No one ever had kids because the cost-benefit analysis looked good. (Maybe some immigrant brides) In the long run, kids do not make economic sens: in fact, in the long run nothing does. But in the long run, nothing makes as much non-economic sense as kids.
It’s literally only in the long run that kids make sense. But you have to factor in their lifetime and children’s lifetimes, etc. from a pure economic perspective, having children could produce the kind of gains only .001% of the population attain in a lifetime. Children also greatly reduce the risk of absolute destitution later in life.
I know multiple couples who have explicitly not had kids because of a cost benefit analysis.
I suppose the difference between us might be related to social sorting. Perhaps I am unusually likely to know people who do not feel strongly one way or the other, and who therefore decide based on costs and benefits. And perhaps this situation is also highly unrepresentative. But among professionals of various types in the relevant age range in the US, it does not seem uncommon to me.
Short of fairly revolutionary social engineering, there is not much that a government could do to flip the decisions of these cost-benefit minded people.
Not to dissuade you from interesting quantitative research, but we are far more prosperous than ever, it's not lack of resources that prevents people from having kids. Who has the most children in Europe? Gypsies and immigrants - the poorest lot.
Most young people just do not envision themselves having children early, perhaps when they're 30, and then it's suddenly hard to find a partner or fertility problems kick in. Hedonistic individualism, feminism, overall value confusion.
Someone posted online the minutes of a meeting of the french government about legalizing the pill. De Gaulle and his ministers were aware that the pill would further suppress fertility but hoped that it will be counterbalanced with the help of family and housing spending while others pointed that no interventions are known to work.
The most important question facing our species and in 60 years we made no progress.
Hess of "More Birth" @MoreBirths has dug there yet: "it's culture, stupid". He also replicates his finding in Mongolia and Georgia (the caucasian state). If you want to raise the TFR, you must make childbearing cool/prestigious.
A problem with many of these spending programs is that whatever is left after government inefficiency and corruption goes to poor families, or those that are under some arbitrary income levels. We know what that means, stuff like locals having to pay for private childcare because public ones are full of immigrant kids. And parents being discouraged from being productive so they stay under that income threshold and keep getting benefits. So essentially locals don’t have kids in part because they have to subsidize other’s kids, and bureaucrats.
Tax breaks should work better, but if they mean percentage discounts, instead of absolute deductions. For example, 25% discount on your income tax for each child after the first one, so you pay 0 on your 5th child (you’d still pay some other kinds of taxes). I know you studied that in this article but I think we’d need to see what those deductions really mean. If they mean something like paying €500 less in taxes a year, that’s obviously not going to change anything for productive people, only for parasites.
Anyway, as a libertarian I don’t like any of those options, we should just get rid of most of what governments do and drastically reduce taxes, so that people have much more money and are encouraged to produce more as they’d keep most of it. That, among other things, means people could afford having more kids and spending more time with them.
I think a big factor why people have less kids now in the developed world is because family members rarely stay in the same place, they move to different places to study or work and don’t come back. If where you live you have many relatives, that means you can easily delegate raising the kid among people you trust. That creates like a network effect, the more family around, the more convenient it becomes to grow the family.
- Chile's COVID-related family emergency policy: This program cost around $8.6 billion in 2020 and $21.3 billion in 2021. Those figures agree with the 2.15% (2020) and 7.34% (2021) in cash transfers to families that the OECD reports (*). Thing is, the program was not restricted to families _with_ children, as you can see in the Chilean government site (https://www.desarrollosocialyfamilia.gob.cl/apoyos-sociales) where beneficiaries of the program are defined as households (hogares), not families. Households without children were not paid less on a per capita basis. I know the OECD report (**) says "provided large scale financial aid to support families with children", but this is wrong (they also got wrong the year when it was phased out, which was 2021 not 2022).
- The OECD SOCX Manual (***) states that family payments are "child allowances and credits, childcare support, income support during leave, single parent payments;". We already saw above that this is not correct, as they included payments to childless households in the "family benefits" category. Also, why should we expect that single parent payments would encourage parents to have more children?
