Robin:

Welcome, Lyman Stone, to Minds Almost Meeting with myself, Robin Hanson, and my co-host, Agnes Callard.

Lyman:

It's great to be with you all. I'm glad to be able to come on and chat.

Robin:

So viewers, Lyman is an impressive graduate student still, although I guess you're older than the usual graduate student, who is a world leader in especially empirical fertility research. And I was especially interested in the topic about two years ago where I spent roughly eight months obsessed with that until that threw me into my next topic. So I am honored and especially interested to talk to you about the subject.

Lyman:

It's really good to be with you. And I don't know if I'm a world leader. That'd be very, it'd be nice if I was, but I think I'm perhaps a regional potentate of fertility studies, not world leading.

Robin:

In my part of the world, you stood out as the person most talking about it. And when you talk, you speak authoritatively and seem to know many things. And that's impressive and relevant. Well, so to me, what drew me into fertility was thinking that was a big problem, realizing that was a big problem, because I realized with falling population, innovation would fall. And that's just central to economists view of the world. And then as I dug into it, looking at causes, it seemed like most of the proximate causes were cultural. And then that's made me switch to thinking about how culture works and the causes of that. So I guess that's maybe, I guess a question I might want to start with is just, has the world ever seen a society that saw they had a fertility problem and then fixed it? Is that, is there any historical basis on which we could hope that we would see a problem and fix it because somebody else did it?

Lyman:

There are, yeah, there, yeah, we do. Um, probably the. The example that a lot of people wouldn't find compelling, but you might, is Meiji period Japan, right? So in the Edo period, the Tokugawa shogunate, Japan has a really, really overpoweringly antinatal set of state-promoted cultural norms, and also actual laws. Like, they forbid splitting of inheritances. They promote a new form of Buddhism that provides sort of moral ratification for abortion and infanticide. They suppress the form of Buddhism that promotes fertility, which is called Pure Land Buddhism. Um, uh, it's, it's a really, and we have edicts from this period where they're like, we need to cut the birth rate because we're going to run out of rice. Like, just like bureaucrats say that in like 16th century Japan, we have the right, like it's, it's all there. And we also, Japan has pretty good records. So we actually know fertility did fall. Japanese fertility in the Edo period was adjusting for child mortality. Much of the northeast part of Japan had below replacement rate fertility. Places further from Edo that were further from that region that were less exposed to this policy were still higher. But Japan on the whole has very little population growth in this period. But then in the Meiji period, so mid-19th century, they kind of said, crap, we want to become a large, powerful, modern state. And if you're going to do that, you need workers. And we don't got them because we killed them all as babies. So they launch a new police force whose job is to surveil pregnancies to make sure that they are not aborted and to ensure that newborn children are not killed. And then they also do some various marriage promotion policies. And they also have a state-led suppression of this innovative form of Buddhism that had justified antinatalism. And there's a very vigorous public debate in Japan in the mid-19th century about modernization generally, but fertility specifically is a considerable part of this. And fertility rates go through the roof in Japan. They go from like three or four to like six. And at the same time, educational attainment rises, life expectancy rises, industrial output rises. I mean, this is the story of modernization of Japan. But what everyone ignores with the story of Japan's modernization is that whereas a lot of places fertility has fallen during this process, in Japan it actually rose a lot. Because instead of Japan being coincidentally exposed to kind of the French disease, so to speak, the norms around contraception, small family size, and sort of liberated individualized sex during its economic industrialization, Japan instead is exposed to this state-led pronatal cultural campaign as a reaction against a prior state-led antinatal cultural program.

Robin:

So later on, much of the world tries to emulate the high status, rich West, and then they're emulating low fertility. But was Japan trying to emulate a particular other part of the world at that time that they, so when Japan industrialized, they clearly were seeing the rest of the world and saying, we want to be like them. So were they picking a part of the world, like maybe the U S they saw that has high fertility, we want to be like them.

Lyman:

I know they brought in experts from many different places of the world for various pieces of their modernization program. I know American experts were a considerable part of that. As far as fertility goes, I don't know, I do not recall reading that they had any international input on that particular issue. My understanding is that that one was a little more homegrown. Basically just that they knew they needed workers. They needed workers and soldiers. And they wanted to have this empire. Pretty early on, Meiji Japan was not going to be content with chilling on their home islands. So, I mean, you see pretty early on conversations about Korea, Taiwan, and Manchuria. They don't actually get those until the late 1800s, early 1900s. So they knew they were going to need workers and soldiers, and so they set about making them.

Robin:

Marc Thiessen So then the proximate cause, culturally, would be a lust for empire. Yeah, sure. And then they were inspired by other empires.

Lyman:

But also it's worth noting that the other proximate cause was, and it's no spoiler here that I'm writing a book on this topic, is that there's an explicit overt debate going on for a couple centuries in Japanese literate culture and non-literate culture about infanticide and abortion. So it's not like population management through fertility control was like a thing Japanese people had never heard of, right? It's just that during the Meiji period, instead of kind of the scarcity mindset dominating, the growth mindset dominates, right? Instead of saying, gosh, we're going to have this isolationist policy where we're all just going to stay on the home islands, never speak to the Dutch or the Portuguese and try not to run out of rice. Okay, or millet or whatever. Instead, they say, we're going to beat the Westerners at their own game. We're going to modernize, we're going to industrialize, we're going to have an empire.

Robin:

So is the baby boom after World War II and Europe and the US a similar, like, does, like, we need to be more bigger, more powerful push? Or is, was that a conscious choice? Or is that just a side effect of some other random thing?

Lyman:

There are a lot of different factors going into the baby boom. And the most proximate explanation of what's going on is that between 1930 and 1950, I mean, the housing stock is revolutionized around the world in that period. And it's not just that tons of houses are built. This is the period where like all the new houses are getting electricity, indoor plumbing, uh, air conditioning, uh, better heating. Like the quality of the housing stock rises just incredibly at the same time that price prices in relation to income or not. You get this incredible thing where a house in 1950 is, is bigger, is better quality and is cheaper relative to income than it was in 1930. Um, and so what's going on. And so you can actually see this, that almost the entire baby boom. can be compositionally explained by the increase in homeowning household headship among young people. That in 1930, 1940, there were very few young homeowning heads of household, whereas in 1950, 1960, there are tons of them. And...

