The rate at which we find new innovations is roughly proportional to the size of the world economy. Yes, there likely are other ways to promote innovation, but firms and governments have long been trying to promote local innovation, without much success. So this simple innovation-economy relation will probably long continue.
This analysis seems to rely a great deal on a model in which causality runs pretty strictly from population growth to innovation. But I read history more the other way around. Innovation enabled population growth. Going forward, my intuition is that if innovators were surrounded by much better institutional support, we would see more innovation even if world population were an order of magnitude below what it is today.
By "nations" you mean governments. Governments seem to have the ability suppress innovation, at least somewhat. They can also suppress fertility (1-child policy). I agree that the policies that have been enacted with the supposed intent of increasing innovation mostly have not worked. Same with policies with the supposed intent of increasing fertility. But that does not mean that there are no institutional changes that would have the effect of increasing innovation. I didn't say that governments are likely to enact policies that promote such changes, merely that I suspect that such changes exist in principle.
Innovators are indeed surrounded by excellent institutional support. It's called Venture Capital. Please don't tell me that you were thinking of tax-funded state support.
Consider the perspective of a randomly choosen person from all of human history (behind the veil of ignorance). Do they want to maximize innovation **per year**? Certainly not. They would want to maximize **Innovation per person year**.
You don't want to have more people existing at earlier, less productive, times because that makes the average human life less good. Since a smaller population both decreases innovation and person years that pass each year proportionally we should be indifferent to it.
To put it differently, If a demon caused literally everything in the universe to slow to half speed (according to his infernal clock) would that be bad because now it takes 2 years to increase productivity by 5% instead of one year? Surely not. What's relevant is the amount of human experience which occurs at each level of productivity not some arbitrary unit of time.
Yes, obviously, other things being equal, more innovation is good but when you change the total population or population growth rate other things aren't equal.
My point is simply that intuitively attractive choices there give different verdicts about what scenarios are better than you assume and you really should specify what variable you want maximized and check that leads to plausible conclusions.
For instance, if you want to maximize something like expected consumption (or utility) for a randomly choosen member of the species over all of human history sometimes you prefer to spend less human lives to reach a given level of tech
1) No my claim doesn't imply less people will be born in the future in any way. In fact, such a situation is better not worse for my argument because an even greater fraction of all people are born at high productivity times.
2) The worry isn't that literally innovation ceases but that the total amount approaches a finite limit. While I don't think that's the calculation one should be doing, obviously even on my model you end up worse in the case where the population keeps halving.
But, of course, that's not even sorta realistic since you can't keep halving the population -- nor is it intended to be. But once you avoid that issue it just presses the point even more fiercely: if you know that eventually infinitely many people will be born in both scenarios do you care how quickly their per person production increases in years or maximizing the fraction of people who will be born who live at high productivity times.
You've highlighted some interesting issues, but it's a lot worse than people imagine.
1) The downward shift in productivity tends to stem from the shift from production tradables to service sector, although productivity gains in the former tend to be masked by significant product quality increases.
2) With the exception of the tech developments of the past 30 years, historically the analytical mode of thinking taught in universities actually decreases invention, innovation and tinkering, all of which tended to stem from the operational mode of thinking- smart pragmatic people taking real world problems and solving them in their garages and sheds. A more educated workforce inherently means either a loss of innovation or a narrowing of innovation into specific areas.
3) How much of economic growth in capitalism is dependent upon the growth of new markets and new products? In the past, labour reallocation through saturation plus productivity gain has always proven useful, because there have always been new products for labour utilisation. The largest single expansion occurred as a result of women's clothing. Is the rise of the service sector a signal of the end of new markets and new products? If so, it's a rocky road ahead- made all the more worrying by the current billionaire preoccupation with 'where is the best place to ride out an apocalyptic economic collapse'.
If your theory that innovation is proportional to population is correct, then it should be correct if you test it on past events. Here is a counterexample: After the black death, 1350-1450 Europe had lost 50% of its population. Yet what happened right after was the Renaissance. You would not call this innovation in terms of technology, but you can certainly characterize it as a massive mind-opening time. It is possible that the challenges of under-population will bring about something similar in the future.