- The OECD criteria is not consistent if we are trying to account for payments that might incentivize having children: "For example, income support during parental leave paid by a public insurance fund is ‘public’; legally required continued wage payments by employers to fathers on paternity leave are ‘mandatory private’" (****). Some parental leave payments are considered as public spending on family benefits in cash, while others are not.
To summarize, I don't think "family spending" as recorded by the OECD qualifies as "[paying] people ( in various ways) to have more kids". It's a mixed bag of various payments, and it's not intended to be a comprehensive tally of all payments incentivizing child birth.
Fascinating essay. I think one possible answer might be tax breaks for employers to encourage flexibility towards working women. What seems to be happening is that globally women's employment negatively correlates with TFR, but flex tends to buffer or mitigate the trend.
I wonder whether it might be possible to assign countries policies a value for flex and then compare to TFR. Grok stated the following: In Italy/Poland (low flex), employed women have 0.2–0.3 fewer kids; generous leave narrows gap.
It's worth noting that some of the early evidence shows that educated/remote women see biggest fertility gains (+10–15% intentions).
I think you're right- broadly speaking government spending doesn't improve TFR, but there might be a specific exception for economic freedom and reforms which allow more choice and less economic necessity for women to work full-time.
I don't know how you would go about testing this hypothesis. There are few good databases but I'm sure whether they compile their data in a format which would be useful.
You could also look at Swiss data where women were de facto excluded from the workforce much later than the rest of Europe (until very recently) despite having an otherwise similar society and economy.
That's a great point and highly illuminating. Before full inclusion in the workforce Swiss women had higher TFRs. However, when they did rapidly enter the workforce, without many of the childcare provisions, leave entitlements, and flex arrangements enjoyed by women in other Western countries it led to decline in TFR below the average rate for Europe.
It does seem as though flex and other freedom to choose rights and provisions mitigate against birthrate population decline. Here's the thing- I'm very sceptical of government inserting itself into voluntary arrangements like contracts. This is why I generally tend to argue for slightly higher rates of corporation tax and employer taxes on employees, with a view to providing very generous tax incentives and rebates for approaches which expand choice and personal liberty. Ironically, the best policy in this suite is only tangentially related to liberty- capital expenditure programs enhance liberty by making everyone more prosperous over time. It's one of reasons why societies tend to become wealthier per capita during tight labour market conditions.
Somebody ... Lyman Stone? ... said that what counted is marrying earlier. Married people keep having kids until they are too old to do so. You can imagine how earlier married could drive the 15 years or more married statistic, but is that what the data shows?
The evidence that income/wealth (positively but weakly) cause fertility is fairly solid, but this might not translate to resource gains that come from the government.
IIRC, the "correct" statistical procedure in this case would be to consider this a natural experiment for difference-in-differences with stratified sampling.
That is:
1) Take a time series dataset and mark a set of countries that introduced family spending legislation at a certain date as the treatment group
2) Cluster the data based on reasonable covariates, e.g., GDPpc, postsecondary education rates, etc. such that a treatment country is the center of each cluster. Ensure at least one control country (.e. a country that did not implement any such policy that is similar to one that did) exists to compare to each treatment country
3) Compute the test statistic for the difference-in-differences pooling the DD for each cluster, which can give you a p-value.
This would also get around the nonstationarity of the time series due to COVID
My bet is still with the null, though
One caveat: I recall it being mentioned there are transient effects, that is, people suddenly moving their births, but having a net zero cohort effect, so you may have to look into that and compensate (e.g. by censoring a couple years before and after the legislation is introduced)
Back in 1939 NZ changed taxation in a way that penalised families. At the time NZ's TFR was about 2.5 per woman. It caused a great deal of bitterness among voters who had kicked off a baby boom shortly before the Labour government was elected into power in 1935 (they initiated many social welfare policies that supported working class voters). This boom peaked in 1960 at 4.1 per women, 20 years after the tax penalty was emplaced. Curiously1960 was the year after the tax rules were returned to neutrality. Voters clearly didn't get the message as births dropped precipitously and persistently for the following 20 years. In 1980 NZ's TFR stabilised around 2.1 where it persisted until about 2010, since when it has persistently declined to its current 1.6 per women. God knows why, as family taxes haven't changed
"This boom peaked in 1960 at 4.1 per women, 20 years after the tax penalty was emplaced. Curiously1960 was the year after the tax rules were returned to neutrality."