Robin:

rising fertility everywhere then, because the world, this would be a real wide-term, like not just in Europe, in the US, involved in World War.

Lyman:

So the baby boom is variable by country. It occurs in a lot of places. And it does, housing data isn't as good as we might like it to be, but from the few countries I have been able to put together data on, it basically does proxy for the shift in homeowning headship, which is largely a shift in basically constructing housing. Places that built a lot of housing have bigger baby booms. And interestingly, this has a weird interaction with World War II itself that some of the belligerent countries have quite significant baby booms and some don't. And some non belligerents have baby booms and some don't. And there's a sense in which like your reaction to your cities getting bombed and like how much rebuilding you do versus how much you kind of like let things sit is actually part of what shapes your eventual baby boom.

Robin:

But hasn't housing costs just continued to go down in the last 70 years since World War II and, or even many years, I guess, worldwide?

Lyman:

No, if you look in 1970, okay, if you look at 20 to 35 year olds, like the key age for entrance into family, like we don't care what happens with 50 year olds, but if you look at 20 to 35 year olds, And I'm not just talking men, men and women alike. In 1970, the median house in the U.S. cost about five years of median income for a 20 to 35 year old. Today, it's almost nine years.

Robin:

That's in the US. What about India, Brazil, Nigeria, the whole world? Surely hasn't housing been getting, you know, cheaper compared to income across the world in the last 80 years?

Lyman:

So I don't know everywhere in virtually the entire Western world, it's been getting more expensive. Um, in the developing world, you have a different dynamic. So, and, and I think this is actually really important to get right out of the gate, I think, and you might disagree with this, but. There really isn't a monocausal explanation. It's tempting to think that because fertility is declining everywhere, there's one thing going on everywhere. But the pace and timing of decline are actually quite variable around the world. And they point to multiple factors. So I mean, just declining child mortality is obviously a huge factor here. If more kids live, you don't have as many kids because you don't need to have as many kids to hit your desired family size. Point blank, there's that.

Robin:

If we see this long aggregate trend across the world, do we really want to say that that's a coincidence, that there just happens to be a bunch of local trends that each happen to produce lower fertility and there's no overall reason why they all happen to be in the same direction? We're just living in a coincidence where it looks like there's a big overall trend, but there's not. It's just a bunch of specific local trends.

Lyman:

I don't think it's a coincidence, but I don't think it's a single cause. Right. So, I mean, I think maybe we were discussing this on Twitter and maybe, maybe I use this analogy then, but like, it's not a coincidence that when a pebble lands on a bigger pile of pebbles that it causes an avalanche. Okay. At the same time, that pebble alone didn't do very much. Right. So the chain of cause and effect can be multi-causal without it being coincidental that these things interact, right? So, for example, falling child mortality explains a huge share of declining fertility, okay? And the spread of developmental idealism explains a very large share as well. That's basically kind of this aspirational norm oriented towards the West that sees small family size as part of the project of modernity and self-improvement. But And these two things are separate. And we do see places where they move separately, where child mortality falls, preferences don't, and fertility responds just to child mortality. Or we see places where preferences change, and child mortality doesn't, and fertility responds to just preferences. We also see places where they both change, where neither changes. But the point is they both actually matter. And we can see this in good empirical work.

Robin:

Those are both global trends, right?

Lyman:

Well, no, there are particular trends because in some places child mortality doesn't fall, right? Now, but they do have a global relationship to one another, right? There are underlying technological, cultural, and political factors that influence both of them, though not in the same ways. But that doesn't mean that there's like one cause driving all of this, right? Because at any point, you know, somebody can stick their hand out and stop the next pebble from falling. Right? Or at any point, random chance could have led to things going a slightly different way. And we can see that in a place like Israel, where Israel is just as modern and developed as other places. But by a couple of random chances, things went a little bit different.

Robin:

Israel is quite exceptional. Sorry. Yeah.

Agnes:

It seems to me that already there are like two very different kinds of causes that you have brought out. So one is like the kind of cause that was operative in the Meiji period in Japan, which is a big vision, right? Like a big ambition. that then drives the demand for more people, where that seems really different from, like, lower cost of housing, to me. Those seem like very different kinds of causes. Yeah. And, like, I can very well imagine that the second, the others that are similar to lower cost of housing, you know, decrease of travel tariffs is of that same kind. And... That is, the second kind is not intentionally oriented towards, it's kind of accidental, whether you're going to get fewer or more children. Whereas the first kind, the greater ambition, there seems to be a bigger connection. I wonder whether There are more causes of that first kind. Either way.

Lyman:

Either... Like big vision type causes?

Agnes:

Yeah, or like, you know, like you might say, like, if a country gets into a state of, let's say, The developmental ideal isn't intentional.

Robin:

cause. It's the other direction.

Agnes:

I thought of that as a possibly, possibly. I actually think it's not obvious which one that goes in.

Lyman:

Yeah. And I think that one is interesting because in some places it is intentional and in some places it's probably not. So I get the elite level developmental idealism is probably intentionally advanced as a way to reduce fertility. But at the popular level, it's probably adopted with the oblique sort of...that is, people don't...at the mass level, people don't adopt developmental idealism because they want to lower fertility. They lower fertility because they believe developmental idealism. But at the elite level, it can be flipped. Um, so they're actually, I think this big, the question of like, are there big macro visions that are there other kinds of big macro visions that matter? Um, uh, and the answer is like, yes, obviously. So, uh, like ideas of transcendence. So religion is going to be like your most common one of those, but, um, it is. It is always awkward to bring it up, but the Nazi pronatal policies that were for Aryans worked. Fertility rose very sharply for Aryans until they started getting blown up by American bombs. So, big expansive visions. of your society growing and being mighty and powerful might be more, they might be, they might have effects. You can also, is it kosher to share screen in this? Can I share a graph?

Agnes:

You can only some, you should describe it as well because.