That's enough years of innovation that I think fertility will be a moot point by then. Either we or our descendants will have radical life extension (if not medical immortality), or they'll have innovations that make it possible for societies to set their own population levels without the need to coax or coerce regular people into mating pairs (womb tanks, robot nannies, etc). Or possibly both.
I'm so confused. How did one of the most insightful thinkers of the last 30 years become so singularly reductionist on this idea? It's entirely possible I'm missing something as I would not assume I'm any more capable than you at thinking about these things (likely much less so).
But you seem so heavily indexed on the 1:1 association of fertility with productivity. Really? With all the advances in individual human productivity through technology and the invention of human+ level AGI in the next few decades, you don't see opportunities for productivity and innovation to develop elsewhere than just absolute number of biological humans?
Perhaps your insight and research has given you more clarity than me, but it also seems possible to me that you've made some early assumptions and then run with them to their deepest conclusions without clearly adjusting for the strength of your priors. What if absolute number of humans isn't the crucial driving force of innovation you assume? Maybe it is, I don't know. But I'm not convinced either way and as a result all these fertility pieces seem like deeply worrying thought pieces that are putting the horse before the cart.
You seem to be blurring different aggregate measures, to describe what seems to be a pessimistic scenario. Let's agree that world output is roughly proportional to population, for a given level of productivity. You are right that, as population declines, we should also expect a decline in total world output. That's a decline in absolute levels.
But innovation is a percentage increase in the current state of productivity. There, you are talking about a decline in the RATE of innovation -- but it still remains positive! Productivity doesn't decline as an absolute measure. Productivity continues to rise, indefinitely. Doesn't it?
A more positive way of describing your same scenario is: innovation continues forever (but at a slower rate), and per-capita wealth increases forever. Each individual life in the future is better off than the lives of the past. That story seems to be the "same" story you are telling, but it "sounds" much more optimistic.
Smith and many subsequent others have taught that division of labor, etc., etc. depends on the size of the market. Reduce the market size sufficiently by reducing the population and at some point productivity *will* slow. True that it would not slow initially with population decline for some period of time.
But note also given huge national debt and unfunded liabilities by essentially all modern governments, once GDP growth turns negative you got big problems as population declines even if productivity is increasing somewhat throughout the population decline.
The data behind one drives innovation seems to point towards not just causality from population, but intertwined growth of ideas. So falling population growth will detract from innovation because the number of intersection points of new Idea creation will reduce, and we will be doing more "maintenance" work. cf https://www.strangeloopcanon.com/p/innovation
I basically accept the framing of the situation. We are in a race to offset the coming population cliff (further accentuated by regulatory over-reach and technological catastrophe). On the other side we have scientific and technological innovation, artificial intelligence, education, improved spreading of knowledge, genetic tinkering and artificial? reproduction. Will we see a transition to a new, better form of innovation (not depending on fertility) before population goes over the cliff?
The future is impossible to predict, but I see it as an interesting race. I guess if forced to make a bet, I would side with progress. More of a gut feeling than logic though.
It seems possible to me that there are thermostatic effects between innovation and population growth. If someone told me that fertility increases as innovation slows I wouldn't laugh at them.
I'm new to this substack, which someone called to my attention after I'd expressed concern about dysgenic fertility in another forum focused on economic issues. Maybe I'm preaching to the choir here, but as I see it the demographic problem in the US and other first-world countries is not just low population fertility but an inverse correlation between child-bearing and native intelligence.
In their unjustly maligned book The Bell Curve, first published in 1994, Murray and Herrnstein noted that median lifetime childbearing in the US varied inversely with female educational attainment: only that of women who failed to graduate from high school was significantly above replacement level; it was approximately at replacement level (2.1) among women who went on to college but did not earn a degree; dipping below replacement level among those obtaining bachelor degrees; and lower still among those with postgrad or professional degrees.