Even more curiously, 1960 was when the first oral birth control pill was available to the public, at least in the United States.
In NZ, GP's could prescribe the Pill to married women from 1959. It became generally available to NZ women (via prescription) in 1970 (if I recall correctly).
I write my comment somewhat tongue in cheek. Tax breaks were clearly too insignificant in comparison to other social changes that occurred for their effect to be isolated.
Maybe you'll get an effect if instead of current benefits your pension would linearly depend on all taxes, tolls and social duties paid by your kids, grand-kids, grand-grand-kids etc...
I suspect that the real driving factor behind TFR in the West is something pretty nebulous like “stable optimism about the future”. Good luck getting data on that. All of these more visible measures are just trying to indirectly nudge that dial.
"The world is doomed" -> less kids are born -> "The world is doomed" seems like a pretty vicious feedback cycle, yes. Someone is just going to have to explain to people that the world being fucked is actually when it's most important to have kids.
No one ever had kids because the cost-benefit analysis looked good. (Maybe some immigrant brides) In the long run, kids do not make economic sens: in fact, in the long run nothing does. But in the long run, nothing makes as much non-economic sense as kids.
It’s literally only in the long run that kids make sense. But you have to factor in their lifetime and children’s lifetimes, etc. from a pure economic perspective, having children could produce the kind of gains only .001% of the population attain in a lifetime. Children also greatly reduce the risk of absolute destitution later in life.
I know multiple couples who have explicitly not had kids because of a cost benefit analysis.
I suppose the difference between us might be related to social sorting. Perhaps I am unusually likely to know people who do not feel strongly one way or the other, and who therefore decide based on costs and benefits. And perhaps this situation is also highly unrepresentative. But among professionals of various types in the relevant age range in the US, it does not seem uncommon to me.
Short of fairly revolutionary social engineering, there is not much that a government could do to flip the decisions of these cost-benefit minded people.
Penny wise, life foolish
Not to dissuade you from interesting quantitative research, but we are far more prosperous than ever, it's not lack of resources that prevents people from having kids. Who has the most children in Europe? Gypsies and immigrants - the poorest lot.
IMO it's almost completely cultural
https://x.com/MoreBirths/status/1944066771050188894
Most young people just do not envision themselves having children early, perhaps when they're 30, and then it's suddenly hard to find a partner or fertility problems kick in. Hedonistic individualism, feminism, overall value confusion.
It's not the absolute cost so much as the opportunity cost (that, and the fact we made child labour illegal and/or automated most of it away.)
Someone posted online the minutes of a meeting of the french government about legalizing the pill. De Gaulle and his ministers were aware that the pill would further suppress fertility but hoped that it will be counterbalanced with the help of family and housing spending while others pointed that no interventions are known to work.
The most important question facing our species and in 60 years we made no progress.
“Only Israel has positive fertility” seems like a good thread to pull on.
Hess of "More Birth" @MoreBirths has dug there yet: "it's culture, stupid". He also replicates his finding in Mongolia and Georgia (the caucasian state). If you want to raise the TFR, you must make childbearing cool/prestigious.
Israeli TFR isn't quite as impressive once you break down which groups are actually having kids (Arctotherium mentions this here.)
https://www.aporiamagazine.com/p/communist-pro-natalism
The 'pill' turned sex into mostly recreation.
A problem with many of these spending programs is that whatever is left after government inefficiency and corruption goes to poor families, or those that are under some arbitrary income levels. We know what that means, stuff like locals having to pay for private childcare because public ones are full of immigrant kids. And parents being discouraged from being productive so they stay under that income threshold and keep getting benefits. So essentially locals don’t have kids in part because they have to subsidize other’s kids, and bureaucrats.
Tax breaks should work better, but if they mean percentage discounts, instead of absolute deductions. For example, 25% discount on your income tax for each child after the first one, so you pay 0 on your 5th child (you’d still pay some other kinds of taxes). I know you studied that in this article but I think we’d need to see what those deductions really mean. If they mean something like paying €500 less in taxes a year, that’s obviously not going to change anything for productive people, only for parasites.
