"My third recommendation is to formulate forward-looking plans for immigration, a highly sensitive subject that nevertheless demands strategic forethought."
It is an interesting self-made "problem", and being self-made it would seem the solution is also within reach if the will to act is there.
What do people need to see before having more kids? Time and Money. China has very unequal pay, horrific working hours culture that penalizes home life, sex and age discrimination *out in the open*. ALL of this could theoretically be fixed by the stroke of the legislative pen.
Thanks for this comment, Paul. Time and money certainly matter, and some of the constraints you point to can indeed be improved through policy. But Zhang’s point is that even fixing these may not be enough on their own. East Asian experience suggests that even substantial expansions of childcare support, cash incentives and/or family-policy reforms have often not reversed ultra-low fertility, at best stabilising it or delivering only modest, fragile upticks, because fertility decisions are also shaped by other forces (e.g. socialisation, expectations, relationship skills etc.) formed earlier in life. In that sense, legislative fixes are necessary, but not sufficient.
It is true, what’s done is done. Most of these outcomes are “foretold” as it were, but I get frustrated seeing what I see a not only injustice but willful lack of interest in improving society. It’s not just a China thing, too many narrow in, it’s a human thing. Policy change is necessary of course, but what is needed here is a change of structure. Tax wealth not work. I’m convinced the significant advanced economy to gets this first owns the future.
Don't know if they are right, but a lot of the authors push back on the argument that AI will compensate - they say innovation capacity and AI leadership strongly linked to population. Immigration isn’t about numbers; it’s about preventing epistemic closure. > What's that mean?
By “epistemic closure” I mean the risk that demographic contraction + cultural homogeneity narrows the range of ideas, assumptions, and innovation paths even if technical capacity remains high. Immigration matters less for headcount than for maintaining cognitive variance. AI can scale coordination, but it doesn’t automatically prevent intellectual lock-in.
Thanks for the reply! Interesting point - it makes absolute common sense to me, but are there any studies linking innovation to communities with high levels of migration that you know about?
The demographic contraction framing misses something crucial: China's real challenge isn't population numbers but the social contract between generations. When 400 million workers supported 1.4 billion through industrialization, that was the anomaly - not this return to demographic equilibrium. What happens to a civilization built on upward generational mobility when the pyramid inverts?
I largely agree with the judgment that China’s demographic problem has already entered a phase of accelerating deterioration. While I still think it is very hard to make a firm call that China’s population will fall to 400 million more than 80 years from now, by 2100, I do think the nearer-term projection is much more credible: over the next 25 years, China’s population could decline from 1.4 billion to 1.172 billion, a drop of 16%, roughly equivalent to subtracting the combined populations of Russia and Japan.
I also agree with Prof. Zhang’s argument that China’s ultra-low fertility is fundamentally a problem of institutional environment and life structure, rather than simply a matter of whether families receive enough subsidies. Her framework is useful precisely because it puts educational competition, long working hours, job insecurity, gender division of labor, and the weakening of young people’s relationship-building capacity into the same analytical frame. South Korea is a good example. For many years it has pursued strong pro-natal policies, yet fertility improvement has remained extremely limited because the underlying structures of education, work, and family life have not been fundamentally changed.
I made a similar point in another post, How China’s Housing and Education “Funding Cycles” Suppress Consumption. China’s education spending, relative to per capita income, is significantly higher than in most developed economies. Because educational competition is so intense, families are forced to prepare financially not just for a few extra years, but often for a decade or more of education-related spending. That is one important reason why discretionary consumption remains weak in China, and also one of the structural forces driving fertility lower.
Where I am less convinced is Prof. Zhang’s view that population decline necessarily implies a sharp deterioration in innovation capacity. I addressed this in an earlier post, Beyond Demographics: China’s Shift from Population Dividend to Productivity Dividend. What ultimately determines innovation efficiency is not population size alone, but also the quality of education, capital depth, the level of automation, the organization of scientific research, business institutions, urban agglomeration, and international technological linkages. Japan, South Korea, and Germany have already shown that population aging can weigh on potential growth, but it does not automatically destroy a country’s technological capabilities. And in many of these areas, China still has substantial room to improve.