It stands to reason that the size of a nation's population will have a bearing on its total economic output, but correlation with economic productivity is another matter. According to the World Bank, there are six nations in the world where median lifetime childbearing is 6 or more: Mali, the Central African Republic, the Democratic Republic of Congo, Chad, Somalia, and Niger. If high fertility is the key to innovation and productivity those nations should be at the top of the heap in terms of per capita GDP and low-fertility countries like the US, Singapore, and Norway should be at the bottom. But it's the other way around. Annual GDP per capita is $76,330 in the US, $82,808 in Singapore, and $106,177 in Norway -- well above the worldwide average of $12,744. In contrast, none of those six most-fertile nations has per-capita GDP anywhere near the worldwide average, let alone that in the US, Singapore, or Norway. That of Mali is only $833; of the CAR $427; of the Democratic Republic of Congo $654; of Chad $717; of Somalia $592; and of Niger $585.
You may choose to believe that the poor economic performance of those nations is a legacy of colonialism, racism, corporate capitalist exploitation, or the like, but FYI: according to a recent comparative study by Richard Lynn and David Becker median IQ in Mali is 59.76, in the Democratic Republic of Congo it's 64.8; and in Somalia it's 67.65. (Lynn and Becker did not include estimates for the CAR, Niger, or Chad.)
In light of the foregoing, I surmise that while massive migration into the US from third-world countries like Guatemala (where, as calculated by Lynn and Becker, median IQ is 47.72) may stimulate economic output, it will have a perverse effect on per-capita productivity, which is the true measure of prosperity.
I would love to see the economics equation that shows any relationship whatsoever between population headcount, economic growth and innovation. Heck they should be shipping people by their hundreds of millions into Silicon Valley to further increase its rates of innovation and growth.
The future calculation can be sharpened by removing the hypothetical D economy and focusing on how many more years of our current annual innovation we have left before the end of the universe if we follow the H economy path. The math becomes simpler. Assuming innovation is proportional to the number of human-years until the heat death of the universe, we will produce:
Human-years left = 2 * H + S, where S is the number of years before the population switches to decline.
If we have 30 years before the switch from growth to decline, and the population halving period is 60 years, then we have 150 years of this year's innovation left before we are all dead or Amish.
Here's my picture of aHere's my picture of an optimistic world with another 21 years worth of innovation at the rate that we have seen in recent decades. https://worldbuild.ai/hall-of-mirrors/
The AGI is moot because it is kept isolated, allowing a realistic world, e.g. one without AGI that soon. The contest called for optimism, so I tried to depict perhaps a 1 sigma better than expectation world, thus a world with perhaps half a sigma more than expectation technological growth. It seems reasonable in that case to treat it as 30 years worth of innovation in expectation. I don't discuss indefinitely extended fertility, (which the world was assumed to have) or artificial wombs (which it is assumed not to have), but I do discuss life extension (which it doesn't have much of and what it does have is less widespread than increased iatrogenic harms). Over-all, I would say that you are my largest intellectual influencer, and that our world-views are very close to one another, but that's for you to decide.
My main question is, do you really think we need more progress than is depicted to solve the fertility decline? In the model, I am assuming that China at least has solved it.n optimistic world with another 21 years worth of innovation at the rate that we have seen in recent decades. https://worldbuild.ai/hall-of-mirrors/
The AGI is moot because it is kept isolated, allowing a realistic world, e.g. one without AGI that soon. The contest called for optimism, so I tried to depict perhaps a 1 sigma better than expectation world, thus a world with perhaps half a sigma more than expectation technological growth. It seems reasonable in that case to treat it as 30 years worth of innovation in expectation. I don't discuss indefinitely extended fertility, (which the world was assumed to have) or artificial wombs (which it is assumed not to have), but I do discuss life extension (which it doesn't have much of and what it does have is less widespread than increased iatrogenic harms). Over-all, I would say that you are my largest intellectual influencer, and that our world-views are very close to one another, but that's for you to decide.
My main question is, do you really think we need more progress than is depicted to solve the fertility decline? In the model, I am assuming that China at least has solved it.
This analysis seems to rely a great deal on a model in which causality runs pretty strictly from population growth to innovation. But I read history more the other way around. Innovation enabled population growth. Going forward, my intuition is that if innovators were surrounded by much better institutional support, we would see more innovation even if world population were an order of magnitude below what it is today.
Sure, innovation per econ unit might increase, but nations have been trying a lot of things to induce that and mostly failing for a long time.
