If you doubled the amount of residential housing in New York City, what happens? That depends on how much prices drop versus the population increasing. You are going to get a J-shaped fertility impact curve overall, and it is unclear where we currently are on that curve.
My presumption is that at equilibrium, if we doubled NYC’s residential space, we would grow the population from something like 8 million to 14 million, with a substantial drop in rents and increase in average apartment size, and local fertility would rise substantially, as would fertility elsewhere.
Whether that dominated the compositional shift is unclear, if you did not use the gigantic wealth effect wisely. If you used the wealth effect in substantial part as a child subsidy? Now it is not even close.
I think a similar thing is happening in this study that population density predicts lower fertility. Yes, it predicts lower fertility, but that is largely due to predicting higher space costs. And also historically cities being unhealthier much more than they are now, and children being more valuable in the countryside, where again we want to drive that to zero.
Lyman Stone looks into that question using old NLSY 1979 cohort data. The data available is limited, but what we do have is quite useful.
That’s an 0.04 hit to fertility per year of living with parents per year. Whereas time in a metro area is positive.
Here’s what you get correcting for some stuff:
Controlling for marital status feels like it should reduce the impact of living with parents. Yet we see almost no change.
Either way, we should worry about reverse causation and correlation. Yes, there are socioeconomic and marital status controls here, but presumably people who live with their parents are often doing so exactly because they are not ready to raise a family.
This might be a story about resources, optionality and preference fulfilment then? And yes, the more children you want the more children you get in general, but only to a limited extent. I am guessing that preferences shift a lot, if the correlation is this low.
So the idea is that if you are doing well, able to live in a metro area or own house, then you are in much better position to bargain for and get what you want. You can stay on your own or choose a partner that matches your preferences. If you are not doing well, you might compromise on children in either direction.
Lyman Stone suggests the issue is not urbanization, it is small houses that do not lend themselves to starting families.
So yes, it would be great if every 24 year old could get their own two bedroom starter house or apartment in the places they want to live, even New York City or San Francisco. It would also be the first time in history.
I do know that supply reduces price. However.
You know what it would take to make housing that cheap?
The cube.
Otherwise, all you are doing is forcing those 24 year olds to get roommates because they are forced to buy 2-4 times ‘as much house’ as they actually need.
Roommates are presumably actively bad for family formation versus living alone. Whereas if you rent one bedroom apartments (or studios, or dorm rooms) to single people, they gain disposable income and opportunity, and can then move later when they are ready. Why in the world would you want to tie a 24-year-old down to a 2-bedroom apartment or starter house and mortgage they don’t need, making it hard to move?
Yes, you want to build so much housing that the prices crash, build baby build as much as possible, but it can only take you so far.
Causes
As usual, you can run massive correlational studies on fertility, but they have the usual issues with correlational studies no matter how large. Density hurts fertility, and hurts it more for lower incomes, but how much of that is causation seems nonobvious.
Other times, it is easier to identify.
I never fly first class (except once I got a random free upgrade). I don’t get it. Even without the kids, why wouldn’t you instead want a ‘first class’ house? But the broader point matters far more.
Lyman Stone analyzes how much fertility decline is tied to income and development.
Her first argument is that the traditional U-curve findings, that when income rises very high fertility increases again above replacement, are based on tiny portions of larger surveys, and are statistically unreliable. Even worse, they have timing issues, as income varies with age.
Perhaps one could say that income relative to expectations and social position predicts individual fertility? That seems like the actual mechanism. As you get higher income (perhaps excluding female labor income, because of the substitution problem) relative to the perceived financial cost of children, you get more children. The problem is that if rising income also raises perceived costs more, you go backwards.
Her core argument is that what we actually have is Simpson’s Paradox. That what’s going on is that compositional changes in income cohorts are creating a U-curve that isn’t a good way of understanding the situation:
Or this example of looking at Amish and Ultra-Orthodox Jews:
This points to the hypothesis that causality is in both directions. Children impact income, so you may not be measuring what you think. Although that suggests that very high income groups are even higher fertility than they look.
And yes, these graphs are quite interesting. The first is slanted by household structure (e.g. ‘do you live with your parents’?)
This is married women not living with their parents:
So what matters is not being 70%+ of the income share. These numbers only look at years when the woman is married, which is why TFR is in the 4 range this whole time, although I’m still confused why it pushed it up that high. Perhaps we really should be focusing on getting more marriages to form and last.
Next up she shows schooling is a linear predictor of low fertility at all income levels, listing four reasons:
I would add that school delays economic actility, and one’s ability to get into a life position where one can get ready to have a family, and we now strongly discourage family formation during one’s education.