Population decline will indeed put pressure on real estate, basic consumption, and parts of labor-intensive supply. But China differs from many aging economies in one crucial respect: it still possesses a powerful industrial system, dense infrastructure, and strong policy execution capacity. That means China may not follow the same path as countries where demographic decline translates more directly into broad-based stagnation. A more plausible scenario is slower aggregate growth, rising pressure on pensions and social welfare systems, and contraction in some sectors, while at the same time the impact is partially delayed and offset through robotics, AI, industrial upgrading, and higher labor-force participation.
This reminds me of programs that native Americans children were forcibly taken and learn Christianity . But after this they still chose to be with their origin
"While our per capita income is far lower than Hong Kong’s, the price of domestic services here is higher and the quality lower, largely because Hong Kong employs Filipino domestic workers."
In other words, there is more income inequality in Hong Kong than in China — due to immigration. Are you really sure you want to go that route?
Crying over declining fertility is trifling. In 80+ years China will have 400 m people. Given that fertility is declining fast everywhere (the Earth is defending itself), it will still be a very populous country. Projecting how the world will look like in 80 years is just like looking into a crystal ball or tea lives or stars. No one knows. In 1950, no one could envisage cell phones, or robots of today. Finding a solution to low fertility by relaxing competition in education is laughable. North Atlantic countries did that and the results are generations of mediocre work force. And the fertility rates dropped nonetheless. Beijing University professor is dead wrong. In any case gaokao is at the age of 18, too early for reproduction.
I think it’s less the earth defending itself and more incentive/disincentives changing for modern urbanites, but where I’d agree is that China has a history of over-engineering solutions to social problems… That being said, I think it an ageing population presents problems that we can’t expect tech to find a quick fix too. Also, don’t think education point is as laughable as you suggest - raising a kid that hyper-competitive environment would be a minus for me.
Nothing wrong with competitiveness. The oldies of today all grew up in such environment. And we were actually learning at school. Nowadays, not only competition is suppressed and mediocrity favored, but the kids do not learn much at school. Good luck with the future.
To be fair, when Zhang refers to addressing intense competitiveness, I don't believe she is advocating reducing standards, but rather making the standards broader so as to reward a broader range of abilities than attaining high marks in often quite mechanical tests. Whether such a change would be possible without opening the way to corruption is a fair question, but as a teacher of university students she does at least elaborate on the challenges she had encountered within the current system of turning exemplary Gaokao candidates into original researchers:
"[Some students], having become accustomed to intensive practice-exercise work, remain fixated on whether they can improve themselves by doing more practice problems, often without developing the awareness needed to construct their own knowledge frameworks from the subjects they study. Yet truly mastering advanced knowledge and transforming it into creativity or productive capacity depends not on exam-oriented techniques, but on the ability to synthesise knowledge, think independently, and collaborate meaningfully."
The problem with prof. Zhang’s arguments is that she is linking the strictness of exams to declining fertility. Declining fertility is a worldwide phenomenon. It has nothing to do with the strictness of university exams. They used to be very strict in Europe. Their relaxation did not prevent the fertility downward trends.
Make education the gatekeeper and find it is the slave plantation owner?
We did that in America.
We’re rethinking that.
The point of having children has become to get them the best education- for many Americans. The children exist as grist for the education mills. As the wheels grind ever finer don’t be surprised the young don’t want to have children. That is our experience in America.
Congratulations on making our mistakes, just with far more focus.
It is an interesting self-made "problem", and being self-made it would seem the solution is also within reach if the will to act is there.
What do people need to see before having more kids? Time and Money. China has very unequal pay, horrific working hours culture that penalizes home life, sex and age discrimination *out in the open*. ALL of this could theoretically be fixed by the stroke of the legislative pen.
So do it.
Thanks for this comment, Paul. Time and money certainly matter, and some of the constraints you point to can indeed be improved through policy. But Zhang’s point is that even fixing these may not be enough on their own. East Asian experience suggests that even substantial expansions of childcare support, cash incentives and/or family-policy reforms have often not reversed ultra-low fertility, at best stabilising it or delivering only modest, fragile upticks, because fertility decisions are also shaped by other forces (e.g. socialisation, expectations, relationship skills etc.) formed earlier in life. In that sense, legislative fixes are necessary, but not sufficient.