Lyman:

Okay, yeah. So there's also a group in Finland. They're called the Lestadians. They are a pietistic sect of Lutheranism. They're a conservative branch of Lutheranism, but they are different from other conservative religious high fertility sects. If you think of like the Amish, they are, you know, Luddite, separatistic, politically quiescent for the most part. The Amish are not going to take over society for the simple reason that their model of the world doesn't scale. They literally depend on non-Amish people to pave their roads, to deliver their babies at the hospital, because they use hospital births, these types of things. their model of the world doesn't scale. They cannot actually maintain their societies without non-Amish people. So if Amish people start to become a huge share of population, exit from Amish communities just scales with it. The Hasidim, ultra-Orthodox Jews, probably doesn't scale without major change because, again, they're really dependent on other people operating a lot of their society. The Leshtadians are interested because they're not like that. They have normal education levels. They're not separatistic. They live in the same towns, neighborhoods, streets. They're not technologically Luddite. They are just They just think babies are great. And they have some other theological differences. But they are, to all appearances, just normal Finns. They live in Finland. Actually, there's a sect of them in Wisconsin, too, but they're quite small. Or in Minnesota, maybe, something like that. And there's actually several legislators in the Finnish parliament right now who are Lestadians. In fact, I think one of them got arrested recently. There was a whole free speech case involving a conservative Lutheran who quoted a Bible verse on Twitter and got arrested for it in Finland because it was hate speech. Turns out the Bible's hate speech. But they won their case. They won. But regardless, so this group, the Lushtadians, are interesting because they have really high fertility, even though they are not, like, low education or separatistic. So I can share the screen now.

Robin:

I've made you a co-host, so you're able to.

Lyman:

But I can't, why can't I see it?

Robin:

At the bottom, in the middle of the work, it remains share button.

Lyman:

Okay. So let me, so I can share the screen. And basically this is a graph with two lines. There's a red one for a town called Larsmo, which is a town that's like 75% Lushtadians. It's like their big hub. And then the blue line is Finland. And the red line from 1942 to the present, to 2024, is basically just flat at 3.5 kids on average. Okay? There's like no change. The blue line for Finland, you see the baby boom. Then you see like steadiness around 1.6, 1.7 from like 1970 until 2008, 2010. And then you see a decline to like 1.2 today. In fact, Lushtadian fertility has actually gone up in the last 10 years. And so, like, this is interesting because this, to me, even more than, like, the Amish. Because with the Amish, you can say, well, is it their unique cultural values or is it that they're, like, you know, separatistic, uneducated farmers or something like that? But in the Lushtadians, you can really see that, like, no, it's just their sense of their vision of their life. And so, yeah, I think there are probably other visions of life laying around that could motivate fertility. You can think of Mormons as another exception case. You can think of Roma as an exception case, right? Roma in Europe have relatively high fertility and they have a rather different vision of the good life. I could trot out a few other examples. But the point is, there are probably other big visions laying around, and there's probably others that could be innovated and promoted, but we don't always know what they are.

Agnes:

Can I ask a question about the Lushtadians? Yeah. What's their retention rate?

Lyman:

Offhand, I don't know it exactly, but I know that it's not low. Like, it's not unusually low. In fact, I think I know someone who has that data, but my understanding is that it is not low.

Robin:

Like two thirds maybe, or 90. Maybe something like that.

Lyman:

It's not as high as the Amish. The Amish are like 80, 85%.

Robin:

Amish are like 93.

Lyman:

No, no, the Amish are not.

Robin:

That was the number I saw.

Lyman:

Yeah, no, the Amish are like 85, but it also kind of depends on how you count the Amish, but yeah. But the Amish like to, there is a narrative that their retention is over, is like 96% or something, but no, it's like 85 or 90.

Robin:

So we should be able to look at the rate at which these different visions show up in the world to predict the rate at which we expect them to show up in the future. So we might say groups like the Amish or Heretim or the Lestadians, I'm sorry, I mispronounced it, haven't just shown up very often. These are rare things to appear in the world. And if their rareness continues at the same rate, you might say, well, we should mostly expect the existing ones to continue their trajectory and then new ones to show up rarely. And then that gives us a basis for predicting a future, which is to say, you know, main world population will peak and decline, and then these other ones will rise. And then unless our mainline civilization finds a way to be much more lusting after empire or some other inspiration that will turn us to, we must take Greenland. Yes. Well, that's the sort of thing we could do it. But that's why I might ask, what are the historical precedents for societies changing their attitudes? That is, could we get elites like Meiji Japan to all get together and say, we want empire here? Is that at all plausible for our, you know, Europe and US to, or even China to do in the next century? Or is it mostly going to be are being replaced by these small fertile subcultures. I wanted to mention, I think the Amish and even the Heretim are capable of change. So like Christians basically took over by growing over three centuries and they made a lot of changes in that time in order to do that. But so change is quite possible in these groups.

Lyman:

So the Christian conquest of Europe is kind of an interesting example that is actually, it was the, another example I was going to give, cause I have some cool graphs on it. So, um, we actually know the religious composition of Egypt going back about 2000 years because we have so many papyrological records and we know that like your given name tells us a lot about your religion. Um, like we, we know that like if, if your name is like Amun-Ra, low odds your parents are Christians. And likewise, if your name is Muhammad, pretty decent odds your parents are Muslim. And from that, and we also actually have tax records from the Islamic period, since non-Muslims pay different tax rates. And so we actually have records, and again, I'm sharing a screen from a recent post I did. This is the historic religion in Egypt over time. And what you can see is the Christian share doesn't rise slowly over 300 years. It rises like between like 265 and, uh, and, and like 400, it's like 150 years. It goes from like 4% to like 75% of Egypt. Um, and likewise, the Muslim share just rockets upward after the Muslim conquest, because most people will just adopt whatever they see as the winning horse. Um, once it looks like a group is getting ahead, most people will bandwagon. Cultural culture is a bandwagon dynamic. Um, when you look like you're winning, everyone wants to be, when you've got the vibes on your side, nothing can stop you. Um, but the other interesting thing is that, uh, the Roman Empire had a fertility turnaround. So I have another graph that merges a bunch of different sources of evidence on Roman fertility and if I can find it. that basically we can see, this is a second graph that looks at a bunch of, it harmonizes a bunch of data for fertility in the Western Roman Empire and the Eastern Roman Empire. In the East, it's undersampled. It's mostly going to be Egyptians and Latin speakers in the East. I definitely have a gap for Greek speakers in the Roman East, so, which is most of the Roman East. So, this is going to be mostly Egypt and the Balkans, basically. But what we see is that in the first century, the West probably had higher fertility than the East. And there's reasons to think that in the first century B.C., or in the last century B.C., so to speak, this gap was even bigger, probably. That fertility in the East was probably around four children per woman in the last century B.C., and it was probably around five and a half in, like, Italy. Maybe six. But then you can see by the third century, that in the Roman West, the average woman's probably having fewer than four children. Whereas in the Roman East, fertility has risen to about five and a half. And in the Roman West, by the fourth century and later, has higher fertility. And what's going on here is basically, I mean, it's a variety of things, but it's largely that the Greek East noticed that low fertility was causing them problems, and also the first. the dynamic here. So it was a combination of, in the latter BC period and into the early period, you start to get more Greek writers complaining about low fertility and being like, we should do something about this. And then you start to see them convert into Christianity. And Christians just have more babies. So, and they don't kill their babies. And so by the fourth century, you, and then as Christianity spreads West, you see a fertility revival there. And you can also see this at the level of emperors. I have data on fertility by religion of emperor. And what you can see is that imperial pagan, across the entire history of Rome, pagan imperial emperors, their wives average like 3.6 kids each, but the wives of Christian emperors average like five kids each. There's a huge Delta between them. Um, so like this is a case of a society that had low fertility. They were aware of it. You can find so many elite Romans and Greeks complaining about low fertility. And then a high fertility ideology came along and won.

