I knew what this graph was before I opened it. Indeed, this graph is also featured at the opening of Pathways of Progress: https://www.lianeon.org/p/starthere
What is interesting, however, is that if we chart GDP doubling time, we can see a steady declining trend until about 1970. That is, GDP growth throughout human history kept getting faster and faster until that year. We are still growing, just no longer accelerating.
I have an essay examining this. Some surmise that perhaps the “singularity” was cancelled. My guess, however, is that GDP is becoming less useful as a measurement index as the economy dematerializes and progress becomes more intangible.
Thank you! I wonder a lot about the 70s - I’d like to do a whole post on what happened there. I suspect it has more to do with the rising costs of energy, but I don’t now.
Avi Collis has some great research on the limits of GDP, particularly around digital goods (like substack!): https://www.avinash.info/research. I do think it’s becoming less useful, especially given the declining marginal utility of wealth.
Great piece! Its a great reminder how good things are. But I’m also on board with your cautions, maybe a little more in the “past performance does not guarantee future results” camp
We are due for a radical change because there does not appear to be a way exponential growth can continue indefinitely. The planet has limits. And as far as I can tell, we are approaching those limits, headed back towards a type of Malthusian population dynamics. One telling statistic is that 90% of land mammal biomass is either human or human livestock
I’m guessing if you could zoom into that graph, that prior to the industrial revolution you’d see more linear GDP gains. Its only since humans unleashed the power of combustion that we’ve enjoyed unrestricted population growth and therefore exponential GDP growth. But it can’t last forever unless we somehow master space travel very quickly.
I tend to think that the growth rate is unlikely to continue at this pace. Robert Gordon is really good on this point, and I plan to do a post on him in the not-to-distant future. But, I'm optimistic that we will continue to grow, just not at the same pace. You're right that it will require a much more efficient use of resources though.
I hope so. But I’m getting more and more pessimistic, thinking we’re heading towards a global Malthusian crisis that will be resolved by world war within a few decades.
Seems to me that a “soft landing” could be accomplished whereby population growth stabilizes and starts to decline while average GDP continues to increase at the healthy clip we’ve become accustomed. This can be accomplished via government incentives and education, especially for girls/women in developing regions.
Alas, the ‘rate of economic growth’ has not been increasing for many decades… And productivity has been declining for the last 2. The long tail of the scientific revolution might not be the gift that keeps giving… and AI might only briefly keep the illusion alive…
Economic growth is usually defined as increasing GDP per capita. The fact that China had the largest GDP in 1800 was because 300 million mostly impoverished peasants could produce more output than Britain's 10 million people, who were not yet 30 times more productive than the Chinese. But fifty years later British GDP was about 1/4 of the world total and dwarfed that of China.
China's population in 1800 had risen about five-fold from Ancient and early Medieval times. During the late Medieval period it rose to about 100 million and to about 140 million by 1600, roughly doubling again over the next two centuries in 300 in 1800 and to 430 million 1850. All of this before any industrialization.
Thanks for sharing this link - this is helpful. That said, I'm not sure I see the point of saying "I don't need the evidence of economic output when I know that population can be modelled by an exponential graph." The authors' theoretical insight - population has grown exponentially - is only interesting because it matches empirical outcome. The theory is interesting because it explains the result, not vice versa.
I do agree that population and economic output are going to be positively correlated to some extent. But, to say that makes the result "useless" misses the point. The fact that the world produced virtually nothing for thousands of years, and then, starting the industrial revolution, started to grow, enabling population growth, is to me one of the most interesting facts in the world. It wasn't like someone in 0BC was sitting around saying "just give it 2000 years, then we'll be driving cars, eating fast food, and taking penicillin."
Even if we abstract away from output, and just use the graph on page 9 of the PDF, above, it's still fascinating (and shocking) how much growth we have, and when it occurred. I prefer the economic one, as I think it suggests something more causal, but I could see the argument here.
Your point on the limitations of cross-country comparisons of GDP though is absolutely correct, and often lost. Still, this is world economic output, so I think there's something there.
I was critiquing the graph has providing no information other than the stunning imagery to laypersons. If a logarithmic scale were used it would show more clearly how growth has accelerated. And a second plot of per capita GDP on the right axis would give a complete picture of the explosion of progress over time.
For example, the per capita plot would show the explosion after 1870 due to the spread of the industrial revolution out of Europe and the US which J Bradford Delong writes about. The overall plot would show the effects of the pre-industrial development that as I indicated in my citation, was substantial.
I knew what this graph was before I opened it. Indeed, this graph is also featured at the opening of Pathways of Progress: https://www.lianeon.org/p/starthere
What is interesting, however, is that if we chart GDP doubling time, we can see a steady declining trend until about 1970. That is, GDP growth throughout human history kept getting faster and faster until that year. We are still growing, just no longer accelerating.