It is true, what’s done is done. Most of these outcomes are “foretold” as it were, but I get frustrated seeing what I see a not only injustice but willful lack of interest in improving society. It’s not just a China thing, too many narrow in, it’s a human thing. Policy change is necessary of course, but what is needed here is a change of structure. Tax wealth not work. I’m convinced the significant advanced economy to gets this first owns the future.
China is already compensating for demographic decline with synthetic capacity.
The problem is not labor, it’s coordination and diversity.
Immigration isn’t about numbers; it’s about preventing epistemic closure.
Don't know if they are right, but a lot of the authors push back on the argument that AI will compensate - they say innovation capacity and AI leadership strongly linked to population. Immigration isn’t about numbers; it’s about preventing epistemic closure. > What's that mean?
AI is a tool, where are the hands?
By “epistemic closure” I mean the risk that demographic contraction + cultural homogeneity narrows the range of ideas, assumptions, and innovation paths even if technical capacity remains high. Immigration matters less for headcount than for maintaining cognitive variance. AI can scale coordination, but it doesn’t automatically prevent intellectual lock-in.
Thanks for the reply! Interesting point - it makes absolute common sense to me, but are there any studies linking innovation to communities with high levels of migration that you know about?
You can't turn AI into consumers. That's the biggest problem.
The point of immigration is indeed epistemic closure … just not the way people think it means. 💀
The demographic contraction framing misses something crucial: China's real challenge isn't population numbers but the social contract between generations. When 400 million workers supported 1.4 billion through industrialization, that was the anomaly - not this return to demographic equilibrium. What happens to a civilization built on upward generational mobility when the pyramid inverts?
Absolutely, not so much absolute population as the ageing of the population.
If China permits Immigration, it will be the end of their civilization
Immigrants bring democracy, liberalism and woke communism
Depends on the immigrant surely!
I think China already has communism.
That last word is hilariously employed.
🤣
@Paul Millerd default-path-ness as a national security problem 😂
I largely agree with the judgment that China’s demographic problem has already entered a phase of accelerating deterioration. While I still think it is very hard to make a firm call that China’s population will fall to 400 million more than 80 years from now, by 2100, I do think the nearer-term projection is much more credible: over the next 25 years, China’s population could decline from 1.4 billion to 1.172 billion, a drop of 16%, roughly equivalent to subtracting the combined populations of Russia and Japan.
I also agree with Prof. Zhang’s argument that China’s ultra-low fertility is fundamentally a problem of institutional environment and life structure, rather than simply a matter of whether families receive enough subsidies. Her framework is useful precisely because it puts educational competition, long working hours, job insecurity, gender division of labor, and the weakening of young people’s relationship-building capacity into the same analytical frame. South Korea is a good example. For many years it has pursued strong pro-natal policies, yet fertility improvement has remained extremely limited because the underlying structures of education, work, and family life have not been fundamentally changed.
I made a similar point in another post, How China’s Housing and Education “Funding Cycles” Suppress Consumption. China’s education spending, relative to per capita income, is significantly higher than in most developed economies. Because educational competition is so intense, families are forced to prepare financially not just for a few extra years, but often for a decade or more of education-related spending. That is one important reason why discretionary consumption remains weak in China, and also one of the structural forces driving fertility lower.
Where I am less convinced is Prof. Zhang’s view that population decline necessarily implies a sharp deterioration in innovation capacity. I addressed this in an earlier post, Beyond Demographics: China’s Shift from Population Dividend to Productivity Dividend. What ultimately determines innovation efficiency is not population size alone, but also the quality of education, capital depth, the level of automation, the organization of scientific research, business institutions, urban agglomeration, and international technological linkages. Japan, South Korea, and Germany have already shown that population aging can weigh on potential growth, but it does not automatically destroy a country’s technological capabilities. And in many of these areas, China still has substantial room to improve.