Robin:

Right. But that's not so much that the elites said, Oh, we need to become Christians so that we can be more fertile. Right. That's correct. Just being displaced by another culture. That's right.

Lyman:

So what did happen is that the elites came along and said, our system is failing. Actually, there's a great book about this called the final pagan generation. I highly recommend, um, that they basically said our system is failing. Like being an elite Roman man in the second and third century sucked. Like, it just, for a variety of reasons, this was actually not a great time to be an elite Roman man. And so people were looking, particularly elite men, were looking for, like, what's a better way to do this? Elite men and also some urban low-class men and women. And Christianity was a strong candidate. There were other candidate ideologies. Christianity beat them in no small part because of its fertility culture. Rodney Stark has a good book about this, about the rise of Christianity and how fertility was probably a big part of it. But the point is, actually, elite Romans, it's not that they said the problem in our culture is low fertility per se. It's that they said, clearly our culture is dysfunctional on multiple levels because they could perceive that the economy had been not as great as it was during the great early imperial period and all these things. And they went looking for ideologies that appeared to be good candidates as successors. And ultimately Christianity wins that battle.

Robin:

When we see your face again, take away that.

Lyman:

Oh yeah.

Robin:

Yeah. Um, Well, I mean, almost always when there's conversion, people give reasons or excuses for their conversion, uh, that, you know, the question is whether the conversion is caused by the reasons they give later or whether they were jumping on a bandwagon, as you described. Right. So when we look forward to our future, the question is, you know, if nothing else, the Amish, Heretim, et cetera, will just rise in population and then if they make the proper adaptations basically displace us. And at the point they displace us, people will be complaining about how they're doing better and giving that as the reason for switching. But that's different than a scenario where the rest of us do more like what Japan did and said, oh, we have a problem. Let's deal with it.

Lyman:

I think, though, there's kind of three scenarios here. One is the Amish replace us. And that is not going to happen. Even among the Amish, I have a paper coming out on this in the next few weeks. Among the Amish, we already observe a fertility transition happening. Their fertility is falling. And it's falling particularly among the Amish sects that have started to try to make compromises, which are the fastest growing sects. Um, so the Amish sects that has started to let people use a phone under certain circumstances through these things, they're having a fertility transition. They can't keep it up. Um, because separatism is strict. Separatism is a two-edged sword. Your community, like. You can maintain your norms, but also you just have a scale limit. So the Amish aren't going to take over. The other two options are basically conversion. That is, we all convert to Christianity, so to speak, or whatever the next ideology is. or adaptation. That is, we find a way to modify our current hegemonic ideology. And I think the case of Christianity, you know, the Meiji period points to a modification, although they did have to basically switch the version of Buddhism that they adhered to. Whereas the Roman case, the late classical case, points to conversion. And I think that's the interesting question, is are we a society on the cusp of a major conversion? Or are we a society with the capacity for a major adaptation?

Agnes:

Can I can I follow up on this? So I think that's really interesting because like I've heard Robin push this like they're going to replace us line a lot. And I'm sort of persuaded by what you're saying, which is that's not really how it works. That is what happens is like. We jump into that, like how people converted to Christianity, right? We will become those people in one way or another. And then the question is, OK, are there two different forms of jumping? That is, one kind is conversion to a different religion, and then the other kind you're describing as like adaptation Where, as you just noted, it seemed to include conversion. That is, there was this change in the form of Buddhism. And like, I actually wonder whether we should just say, well, look, the thing that's going to happen is that our ideas are going to change. and our ideology is going to change. And in fact, maybe we can just predict it's going to change to a more pronatal ideology in some way. But it's really, really hard to predict in exactly what way will our ideology become more pronatal. And so we should be reassured. Everything's gonna be fine.

Lyman:

So I think we shouldn't be reassured, though, because what we know about heritability of fertility is that it's not. There is no long-run heritability in fertility. At the two-generation level, it's literally zero. That is, grandparents have almost no predictive power on grandchild fertility. Even among people who remain within the same cultural group, correlations are low across multiple generations. At the genetic level, we know that the genes that predict higher fertility vary extensively across similar industrialized high education societies, and they vary extensively across proximate cohorts, that is, across short periods of time. And while there are cases of sort of cultural groups with higher fertility out-competing other groups. There are. In fact, I just published an article literally today where I argued that this is basically why monogamy is dominant and why monogamy will always be dominant. over polygyny. But there are cases of this. There are almost all kind of exceptions that prove the rule. So, like, I think I would say very confidently that we are not going to have a future where, like, 60% of people identify as LGBTQ. Okay? Because there are, like, very clear reproductive fitness reasons why that will not happen. Um, but a situation in which straight people in the future all only have one kid, totally plausible. Um, because there's.