I have an essay examining this. Some surmise that perhaps the “singularity” was cancelled. My guess, however, is that GDP is becoming less useful as a measurement index as the economy dematerializes and progress becomes more intangible.
Thank you! I wonder a lot about the 70s - I’d like to do a whole post on what happened there. I suspect it has more to do with the rising costs of energy, but I don’t now.
Avi Collis has some great research on the limits of GDP, particularly around digital goods (like substack!): https://www.avinash.info/research. I do think it’s becoming less useful, especially given the declining marginal utility of wealth.
That is the leading theory, energy constraints. Mainly, the stifling of nuclear power.
Great piece! Its a great reminder how good things are. But I’m also on board with your cautions, maybe a little more in the “past performance does not guarantee future results” camp
We are due for a radical change because there does not appear to be a way exponential growth can continue indefinitely. The planet has limits. And as far as I can tell, we are approaching those limits, headed back towards a type of Malthusian population dynamics. One telling statistic is that 90% of land mammal biomass is either human or human livestock
I’m guessing if you could zoom into that graph, that prior to the industrial revolution you’d see more linear GDP gains. Its only since humans unleashed the power of combustion that we’ve enjoyed unrestricted population growth and therefore exponential GDP growth. But it can’t last forever unless we somehow master space travel very quickly.
I tend to think that the growth rate is unlikely to continue at this pace. Robert Gordon is really good on this point, and I plan to do a post on him in the not-to-distant future. But, I'm optimistic that we will continue to grow, just not at the same pace. You're right that it will require a much more efficient use of resources though.
I hope so. But I’m getting more and more pessimistic, thinking we’re heading towards a global Malthusian crisis that will be resolved by world war within a few decades.
Seems to me that a “soft landing” could be accomplished whereby population growth stabilizes and starts to decline while average GDP continues to increase at the healthy clip we’ve become accustomed. This can be accomplished via government incentives and education, especially for girls/women in developing regions.
Absolutely fantastic piece..!
Alas, the ‘rate of economic growth’ has not been increasing for many decades… And productivity has been declining for the last 2. The long tail of the scientific revolution might not be the gift that keeps giving… and AI might only briefly keep the illusion alive…
Very cool read. The jump in worldwide progress recently is pretty remarkable.
I hate that graph, it near worthless it would look the same even if they had been economic growth and GDP simply reflected population growth.
But that’s what makes the story so great! Population growth wasn’t possible before economic growth! Malthus was right, but only for a time. Also,
Economic growth is usually defined as increasing GDP per capita. The fact that China had the largest GDP in 1800 was because 300 million mostly impoverished peasants could produce more output than Britain's 10 million people, who were not yet 30 times more productive than the Chinese. But fifty years later British GDP was about 1/4 of the world total and dwarfed that of China.
China's population in 1800 had risen about five-fold from Ancient and early Medieval times. During the late Medieval period it rose to about 100 million and to about 140 million by 1600, roughly doubling again over the next two centuries in 300 in 1800 and to 430 million 1850. All of this before any industrialization.
https://www.academia.edu/39054520/Introduction_to_Social_Macrodynamics_Part_2_Secular_Cycles_and_Millennial_Trends?rhid=29171717401&swp=rr-rw-wc-35658524
Thanks for sharing this link - this is helpful. That said, I'm not sure I see the point of saying "I don't need the evidence of economic output when I know that population can be modelled by an exponential graph." The authors' theoretical insight - population has grown exponentially - is only interesting because it matches empirical outcome. The theory is interesting because it explains the result, not vice versa.
I do agree that population and economic output are going to be positively correlated to some extent. But, to say that makes the result "useless" misses the point. The fact that the world produced virtually nothing for thousands of years, and then, starting the industrial revolution, started to grow, enabling population growth, is to me one of the most interesting facts in the world. It wasn't like someone in 0BC was sitting around saying "just give it 2000 years, then we'll be driving cars, eating fast food, and taking penicillin."
Even if we abstract away from output, and just use the graph on page 9 of the PDF, above, it's still fascinating (and shocking) how much growth we have, and when it occurred. I prefer the economic one, as I think it suggests something more causal, but I could see the argument here.
Your point on the limitations of cross-country comparisons of GDP though is absolutely correct, and often lost. Still, this is world economic output, so I think there's something there.
I was critiquing the graph has providing no information other than the stunning imagery to laypersons. If a logarithmic scale were used it would show more clearly how growth has accelerated. And a second plot of per capita GDP on the right axis would give a complete picture of the explosion of progress over time.
For example, the per capita plot would show the explosion after 1870 due to the spread of the industrial revolution out of Europe and the US which J Bradford Delong writes about. The overall plot would show the effects of the pre-industrial development that as I indicated in my citation, was substantial.
Using a linear axis masks all those cool details.