Population decline will indeed put pressure on real estate, basic consumption, and parts of labor-intensive supply. But China differs from many aging economies in one crucial respect: it still possesses a powerful industrial system, dense infrastructure, and strong policy execution capacity. That means China may not follow the same path as countries where demographic decline translates more directly into broad-based stagnation. A more plausible scenario is slower aggregate growth, rising pressure on pensions and social welfare systems, and contraction in some sectors, while at the same time the impact is partially delayed and offset through robotics, AI, industrial upgrading, and higher labor-force participation.
Excellent article.
This reminds me of programs that native Americans children were forcibly taken and learn Christianity . But after this they still chose to be with their origin
"While our per capita income is far lower than Hong Kong’s, the price of domestic services here is higher and the quality lower, largely because Hong Kong employs Filipino domestic workers."
In other words, there is more income inequality in Hong Kong than in China — due to immigration. Are you really sure you want to go that route?
Crying over declining fertility is trifling. In 80+ years China will have 400 m people. Given that fertility is declining fast everywhere (the Earth is defending itself), it will still be a very populous country. Projecting how the world will look like in 80 years is just like looking into a crystal ball or tea lives or stars. No one knows. In 1950, no one could envisage cell phones, or robots of today. Finding a solution to low fertility by relaxing competition in education is laughable. North Atlantic countries did that and the results are generations of mediocre work force. And the fertility rates dropped nonetheless. Beijing University professor is dead wrong. In any case gaokao is at the age of 18, too early for reproduction.
I think it’s less the earth defending itself and more incentive/disincentives changing for modern urbanites, but where I’d agree is that China has a history of over-engineering solutions to social problems… That being said, I think it an ageing population presents problems that we can’t expect tech to find a quick fix too. Also, don’t think education point is as laughable as you suggest - raising a kid that hyper-competitive environment would be a minus for me.
Nothing wrong with competitiveness. The oldies of today all grew up in such environment. And we were actually learning at school. Nowadays, not only competition is suppressed and mediocrity favored, but the kids do not learn much at school. Good luck with the future.
To be fair, when Zhang refers to addressing intense competitiveness, I don't believe she is advocating reducing standards, but rather making the standards broader so as to reward a broader range of abilities than attaining high marks in often quite mechanical tests. Whether such a change would be possible without opening the way to corruption is a fair question, but as a teacher of university students she does at least elaborate on the challenges she had encountered within the current system of turning exemplary Gaokao candidates into original researchers:
"[Some students], having become accustomed to intensive practice-exercise work, remain fixated on whether they can improve themselves by doing more practice problems, often without developing the awareness needed to construct their own knowledge frameworks from the subjects they study. Yet truly mastering advanced knowledge and transforming it into creativity or productive capacity depends not on exam-oriented techniques, but on the ability to synthesise knowledge, think independently, and collaborate meaningfully."
The problem with prof. Zhang’s arguments is that she is linking the strictness of exams to declining fertility. Declining fertility is a worldwide phenomenon. It has nothing to do with the strictness of university exams. They used to be very strict in Europe. Their relaxation did not prevent the fertility downward trends.
Make education the gatekeeper and find it is the slave plantation owner?
We did that in America.
We’re rethinking that.
The point of having children has become to get them the best education- for many Americans. The children exist as grist for the education mills. As the wheels grind ever finer don’t be surprised the young don’t want to have children. That is our experience in America.
Congratulations on making our mistakes, just with far more focus.
You think education is hyper-competitive in America in the same way? Just curious.
In terms of the striver parents and social climbers yes.
Standards?
Probably not but that misses the real point.
🆙⬆️ is the point, not excellence.
What other nations do is their business entirely of course.
The point here 🇺🇸 is power and money.
It’s almost entirely political, indeed here 🇺🇸 the school board is the most powerful local government.
It’s also our 🇺🇸 chief arm of Imperialism; we seldom invade-we educate their young elite and wait.
What’s wrong with only 400 million people?
Nothing. It’s the population structure that’s so dire. You’d have more retirees than children. She addressed this in the article
Exactly - some do make the argument that bigger population = bigger pool of talent etc., but it's the ageing population that is the big problem.