Anyway, as a libertarian I don’t like any of those options, we should just get rid of most of what governments do and drastically reduce taxes, so that people have much more money and are encouraged to produce more as they’d keep most of it. That, among other things, means people could afford having more kids and spending more time with them.
I think a big factor why people have less kids now in the developed world is because family members rarely stay in the same place, they move to different places to study or work and don’t come back. If where you live you have many relatives, that means you can easily delegate raising the kid among people you trust. That creates like a network effect, the more family around, the more convenient it becomes to grow the family.
A few caveats:
- Chile's COVID-related family emergency policy: This program cost around $8.6 billion in 2020 and $21.3 billion in 2021. Those figures agree with the 2.15% (2020) and 7.34% (2021) in cash transfers to families that the OECD reports (*). Thing is, the program was not restricted to families _with_ children, as you can see in the Chilean government site (https://www.desarrollosocialyfamilia.gob.cl/apoyos-sociales) where beneficiaries of the program are defined as households (hogares), not families. Households without children were not paid less on a per capita basis. I know the OECD report (**) says "provided large scale financial aid to support families with children", but this is wrong (they also got wrong the year when it was phased out, which was 2021 not 2022).
- The OECD SOCX Manual (***) states that family payments are "child allowances and credits, childcare support, income support during leave, single parent payments;". We already saw above that this is not correct, as they included payments to childless households in the "family benefits" category. Also, why should we expect that single parent payments would encourage parents to have more children?
- The OECD criteria is not consistent if we are trying to account for payments that might incentivize having children: "For example, income support during parental leave paid by a public insurance fund is ‘public’; legally required continued wage payments by employers to fathers on paternity leave are ‘mandatory private’" (****). Some parental leave payments are considered as public spending on family benefits in cash, while others are not.
To summarize, I don't think "family spending" as recorded by the OECD qualifies as "[paying] people ( in various ways) to have more kids". It's a mixed bag of various payments, and it's not intended to be a comprehensive tally of all payments incentivizing child birth.
(*) PF1.1 from https://www.oecd.org/en/data/datasets/oecd-family-database.html
(**) Page 3: https://www.oecd.org/content/dam/oecd/en/publications/reports/2023/01/the-rise-and-fall-of-public-social-spending-with-the-covid-19-pandemic_0c92ca71/12563432-en.pdf
(***) Page 9: https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/social-spending/socx_manuel_2019.pdf
(****) Page 15 of SOCX manual.
Fascinating essay. I think one possible answer might be tax breaks for employers to encourage flexibility towards working women. What seems to be happening is that globally women's employment negatively correlates with TFR, but flex tends to buffer or mitigate the trend.
I wonder whether it might be possible to assign countries policies a value for flex and then compare to TFR. Grok stated the following: In Italy/Poland (low flex), employed women have 0.2–0.3 fewer kids; generous leave narrows gap.
It's worth noting that some of the early evidence shows that educated/remote women see biggest fertility gains (+10–15% intentions).
I think you're right- broadly speaking government spending doesn't improve TFR, but there might be a specific exception for economic freedom and reforms which allow more choice and less economic necessity for women to work full-time.
I don't know how you would go about testing this hypothesis. There are few good databases but I'm sure whether they compile their data in a format which would be useful.
You could also look at Swiss data where women were de facto excluded from the workforce much later than the rest of Europe (until very recently) despite having an otherwise similar society and economy.
That's a great point and highly illuminating. Before full inclusion in the workforce Swiss women had higher TFRs. However, when they did rapidly enter the workforce, without many of the childcare provisions, leave entitlements, and flex arrangements enjoyed by women in other Western countries it led to decline in TFR below the average rate for Europe.
It does seem as though flex and other freedom to choose rights and provisions mitigate against birthrate population decline. Here's the thing- I'm very sceptical of government inserting itself into voluntary arrangements like contracts. This is why I generally tend to argue for slightly higher rates of corporation tax and employer taxes on employees, with a view to providing very generous tax incentives and rebates for approaches which expand choice and personal liberty. Ironically, the best policy in this suite is only tangentially related to liberty- capital expenditure programs enhance liberty by making everyone more prosperous over time. It's one of reasons why societies tend to become wealthier per capita during tight labour market conditions.