Robin:

So it seems to me like this is about timescale. That is, uh, you can see events that might happen in the future at a certain rate, one of which is a new ideology spreads among us. Another is, you know, some subgroups just maybe parts of them at least continue to have high fertility and successfully grow. The question is the rate at which these scenarios happen. That is, are we talking a 1 or 2 or 10 century decline before a revival again? So that's the key question. How often do these things happen?

Lyman:

Yeah. On the genetic timescale, I'll say for... You're right. Not even including a genetic fix, a cultural fix would be the main... For the most powerful genes that we know of, for them to reach fixation in the population will take 3,000 to 7,000 years.

Robin:

current rates. So I think the cultural rates are definitely higher than that, but right. So some people, someone who was optimistic thinks, oh, in another 50 years, of course. we will see these changes. But now I think we should just look at past rates of these cultural changes as our best indication of future rates. How often have major ideological religions or movements shown up and gotten a big fraction of the world, or how often does... So again, I think, say, the Amish are decentralized like the Haredim, so even if large fractions misstep the other parts can continue. So it'll take longer, but they do have an internal selection effect whereby whichever parts of them can maintain their high fertility can continue with that. In contrast to the Mormons who had high fertility but then had a very centralized control to enforce integration into larger society. But again, these are these two paths you outline. One is some outside groups, like maybe the Finns you know, just continues to grow and displace. Another is that some new movement spreads among us, not directly by fertility, but both of those rates seem to me pretty low. So, and the rate at which something like Japan happened, where the whole society gets together and decides we want to do, we need to do this thing, so let's make a whole new plan. That's also quite low. And that's, you know, Japan is unusually the sort of place that's done that over the, over the, you know, world. You know, the key question is how long of a decline to expect. I don't think human extinction is on the table, but a long decline might well be on the table.

Lyman:

So the interesting question here would be if you had a database where you could say, like, in a given 50-year period, what are the odds that a country's position in the global distribution of fertility will rise by more than two standard deviations between two 50-year periods? That would be a very interesting thing. We do not have very many, the UN's database goes back a total of 70 years. If you are optimistic about historic data reconstructions, you could go back to 1700 using like big crowdsourced genealogical data. If as historic and ancient DNA reconstructions get better, We are starting to be able to say a little bit about reproductive demography in deep history. We're able to say more, but we are still not quite at the place where we can truly walk from phylogenetic trees towards family sizes. We can only talk about sort of all we can really capture is kind of like collagen-y stuff and cousin marriage. But eventually we may get to the point where our understanding of the human, human genetic tree is enough that we can talk about family size in the human, in the deep human genetic past. Um, but we're not there yet. Um, so this is definitely going to be an interesting thing to look at. Uh, but as of yet, I don't think we have a way to test the theory, um, that like, you know, we don't have a way of saying, like, what is the frequency with which a country that was, you know, appreciably below the mean of the distribution moves to appreciably above? And not because other countries fell, but, like, because they rose or something. Like, we don't have that data set right now. We really only have one major cultural episode, and that is the post-war period that we have widely accepted data on. At the same time, I will say, Culture can change in the blink of an eye. I mean, the data I showed on Egypt right there, I mean, Egypt went from zero Muslims to like 50% Muslims in about 10 years. Now, that involved a military conquest. But there are non-military cases of this. I have a paper in review, actually not in review, excepted. It's going to be published at the Journal of Population Economics where we show that In the post-Soviet country of Georgia, their fertility rose from 1.6 to 2.2 in 18 months, and it remained above 2 for like 12 years because of a specific cultural intervention by a popular religious leader. Now it's below replacement today. After COVID it fell, like it's still kind of declining, but it's now higher than any of the Muslim countries in the area. So that's interesting. Armenia also is now higher than Azerbaijan. So there's interesting trivia for you.

Robin:

I feel like one key relevant datum there here is how in fact people are framing fertility issues in light of our current sort of world monoculture. That is, if it was framed, the more it's framed as an incidental side parameter that people don't actually care that much about, then the more prospect there is for it to make big changes. But the more centrally it's tied to central features of our shared ideology, then, you know, if it's framed negatively, then that's more of an obstacle to changing And it seems to me as we are seeing more discussion of fertility, it is getting a substantial left-right framing. And what I see as a predominant framing is many people say, fearing that, you know, higher fertility is going to come directly at the expense of gender equality. and that it's all, you know, an attempt to undermine women and to promote white race or things like that. And that framing doesn't bode well for a ideological revolution because that's pretty core to modern ideology. So a key question is whether it can be disentangled from that framing on what timescale.

Agnes:

But if you take climate change, right? That's as politicized as anything. And people very much feel like, oh, caring about climate change is... At one point I read that it was the most clear indication of politically, if you're left or right, is how you feel about climate change. But that doesn't mean that we haven't done anything about climate change. Plenty of people think we haven't done enough, but it seems to be an issue. But we're doing quite a bit. Right. We do quite a bit. And so it's not obvious to me that the fact that an issue is politicized means we don't do anything about it. Well, it's like doing nothing.

Robin:

But for fertility, there's this objective metric of enough, i.e., if you don't get above replacement, you still have decline. Same with global warming. There's an objective metric of enough. And objectively, we just aren't doing enough by the objective measure of enough from global warming. It's still rising.

Agnes:

It's not clear that we would be doing more if it weren't politicized. It's not clear to me that politicization is the problem. I want to just ask Lyman a question about, like, you know, as somebody who's looking at the data, forget about politicization, just looking at the data. Like, the thing you just said about, you know, like, our best data comes from post-World War II, and that's relatively, like, you know, limited timescale. It's like, do we have reason? How worried do we have reason to be, is what I want to know.

Lyman:

You mean, like, what do you mean by that?

Agnes:

Well, I mean, um, like, if it's just pretty unclear... how these things write themselves and what kind of timescales that happens in. If we go back to Malthus or something, who I'm a fan of, I read Malthus, his reasoning is really good for explaining why we have a terrible cycle and the only way to get out of it is to reduce population.

Lyman:

He was also mostly right, actually.

Agnes:

Right. Right. I think that that's right. But, you know, there's just important, there are like important things he didn't predict. Right. And and those things turn out to be crucial. Um, and so, so, like, like, that puts me in, like, a kind of skeptical mindset, you know? Like, maybe this is a really big problem, maybe it's not a really big problem, maybe we don't have enough information to know whether it's a really big problem or not. And so I kind of want you to tell me, am I just being, like, um, one of those, like, overly skeptical people who is trying to comfort herself by being like, we don't really know? Or is that actually a reasonable response to the situation?