The core argument Maxwell Tabarrok is making is that labor supply is now declining as a function of labor productivity. People value their leisure time and non-work activities, so they are satisficing on work and income. Which means that as wealth and productivity increase further, hours worked will decline and the opportunity cost of children will go down, and fertility will go up.
I do not think that is a good way to think about this, and the graphs he provides are unconvincing. Instead, I would go back to my notion above of anticipated cost (including opportunity cost) of children versus available surplus under culturally expected and legally required patterns of consumption.
So I would instead say: Beware the Iron Law of Wages.
The Iron Law of Wages
It is both, if you ask why they believe the crazy thing.
Lyman Stone offers this version via Maxwell Tabarrok, which highlights where the people are, and more importantly where they are not, which is at the upper right.
One can also steal from Robert Anton Wilson, and refer to the problem as The Revolution of Rising Expectations.
The Iron Law of Wages asserts that in the long run, wages tend towards the minimum necessary to sustain the life of the worker.
The logic is obvious in an otherwise static Malthusian context. Solve for the equilibrium, and there is only one answer. The population increases until the point where the marginal product is equal to that required for replacement rate fertility.
What happens when instead productivity is rapidly increasing, and we are growing wealthier?
Wages must rise, so they do not tend towards the minimum necessary to sustain life.
Instead, the minimum necessary to sustain life tends towards wages.
This happens through a combination of regulatory fiat requiring the purchase of more and higher quality goods, through various forms of artificial and real scarcity, increasingly expensive status competitions, and shifts in cultural expectation so that we consider more and higher quality goods necessary to sustain life.
Then consider what happens when culture, together with birth control, shifts to make it considered ‘sustaining life’ to sustain yourself without raising a family let alone a large one, and the requirement adjustments render children unable to work and expensive to raise.
For a time you get radically, horribly out of equilibrium. Expectations for living standards zoom past the Iron Law. People trying to meet those expectations are suddenly unable or barely able to raise families while staying consistent with cultural expectations and legal requirements, and many choose to opt out, can’t make it work or settle for only one child. Fertility falls well below replacement.
Then this risks becoming self-sustaining as it further shifts culture, and those trying to raise families must compete with those who give up on that. If adjustments are not made, the people die out, and their civilization falls.
South Korea
South Korea’s fertility nightmare seems best summed up as a symptom of being a nightmare in general?
Let’s not mince words. If 80% of your young people think of your country as ‘hell’ and 75% want to leave, then it matters little that South Korea is some economic miracle. The economic miracle exists so that the people may benefit. The people are not benefiting, to the point of choosing to cease to exist. Why is no one noticing this? Well, no one except everyone who makes South Korean media, which is both quite good and also constantly shouting this from the rooftops if you’re listening.
Things are so bad that dog strollers are outselling child strollers. They have technically declared an emergency, but they are not at all acting like they have an emergency.
Snowden Todd in addition to the usual suspects of education and sexism and geographic concentration proposes that part of the problem is too much small business, and the lack of large company jobs keeps people from settling down.
That’s the up front pitch. Instead Snowden paints a portrait of a country on a decades long quest to pursue GDP-style prosperity at any cost, with government and a handful of big corporations colluding throughout, wages suppressed and overtime the default.
Those big companies, the chaebols, are big and productive, but only combine for 14% of jobs at places with 250+ employees, versus 58% in America. Whereas the rest of the economy is not so productive.
The obvious first question is why aren’t you setting up shop in South Korea?
It seems like an amazing place to run a business. Everyone is highly educated. Everyone is disciplined and happy to work tons of overtime. You are competing for workers against horribly inefficient small businesses paying horrible wages.
If you are working for one of these small businesses, should you not found a new company instead? It doesn’t have to be a startup rocket ship.
The second question is why would you stay? What good is having a wealthy country if this is how you must live in it?
On the direct fertility question, yeah, the problem does seem overdetermined. You work long hours for low pay with little prospects, and if you have a child they get this elite education to suffer the same fate. Does not seem tempting.
Georgia (the Country)
Married births in Georgia spiked much higher in the late 2000s and mostly stayed high. What happened?
They had a great symbolic weapon to deploy. What else could serve this roll? Obviously ‘money’ but status can plausibly be a lot cheaper.
He then contrasts this with South Korea, where he says your status demands on where you work, which is based on intense early life zero sum competitions between students, hence all the super expensive private tutoring.
The obvious response to the situation in South Korea would be to opt out of it. Accept that your children might be low status in the eyes of others, but if you can pass on the willingness to accept this and keep going, you inherit the country. What use is high status with one or no grandchildren? Alas, this is not a popular way of thinking.