Yes, causality can go both ways. But will be hard for innovation to induce more fertility soon; the social forces arrayed against fertility are many and strong: https://www.overcomingbias.com/p/escalating-signals-cut-fertility
By "nations" you mean governments. Governments seem to have the ability suppress innovation, at least somewhat. They can also suppress fertility (1-child policy). I agree that the policies that have been enacted with the supposed intent of increasing innovation mostly have not worked. Same with policies with the supposed intent of increasing fertility. But that does not mean that there are no institutional changes that would have the effect of increasing innovation. I didn't say that governments are likely to enact policies that promote such changes, merely that I suspect that such changes exist in principle.
Oh sure such changes probably exist. But assuming no such changes seems reasonable for my purposes above.
Innovators are indeed surrounded by excellent institutional support. It's called Venture Capital. Please don't tell me that you were thinking of tax-funded state support.
I was thinking of less government involvement (e.g. regulation), not more
Consider the perspective of a randomly choosen person from all of human history (behind the veil of ignorance). Do they want to maximize innovation **per year**? Certainly not. They would want to maximize **Innovation per person year**.
You don't want to have more people existing at earlier, less productive, times because that makes the average human life less good. Since a smaller population both decreases innovation and person years that pass each year proportionally we should be indifferent to it.
To put it differently, If a demon caused literally everything in the universe to slow to half speed (according to his infernal clock) would that be bad because now it takes 2 years to increase productivity by 5% instead of one year? Surely not. What's relevant is the amount of human experience which occurs at each level of productivity not some arbitrary unit of time.
I'll write separately about why innovation is good.
Yes, obviously, other things being equal, more innovation is good but when you change the total population or population growth rate other things aren't equal.
My point is simply that intuitively attractive choices there give different verdicts about what scenarios are better than you assume and you really should specify what variable you want maximized and check that leads to plausible conclusions.
For instance, if you want to maximize something like expected consumption (or utility) for a randomly choosen member of the species over all of human history sometimes you prefer to spend less human lives to reach a given level of tech
If you want to maximize
1. Having more kids now does not mean less are born later, which your claim implies.
2. The worry here is an eventual end to innovation
Fewer.
1) No my claim doesn't imply less people will be born in the future in any way. In fact, such a situation is better not worse for my argument because an even greater fraction of all people are born at high productivity times.
2) The worry isn't that literally innovation ceases but that the total amount approaches a finite limit. While I don't think that's the calculation one should be doing, obviously even on my model you end up worse in the case where the population keeps halving.
But, of course, that's not even sorta realistic since you can't keep halving the population -- nor is it intended to be. But once you avoid that issue it just presses the point even more fiercely: if you know that eventually infinitely many people will be born in both scenarios do you care how quickly their per person production increases in years or maximizing the fraction of people who will be born who live at high productivity times.
You've highlighted some interesting issues, but it's a lot worse than people imagine.
1) The downward shift in productivity tends to stem from the shift from production tradables to service sector, although productivity gains in the former tend to be masked by significant product quality increases.
2) With the exception of the tech developments of the past 30 years, historically the analytical mode of thinking taught in universities actually decreases invention, innovation and tinkering, all of which tended to stem from the operational mode of thinking- smart pragmatic people taking real world problems and solving them in their garages and sheds. A more educated workforce inherently means either a loss of innovation or a narrowing of innovation into specific areas.
3) How much of economic growth in capitalism is dependent upon the growth of new markets and new products? In the past, labour reallocation through saturation plus productivity gain has always proven useful, because there have always been new products for labour utilisation. The largest single expansion occurred as a result of women's clothing. Is the rise of the service sector a signal of the end of new markets and new products? If so, it's a rocky road ahead- made all the more worrying by the current billionaire preoccupation with 'where is the best place to ride out an apocalyptic economic collapse'.
If your theory that innovation is proportional to population is correct, then it should be correct if you test it on past events. Here is a counterexample: After the black death, 1350-1450 Europe had lost 50% of its population. Yet what happened right after was the Renaissance. You would not call this innovation in terms of technology, but you can certainly characterize it as a massive mind-opening time. It is possible that the challenges of under-population will bring about something similar in the future.