Lyman:

Um... I would say that we have lots of reasons to be pretty concerned about fertility being quite low. There are many reasons to think it will cause bad effects for the pace of innovation, economic growth, unless super AIs can just do all the innovation and growth for us, in which case maybe who cares. Um, though I'm skeptical of that. I'm actually, I actually agree with Tyler Cowen did a nice interview with Dwarkesh Patel a little while back.

Agnes:

I'm going to interrupt you because for me, did, did, did Lyman cut out for you, Robin?

Lyman:

No.

Agnes:

Oh, okay.

Lyman:

I think you cut out actually.

Agnes:

Oh, I'm sorry. My fault. But I missed, I missed what you said. So, so you, you were starting to say, um, there are many reasons why we have, we can be worried because of the economic growth and then can you just

Lyman:

Um, yeah, so like, uh, if we get less, if, if population, if each generation is smaller than the last and thus has, um, fewer potential innovators, fewer economies of scale, um, and, uh, and frankly, more care and upkeep obligations, both for the older generation and also for the institutions and infrastructure that they've built. And I would argue that like cultural norms and institutes and institutions have upkeep costs, right? Like, Once you make a law saying such and such must be done, somebody has to do it even if it's stupid to do it, right? Like the Ancien Regime has a maintenance cost. So there are certain things that are just going to eat a lot of human potential as population declines. We can expect innovation to decline. I wrote a whole paper about the linkage between population and like innovation and entrepreneurship. The short of it is it's bad. That's all quite real and problematic and concerning. Tyler Cowen and Dhorkesh Patel had a nice conversation a few weeks or months back where Tyler was actually arguing that even if we do get super powerful AI, that it's probably not going to cause bajillion percent economic growth. it's probably going to be like another 0.1 or 0.2% growth per year, which over a century compounds to quite a lot of extra growth. But it's not like suddenly we don't need human labor. And I think the most interesting point in this is that if you look at the actual budgets of AI companies, they still spend more on human labor than they do on compute. And we're like, oh, these massive data centers. And I'm like, they're still paying more in salaries. At least this is data that I saw. Maybe that was just on the training costs or something. I don't know.

Robin:

But it's like you've- Marc Thiessen I guess I was more worried about, can we predict the continued decline of fertility rather than the consequences of fertility? Is it just going to reverse and do we have to worry about it continuing?

Lyman:

So I just totally was just off on my own spiel there.

Agnes:

Yeah. Is it going to reverse?

Lyman:

We have lots of examples of fertility trajectories reversing. they usually come with things we can recognize as cultural changes. I talked about Meiji Japan. I talked about the Christianization of the Roman Empire. I'm right now sort of combing through some old crowd-sourced genealogical data to see if there's anything interesting there. It looks like there may have been a 17th or an 18th century fertility increase in like Austria-Hungary that I'm not totally sure what was going on. But it is kind of modest. It's not clear if it appears in every data or in all the different data sources. We can definitely see some interesting stuff happening in like the post-Soviet countries, particularly the Central Asian ones, with like re-Islamization, so to speak. But we don't have a huge data set. So like we don't we cannot say, oh, we've got all these examples of fertility just like naturally moving back towards two. And I will say, the UN actually assumes this will happen. They have a statistical model that they use for forecasting that The long and short of it is, it's not a crazy model, but it really, really ends up assuming that the experience of the post-Soviet countries in the 2000s is like a universal law of gravity. That is that what goes down must come up. Because during the 1990s, post-Soviet fertility crashed, and then it came back up in the 2000s. And their model learns. It's like a Bayesian inference type model. And so it learns from that experience. And then it just forecasts that what goes down must come up. But the problem is all the places that came back up, they came back up if and when they implemented pronatal policies. So if you actually look at all these different post-Soviet countries, their recoveries are all at different times and they're on different scales. And the timing is spaced out by 10 or 20 years. This is not one or two year differences. And when you actually start trying to use Google Translate on Kazakh and on Uzbek and on all these different things, you realize, oh, the fertility recovers when the government reinstituted a bunch of pronatal policies that used to exist during the Soviet period. Like, that's when they were covered. So it wasn't just that it just came back up, even in that case. It was that governments did stuff.

Robin:

And didn't it go above replacement, or are these just?

Lyman:

Yeah, seeing quite a few of them. I mean, in Georgia it did. In some of the Central Asians it did. I don't think Mongolia ever actually went below replacement. But some of them added, most of them added anywhere from 0.2 to 0.8 kids to their TFR.

Agnes:

So is there an explanation for why, like, Hungary's recent pronatal policies haven't been more successful? Or am I wrong in thinking that they have been less successful than, say, what you're describing?

Lyman:

I'm going to have to share another screen. I love graphs. I'm a graph addict. So one of the problems in fertility is that we have lots of measures, and there's debate about which ones are right. In principle, what we want to know is completed fertility. That is, people's actual ultimate family size. Not when they had children, but how many did they have. Usually that's what, there are reasons to care about when they had children as well, but in general, completed family size is the main thing we're focused on. But estimating changes in that from annual fertility data is quite difficult. So the best indicator that we have, or the indicator we think is best at doing that, is called Tempo and Parity Adjusted Total Fertility Rate, nicknamed TFRP star. And we have that data for Hungary through 2020, so we don't have it for 2021 to 2024 when things maybe have taken a little bit of a negative turn. But what we can say is that for that indicator, Turkey, there were only two countries in Europe that we have the data for where that indicator, sorry, there were three countries in Europe for that indicator where their fertility, where their TFRP star value actually rose after 2012. So after Hungary started implementing these policies is where I cut it. And those countries are Hungary, where the TFRP star rose, Czechia, which also implemented a ton of really generous pronatal policies at the exact same time, Um, but that didn't get as much press and Portugal and for Portugal, I must say I have no explanation. I have no idea what happened in Portugal. I've tried to look, I don't understand it. Um, so beats me on that one. Um, everywhere else, TFRP star fell. And now I can show you a graph of what TFRP star actually looked like over time for Hungary. Um, So around here 2012 or 2013 is when Hungary implemented these policies. And you can see TFRP star fell, it's on the right axis on this graph, fell from about 1.95 kids to 1.45 by 2011. And then it rose to 1.55 by 2020, and the rise is after these policies were implemented. That's not a huge increase, but it's an increase. And almost nowhere else had an increase. And at the same time, what happened is there was a massive increase in marriage. And it's because actually most of what Hungary did was not subsidize fertility. They canceled programs that were already subsidizing fertility, picked up the money, and moved it to other programs to subsidize fertility. But the difference is the old programs subsidized fertility generally, and especially for poorer women, and the new programs specifically subsidized marital fertility, and thus marriage. So you see this huge shift into marriage and a modest shift in fertility.