Here Johann Kurtz extends the argument that status is the thing that counts, and that the newly low status of stay at home moms is the thing we have to fight. This seems super doable if we decide that we care. The issue is that so far we don’t care enough.
Japan
A third of unmarried adults in Japan aged 20-49 have never dated.
China
That does not sound all that comforting if you don’t think AI changes everything. Yes, you have 25 years before you get to what is happening to Japan, but that is not so long, and from there things look to accelerate further.
Essentially China was fine until about 2017, then things declined rapidly and even more so with Covid. They are five years or so into the new very low fertility period. In terms of overall population numbers that will take a while to have its full impact, but it will compound rapidly.
As usual, notice the ‘and then a miracle occurs’ on the later part of the chart. Why should we expect things to stabilize after 2055? It is not impossible, but that seems like denial if you think it is the baseline scenario.
The weird part of such projections is that even those who face the music in the near term somehow think the music will stop.
By these projections, things kind of mostly stabilize. Why should we expect this?
Yes, no matter what happens with AI we know for sure that 2100 will look a lot different from 2024. It still seems sensible to project the baseline scenario properly, that is what properly motivates us to pursue the right changes, and we have no reason to presume that tech or cultural changes will tend to work in our favor here. So far cultural changes lowering birth rates have snowballed rather than balanced out.
Italy
Italy outright criminalizes surrogacy. Rarely do we see such extreme moral confusion, or such clear cases of civilizational suicide.
Northwestern Spain
The birthrate there is the lowest in Europe. Why?
More Births says this region has a high percentage of people living in apartment towers near the coast, and lots of young people living with their parents, and declining religiosity, and high youth unemployment (although lower than Southern Spain). Essentially the model is simple: Young people are failing to launch and get jobs and houses, so they are less likely to have kids.
Here is an interesting potential alternative explanation, although it still does not bode well for the region.
It seems like everywhere we can point to several of the usual low fertility suspects.
Russia
Russia considering banning ‘propaganda of childlessness,’ and there is discussion of raising taxes on childless families, along the lines of previous bans on other speech. I doubt this alone will have much impact.
Taiwan
Those policies seem fine, but not exceptional, and as others noted company benefits don’t seem to move fertility decisions much. Total compensation matters far more.
Claude estimated that if we account for demographics we should expect something like 0.45% of births to be to TSMC employees. My guess looking at the calculation is this is a modest underestimate, but only a modest one.
The United Kingdom
The fertility rate has dropped to 1.44. There are a lot of responses pointing to various causes that seem especially bad in the UK, especially their housing crisis, but this isn’t out of line with other similar countries.
Ancient Greece
The problem of low fertility is not new. Here is Polybius talking about it in Ancient Greece, blaming it for their fall. His culprit? Men becoming ‘perverted to a passion for show and money and the pleasures of idle life, and accordingly either not marrying at all, or, if they did marry, refusing to rear the children that were born, or at most one or two out of a great number, for the sake of leaving them well off or bringing them up in extravagant luxury.’
Sounds familiar.
Israel
Why is the Israeli birth rate so high, even outside religious communities? The hypothesis offered here is that those religious communities are integrated with more secular ones and seen as worthy of aspiration in at least some senses, so the memes and practices of very high fertility orthodox Jews filter down somewhat to other groups as well. And this is enough to keep the fertility rate at stable levels even among the secular, and has a much bigger effect among those in between.
It is a plausible theory. It suggests that the ‘right kind’ of cultural mingling, that allows us to assimilate ideas from isolated high fertility cultures without the high fertility cultures assimilating ours in exchange, could be part of a solution. It also points back to the status hypothesis, that essentially Israel offers the high-fertility subcultures sufficiently high status that it raises the status of high fertility everywhere.
More Dakka
Money is always the default. As always, if brute force doesn’t solve your problem, then you are not using enough.
There are also other incentives.
Tyler Cowen is not optimistic about getting good returns on the money, but says this and many such experiments are worth running. He worries, what if only 10% of babies were born because of this? That would indeed be a problem, since an 11% rise in births is insufficient.
The part where that means only 4.5 years of tax receipts, and thus a net loss, seems to miss the calculation. The payment is (likely progressive) redistribution, from some Koreans to others. We already do a lot of that without any fiscal payoff. The worry is that this would require marginal tax rates that were too high, and the deadweight loss would exceed the benefits.