That's enough years of innovation that I think fertility will be a moot point by then. Either we or our descendants will have radical life extension (if not medical immortality), or they'll have innovations that make it possible for societies to set their own population levels without the need to coax or coerce regular people into mating pairs (womb tanks, robot nannies, etc). Or possibly both.
I'm so confused. How did one of the most insightful thinkers of the last 30 years become so singularly reductionist on this idea? It's entirely possible I'm missing something as I would not assume I'm any more capable than you at thinking about these things (likely much less so).
But you seem so heavily indexed on the 1:1 association of fertility with productivity. Really? With all the advances in individual human productivity through technology and the invention of human+ level AGI in the next few decades, you don't see opportunities for productivity and innovation to develop elsewhere than just absolute number of biological humans?
Perhaps your insight and research has given you more clarity than me, but it also seems possible to me that you've made some early assumptions and then run with them to their deepest conclusions without clearly adjusting for the strength of your priors. What if absolute number of humans isn't the crucial driving force of innovation you assume? Maybe it is, I don't know. But I'm not convinced either way and as a result all these fertility pieces seem like deeply worrying thought pieces that are putting the horse before the cart.
The great innovators seem to have had a habit of seclusion, or a deliberate distancing from the madding crowd. OK so there was Shakespeare but...
You seem to be blurring different aggregate measures, to describe what seems to be a pessimistic scenario. Let's agree that world output is roughly proportional to population, for a given level of productivity. You are right that, as population declines, we should also expect a decline in total world output. That's a decline in absolute levels.
But innovation is a percentage increase in the current state of productivity. There, you are talking about a decline in the RATE of innovation -- but it still remains positive! Productivity doesn't decline as an absolute measure. Productivity continues to rise, indefinitely. Doesn't it?
A more positive way of describing your same scenario is: innovation continues forever (but at a slower rate), and per-capita wealth increases forever. Each individual life in the future is better off than the lives of the past. That story seems to be the "same" story you are telling, but it "sounds" much more optimistic.
Smith and many subsequent others have taught that division of labor, etc., etc. depends on the size of the market. Reduce the market size sufficiently by reducing the population and at some point productivity *will* slow. True that it would not slow initially with population decline for some period of time.
But note also given huge national debt and unfunded liabilities by essentially all modern governments, once GDP growth turns negative you got big problems as population declines even if productivity is increasing somewhat throughout the population decline.
The data behind one drives innovation seems to point towards not just causality from population, but intertwined growth of ideas. So falling population growth will detract from innovation because the number of intersection points of new Idea creation will reduce, and we will be doing more "maintenance" work. cf https://www.strangeloopcanon.com/p/innovation
I basically accept the framing of the situation. We are in a race to offset the coming population cliff (further accentuated by regulatory over-reach and technological catastrophe). On the other side we have scientific and technological innovation, artificial intelligence, education, improved spreading of knowledge, genetic tinkering and artificial? reproduction. Will we see a transition to a new, better form of innovation (not depending on fertility) before population goes over the cliff?
The future is impossible to predict, but I see it as an interesting race. I guess if forced to make a bet, I would side with progress. More of a gut feeling than logic though.
It seems possible to me that there are thermostatic effects between innovation and population growth. If someone told me that fertility increases as innovation slows I wouldn't laugh at them.
Thank goodness, I'm not sure how much more innovation we can stand.
I'm new to this substack, which someone called to my attention after I'd expressed concern about dysgenic fertility in another forum focused on economic issues. Maybe I'm preaching to the choir here, but as I see it the demographic problem in the US and other first-world countries is not just low population fertility but an inverse correlation between child-bearing and native intelligence.
In their unjustly maligned book The Bell Curve, first published in 1994, Murray and Herrnstein noted that median lifetime childbearing in the US varied inversely with female educational attainment: only that of women who failed to graduate from high school was significantly above replacement level; it was approximately at replacement level (2.1) among women who went on to college but did not earn a degree; dipping below replacement level among those obtaining bachelor degrees; and lower still among those with postgrad or professional degrees.