Very interesting - thank you!
I'd like to see a proper paper on this alleged correlation, rather than just "Grok said so".
https://www.sciencedirect.com/science/article/pii/S0927537125001113
https://link.springer.com/article/10.1007/s10680-024-09719-1
There were also a couple of articles on fertility and remote work in US women, but I didn't include them because there were survey-based.
Good point though. I had my first Grok hallucination the other day (that I know of).
In Japanese data until 2010, the data showed two factors get couples to 2.1 Tfr:
1) length of first marriage over 15 years
2) living anywhere but a major city (In Japan, cities under 1 million)
Somebody ... Lyman Stone? ... said that what counted is marrying earlier. Married people keep having kids until they are too old to do so. You can imagine how earlier married could drive the 15 years or more married statistic, but is that what the data shows?
The evidence that income/wealth (positively but weakly) cause fertility is fairly solid, but this might not translate to resource gains that come from the government.
IIRC, the "correct" statistical procedure in this case would be to consider this a natural experiment for difference-in-differences with stratified sampling.
That is:
1) Take a time series dataset and mark a set of countries that introduced family spending legislation at a certain date as the treatment group
2) Cluster the data based on reasonable covariates, e.g., GDPpc, postsecondary education rates, etc. such that a treatment country is the center of each cluster. Ensure at least one control country (.e. a country that did not implement any such policy that is similar to one that did) exists to compare to each treatment country
3) Compute the test statistic for the difference-in-differences pooling the DD for each cluster, which can give you a p-value.
This would also get around the nonstationarity of the time series due to COVID
My bet is still with the null, though
One caveat: I recall it being mentioned there are transient effects, that is, people suddenly moving their births, but having a net zero cohort effect, so you may have to look into that and compensate (e.g. by censoring a couple years before and after the legislation is introduced)
Re tax breaks for families:
Back in 1939 NZ changed taxation in a way that penalised families. At the time NZ's TFR was about 2.5 per woman. It caused a great deal of bitterness among voters who had kicked off a baby boom shortly before the Labour government was elected into power in 1935 (they initiated many social welfare policies that supported working class voters). This boom peaked in 1960 at 4.1 per women, 20 years after the tax penalty was emplaced. Curiously1960 was the year after the tax rules were returned to neutrality. Voters clearly didn't get the message as births dropped precipitously and persistently for the following 20 years. In 1980 NZ's TFR stabilised around 2.1 where it persisted until about 2010, since when it has persistently declined to its current 1.6 per women. God knows why, as family taxes haven't changed
"This boom peaked in 1960 at 4.1 per women, 20 years after the tax penalty was emplaced. Curiously1960 was the year after the tax rules were returned to neutrality."
Even more curiously, 1960 was when the first oral birth control pill was available to the public, at least in the United States.
In NZ, GP's could prescribe the Pill to married women from 1959. It became generally available to NZ women (via prescription) in 1970 (if I recall correctly).
I write my comment somewhat tongue in cheek. Tax breaks were clearly too insignificant in comparison to other social changes that occurred for their effect to be isolated.
Maybe you'll get an effect if instead of current benefits your pension would linearly depend on all taxes, tolls and social duties paid by your kids, grand-kids, grand-grand-kids etc...
"Fertility rates are terrible in every good country except for Israel."
Are you implying Israel is a good country? Define 'good country'.
Hungary was doing well until the government, like a lot of governments at the time, bullied people into taking a fertility-damaging medication.
Really great analysis work.
I suspect that the real driving factor behind TFR in the West is something pretty nebulous like “stable optimism about the future”. Good luck getting data on that. All of these more visible measures are just trying to indirectly nudge that dial.
"The world is doomed" -> less kids are born -> "The world is doomed" seems like a pretty vicious feedback cycle, yes. Someone is just going to have to explain to people that the world being fucked is actually when it's most important to have kids.