My prediction is that I expect that if they did try $70k baby bonuses, as a lump sum payment, they would get a big impact. I also agree with Lyman that details matter. You 100% want to give this out as a lump sum so people feel it. And as this series has seen several times, South Korea has many other angles they could attack, if they were so inclined.
Another paper that shows child benefits don’t reduce labor supply.
Here is someone who is at least brainstorming about opportunity costs:
The problem is no one is forcing you to be a household. If you massively increase taxes on two-adult households, you get less households, especially what would have been two-income households.
So no, that will not work. If you want to drop the hammer via taxes, you have to tax childless households, or single person households. Or you can subsidize children heavily, which is the same thing, someone has to pay for that.
Bryan Caplan proposes a graduated income tax adjustment (+50%/0%/-20%/-40%/-60%/-80%/-100%) based on number of children, although he would prefer a tax holiday for some years. The advantage of lowering tax rates rather than lump sums is that you improve incentives, you avoid a budget line item and the people most tempted by lower tax rates are plausibly the right people to get to have more kids. Certainly a graduated schedule is better than Hungary’s ‘have four and never pay again’ plan.
I certainly know it would work, including that it would have worked on me personally.
I am almost certainly going to stick with three, but offer me no income tax for life and I assure you I’d have had four years ago.
The obvious issue is that this would get supremely expensive. Everyone earning millions a year would obviously find a way to have six kids, if necessary via surrogates or outright paying potential partners, and raised by those partners or often almost entirely by nannies. That is not exactly the goal, and you’d be paying way more than the market price to get it.
So you would want some cap on the effect, which could blunt how much it works. For those who have liquidity issues or short time preferences, which is most people, you are much more effective per dollar with the lump sum.
Perception
When deciding whether to have a child, it is the perceived costs matter.
One of the biggest perceived costs is the ‘motherhood penalty’ on earnings. Women are afraid they’ll be penalized in the workplace, and be at a permanent disadvantage. It certainly stands to reason that children would interfere with ability to earn money.
But what if that was far less true than people think?
This is a good test, since success with IVF should be a good randomizer. It also is not as crazy as it sounds. The conventional wisdom is that fatherhood increases earnings, because the incentive to step up and earn more outweighs other considerations. Ten years is a long time, but it is a far cry from thinking this lasts for 40, and the trend actively reverses later on.
So in this study, the women whose IVF was successful took a large earning hit in year one, but recover rapidly starting with year two, break even by year 10 and end up with 2% higher overall lifetime earnings.
Using IVF means the study included relatively older prospective mothers. Other data suggests that having children when younger carries a larger earnings penalty. Also this was in Denmark, which likely made things easier in various ways.
A key claim in the post is that, because this finding conflicts with the standard narratives and the stories people want to tell, no one wanted to listen, and it was hard to even get the study published. But that a literature review tells a different story than the conventional one:
I don’t have the time to dive into the literature. Certainly, if women end up earning as much or more in the end, that means (counting raising the children) that they are doing massively more overall work to do it. And the children still cost a lot of money. But we should do our best to avoid giving families and women the wrong idea about the magnitude of this penalty.
Your Own Quest
Bryan Caplan is asked by reader Matt Kuras how to look for a woman who will want lots of kids. Bryan hits some of the obvious suggestions. Be up front about what you want, try multiple dating platforms, indicate some flexibility. He suggests potentially looking overseas, especially since Matt already speaks Spanish.
But Byran cautions (in response to Matt’s request) that going to Utah only makes sense if you go full Mormon. Whereas half the comments are saying, yes, you find this woman in a church, obviously. Certainly that is the percentage play, and has massive benefits, but involves very high particular costs one might not want to accept, as would other religious options.
Another thing several people noted is that saying 3+ kids up front narrows the field a lot, whereas once you have one often you can go from there. Either way, you ultimately have little say in the matter, promises are not reliably kept and preferences change, as they should given how much more one learns. It should help to be clear on what you want, but making a hard commit to big numbers a dealbreaker is not a luxury atheists have these days unless they want to sacrifice a lot everywhere else.
Are you good at predicting your own fertility? A paper asks.
This makes sense. If you are single you are not properly discounting for various things that can go wrong, whereas if you are married you have ‘derisked’ in many ways.
It need not be this hard, but yes, if you become worth over $100 billion then the implied fertility rate is very high. And no, you don’t need to go that far. Once you hit ‘escape velocity’ of wealth and income, you can have as many kids as you like and the money mostly isn’t relevant anymore.
Wow would that be a hard sell, but yes, absolutely, that is how it should work. People without children, who make a lot of money, should face higher tax rates than they do now, whereas those with children should face lower rates. This faces the reality.
What is that reality?
Help Wanted
Oh.