It stands to reason that the size of a nation's population will have a bearing on its total economic output, but correlation with economic productivity is another matter. According to the World Bank, there are six nations in the world where median lifetime childbearing is 6 or more: Mali, the Central African Republic, the Democratic Republic of Congo, Chad, Somalia, and Niger. If high fertility is the key to innovation and productivity those nations should be at the top of the heap in terms of per capita GDP and low-fertility countries like the US, Singapore, and Norway should be at the bottom. But it's the other way around. Annual GDP per capita is $76,330 in the US, $82,808 in Singapore, and $106,177 in Norway -- well above the worldwide average of $12,744. In contrast, none of those six most-fertile nations has per-capita GDP anywhere near the worldwide average, let alone that in the US, Singapore, or Norway. That of Mali is only $833; of the CAR $427; of the Democratic Republic of Congo $654; of Chad $717; of Somalia $592; and of Niger $585.
https://data.worldbank.org/indicator/SP.DYN.TFRT.IN?most_recent_value_desc=false
https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?most_recent_value_desc=false
You may choose to believe that the poor economic performance of those nations is a legacy of colonialism, racism, corporate capitalist exploitation, or the like, but FYI: according to a recent comparative study by Richard Lynn and David Becker median IQ in Mali is 59.76, in the Democratic Republic of Congo it's 64.8; and in Somalia it's 67.65. (Lynn and Becker did not include estimates for the CAR, Niger, or Chad.)
https://www.ulsterinstitute.org/ebook/THE%20INTELLIGENCE%20OF%20NATIONS%20-%20Richard%20Lynn,%20David%20Becker.pdf
In light of the foregoing, I surmise that while massive migration into the US from third-world countries like Guatemala (where, as calculated by Lynn and Becker, median IQ is 47.72) may stimulate economic output, it will have a perverse effect on per-capita productivity, which is the true measure of prosperity.
I would love to see the economics equation that shows any relationship whatsoever between population headcount, economic growth and innovation. Heck they should be shipping people by their hundreds of millions into Silicon Valley to further increase its rates of innovation and growth.
The future calculation can be sharpened by removing the hypothetical D economy and focusing on how many more years of our current annual innovation we have left before the end of the universe if we follow the H economy path. The math becomes simpler. Assuming innovation is proportional to the number of human-years until the heat death of the universe, we will produce:
Human-years left = 2 * H + S, where S is the number of years before the population switches to decline.
If we have 30 years before the switch from growth to decline, and the population halving period is 60 years, then we have 150 years of this year's innovation left before we are all dead or Amish.
Why assume that growth is exponential, instead of logistic (S-shaped)?
There is another factor to consider
Here's my picture of aHere's my picture of an optimistic world with another 21 years worth of innovation at the rate that we have seen in recent decades. https://worldbuild.ai/hall-of-mirrors/
The AGI is moot because it is kept isolated, allowing a realistic world, e.g. one without AGI that soon. The contest called for optimism, so I tried to depict perhaps a 1 sigma better than expectation world, thus a world with perhaps half a sigma more than expectation technological growth. It seems reasonable in that case to treat it as 30 years worth of innovation in expectation. I don't discuss indefinitely extended fertility, (which the world was assumed to have) or artificial wombs (which it is assumed not to have), but I do discuss life extension (which it doesn't have much of and what it does have is less widespread than increased iatrogenic harms). Over-all, I would say that you are my largest intellectual influencer, and that our world-views are very close to one another, but that's for you to decide.
My main question is, do you really think we need more progress than is depicted to solve the fertility decline? In the model, I am assuming that China at least has solved it.n optimistic world with another 21 years worth of innovation at the rate that we have seen in recent decades. https://worldbuild.ai/hall-of-mirrors/
The AGI is moot because it is kept isolated, allowing a realistic world, e.g. one without AGI that soon. The contest called for optimism, so I tried to depict perhaps a 1 sigma better than expectation world, thus a world with perhaps half a sigma more than expectation technological growth. It seems reasonable in that case to treat it as 30 years worth of innovation in expectation. I don't discuss indefinitely extended fertility, (which the world was assumed to have) or artificial wombs (which it is assumed not to have), but I do discuss life extension (which it doesn't have much of and what it does have is less widespread than increased iatrogenic harms). Over-all, I would say that you are my largest intellectual influencer, and that our world-views are very close to one another, but that's for you to decide.
My main question is, do you really think we need more progress than is depicted to solve the fertility decline? In the model, I am assuming that China at least has solved it.