Robin:

I worry that, you know, just looking at the whole data set, you know, there's 100 or so countries over, you know, many decades or even centuries. And overall, there's this huge decline. And then you see, okay, every once in a while, okay, this went up for a while. Isn't that interesting? Maybe they'd add some policies and then it didn't last that long in a reverse. And if I look at the overall trend, I say, well, look. For the world, it's not enough to just every once in a while have a country that happens to do something that raises a bit for a while and then declines again. This just looks bad for the whole world. The question is, what's the chance that some place will not only do something that raises a bit for a little while, but raises it a lot and keeps it up for a long time? Otherwise, we're just all going to go and decline.

Lyman:

Yeah, it seems like low odds to me that we're going to have a global cultural revolution that boosts fertility. However, I don't think it needs to be global. You need... It needs to last, right?

Robin:

If it's local, it can't just be for two decades or something.

Lyman:

You do need something durable. I also think, again, the decline is dramatically overstated by a bias in the data to use births instead of surviving children. If you look at surviving children, that is children who made it to puberty, Um, there is not nearly as monotonic of a decline. There is some, um, but it is a lot more complicated. Uh, and in particular, the scale of decline is much smaller. Um, so I, I do think.

Robin:

Okay. There's a mainly about like the overall trend versus how big are the deviations, right? It's interesting to study the deviations, but if the deviations are mostly small relative to the overall trend, that's the thing to be worried about.

Lyman:

Because child mortality tends to change in a relatively stable way once it begins to change, the scale of deviation is basically unchanged. The scale of within-country deviation over time is basically unchanged by including child mortality. But the mean that you would, say, calculate a coefficient of variation on is much lower. And as a result, once you account for child mortality, you get a lot more cases of countries that had no real decline, or that had very little decline, or that even maybe had an increase. I mean, a lot of countries, their peak fertility in their entire history was in like 1980, once you account for child mortality. Some of them was in like 1990. So, particularly in like less developed countries. I think not accounting for child mortality is a serious systematic error when we try to think about what's really happening with fertility. Now, there's still a decline. There is, on average, a decline. But it's much more modest and much more variable.

Robin:

Okay, but fundamentally, our key question is about the scale of variations. That is, we see a trend, we see modest variations in the trend, they tend to have modest scale and then modest duration. And our stories about the overall reversal have to involve stories of much larger variations, right? And so the question is, how plausible is it that things of the size we've seen, some tail of that distribution, how often or some other effect will produce a much larger deviation?

Lyman:

We have a lot of countries with very large deviations from the global norm and the deviations that don't appear to be immediately going away. Whether that source of deviation can be transported to another country is the big question mark. And the answer is possibly not. We have far fewer cases of a country that was not deviant becoming deviant. Right? That is the rare event. And yeah, that doesn't happen frequently. Depending on the level of spatial analysis you're willing to use, you do see more cases of it. Because there are a lot of places where like a US state got above average, that was below average, went way above average for like 50 years at a time, like two or three generations.

Robin:

Does that make its neighboring states also go up? I mean, it's more that these are independent, right? The more that these are actually independent, the more that we have to worry that, you know, there isn't going to be an contagion that produces a bigger effect, right?

Lyman:

Yeah, so I don't think that we have an example of one country jumping above, jumping way up, doing something to jump way up, and then it infects all their neighbors. No, we don't have that in the way that we do have that for fertility declines. For example, after the Bradlaugh-Bessant case in 1876, we have a really clear case of contagion via basically the anglosphere. Or, like, proximity to the French language after the mid-18th century, we know has a very clear contagion effect.

Robin:

Marc Thiessen And so those contagious effects you're attributing to this shared development culture, i.e. a shared model of what the West is and what made it great and why people should copy it. And then a key question is, what's the prospect for the decline of that shared culture sense of what makes us great?

Lyman:

Yeah. So an interesting hypothetical here is if China sinks a couple aircraft carriers and takes Taiwan, Because as much as that sounds like it would be like really bad for fertility, there's an interesting sort of counterfactual world in which suddenly everyone decides that like, really, truly this time around, we mean it. The West is not a model. Right?

Robin:

Like, if... But China as a model is no better, right? No, no, no.

Lyman:

But this is the thing is that China's not about to be culturally hegemonic either. Okay? They're just not. Partly they don't, for a lot of reasons, they're not going to become. But it's an interesting question. Is there something that could truly shatter kind of Euro-American pop culture dominance. And the thing that would seem like a possibility to this is basically like a catastrophic military defeat for NATO or for the U.S. And I don't mean like Ukraine grinding on, like a genuine military battle.

Robin:

Well, Russia just wins in Ukraine. Is that enough?

Lyman:

No, I don't think so.

Robin:

Well, then Taiwan isn't that different from Ukraine, is it?

Lyman:

I mean, it's different because it would involve I mean, we have defensive commitments to Taiwan. It'll involve a lot of dead Americans, whereas Ukraine were very clearly like not really trying.

Robin:

to keep Ukraine going. But losing Vietnam didn't really do much for development culture worldwide, right?

Lyman:

That's true. That's fair. That's fair. So maybe I'm wrong. Maybe the US has to lose Alaska or something, or Hawaii. Yes, when China annexes Hawaii, then we'll realize. No. So it's not that I'm wishing for this, to be clear. But I do think you have to start asking, what are the things that could make people be like, oh, Maybe this really won't work. I mean, actually, maybe the more plausible one is like North Korea wins against South Korea, because that's a story where you have a really clear fertility differential as well. North Korea is now, we think, like 1.7, and South Korea is 0.7.

Robin:

Marc Thiessen But they're both for a blow-up replacement.

Lyman:

Danielle Pletka Yeah. Although North Korea was above replacement as of like 10 years ago, probably.

Robin:

Marc Thiessen But the thing we agree on is that we have a pretty strong world culture that's pretty antinatal, and We need to imagine pretty large deviations from that in order to imagine a revival of fertility worldwide. Small deviations aren't going to do it here.

Lyman:

If we want to, I don't know that we want to revive fertility worldwide. And I know that sounds weird, but I'm not over here being like, oh, I really am troubled by China's low fertility rate. I'm like, no, this is great. This is good for the US. I would like countries friendly to America to have a fertility revival. But again, I don't think we need And I don't think many people sincerely desire a global fertility revival. Um, I think most of, I think I would be happy.

Robin:

Even a third of a global revival is still needs a pretty unusual fluctuation from what we've seen. Right.

Lyman:

I would say like, I would be happy, like, I think I would be, like, halfway to good enough if, like, my religious denomination and the five or six other most similar denominations had, like, a TFR of three. Like, at that point, I would be like, my work here is mostly done.

Robin:

Okay. But the prior odds of that have to be much less, right? Yes. That's a much less likely fluctuation.

Lyman:

But what I'm saying is, like, Actually, I think it's entirely possible that a social organization that scale could do that. And what I'm saying is I think actually for most people, the community that actually needs to have higher fertility for their concerns to be met, for their life to remain livable through their grandchildren's life, is not the whole world. It's like not even the entire U.S. trade network. It might not even be the U.S. Because although there will be burdens, the reality is life is not about to become unlivable because of low fertility. It's become somewhat less pleasant. Slower. Yeah. Yeah. But so like I think for a lot of people, The wind condition here is not a cessation of the decline of global fertility. I mean, to be clear, actually, like, because for religious reasons, because I believe in the value of human life, yes, I would actually like global decline to stop because I think humans are great and God loves them. But, um, but besides that, like at like a sociological level, like the actual groups that I need to have higher fertility for me to be like, okay, I'm a lot less concerned is not everyone. Now, politically, it's necessary to do things for everyone. And at a more meta-ethical level, yes, I actually do want good things for all people. I do not wish ill on people, et cetera, et cetera, particularly my fellow co-nationals in the US. But I don't think we need to say, well, this won't fix the entire world. Well,

Robin:

even a big chunk of the world requires a quite unusual fluctuation relative to what we've seen, is just my point.

Agnes:

I think it's really, I'm going to say one last thing and then we stop soon, but you guys can also have your last words. I think it's really interesting that like Lyman, who is, you know, attuned to this issue more than almost anybody has this response to it, which is like, well, I care about is like fundamentally is like, you know, my like, like, pre-local group. I spent, like, two weeks in South Korea, like, a month ago, and I asked lots and lots of people about this question. Like, are you worried? Like, I'm worried for you. Are you worried? And they were not worried. I was just stunned. Like, at the level of, this is not something that bothers them. But they do care, like, okay, my sister's gonna have a kid, or... And that there's something about this issue where thinking about it the way Robin does, like, at the level of the world, where it's the world going, is not... a natural way that people think about it. And it this might be, I mean, this is already true of like, you know, environmental issues, right? That it's hard to think as on as big a scale as we're required to. I think it might even be more true about this issue than about environmental issues, because in some way, it's like it triggers the idea of family. or the idea of group or in group or whatever. And it's like, that's what we're like, what do we care about moving forward into the future? There's just something very deep in like, you know, culture that says, not everybody, my people.

Robin:

I just want to thank Lyman for coming on, because this is a topic of great interest to me, and he's a great expert. And thank you for talking about it here.

Lyman:

I mean, it was great being with y'all. It's a fun conversation. I'll just say, I should clarify because otherwise I'll get tweets. You know, we do, when I say like, you know, what would make me happy? Yeah, I want my group to do well. Yes, I earnestly desire the good of others as well. And I advocate for policies that will benefit everyone. But I actually think, With environmentalism, the focus on the particular is actually a barrier to doing what's necessary, right? Somebody's like, well, I used my reusable straw, so it doesn't really matter what policy happens or something. The individual moral burden is actually an obstacle, because the simple fact is your individual environmental choices really do nothing for the global environment. they're just offset by price functions on the other side, like that's it. That is not the case with fertility, partly because there's no market for children. So there's not like a price offset. But it really is the case that like when you have a kid, there's one more kid. And so like individual moralizing about fertility is probably useful. And I also think that on some level, I think So aside from developmental idealism, there's another argument that I think is credible, that transcendental norms are really important. That is, if you think that you are part of a group and the perpetuation of that group matters, and that if that group is not perpetuated beyond your death, Like if you knew that the group would not be there for your great-grandchildren, you would feel worse about your life today. Okay? If you don't have that kind of feeling, you don't have as many kids. People who have kids are people who feel like they have a stake in groups that are long-term oriented. And this is why a lot of the tech pronatalists are also long-termists. Okay. Okay. There's like actually like a psychological argument here that fertility is inherently hacking into the transcendence function in your brain. That basically it's saying, I need you to value something more than death. Okay.

Robin:

And so I think- And that's not very common in the world today.

Lyman:

Right. Right. So I think on some level, recultivating various kinds of identities is useful. Now, I'm not saying, you know, white nationalism for the babies. Okay. Um, partly because that's also not a useful group. Like white people are way too diverse of a group. Okay. Like what you need is like, like, like South Central Kentuckians. Like heretic philosophers, right, Agnes? Yeah, yeah. Or like the poleis, okay? What we really need is like the polis, okay? Like a relatively human-scale community. Like my denomination is like, you know, on a given Sunday, half a million people, half a million adults are in our churches. So like the functioning worshiping population is about a half a million people. That's a totally socially manageable group size. That's enough that you go to another church, talk to a random person, three degrees of separation, you know someone in common. It's that simple. So like, you need people to be investing in those scale communities and to say, like, it's actually worth it to me to make big sacrifices for that kind of group. And that's not the norm right now, but it is by revitalizing that kind of scale of human sociality, I think that we would possibly be able to touch the transcendence element of kind of the big vision question.

Agnes:

Good note to end on.

Robin:

All right. Thank you very much.