That's reasonable enough, but it was coordination that drove the technology. It was the restrictions on coal use that led people to shift to alternate energy sources. It was that coordinated push that made bringing in natural gas and the infrastructure for it to make economic sense. If it was just about technology without coordination, we'd have seen better air filtration systems for homes and workplaces slowly accelerate, not fuel replacement and emissions control technology.
There's something naive about the way many economists address problems that involve collective action. Collective action requires political and legal mechanisms, and economists don't like that, so they insist that economic forces alone are sufficient.
A good example was the bicyclist problem. People liked bicycles and the mobility they provided, but the roads were awful, cobblestones and mud. There was no way making economic decisions about bicycle technology were going to improve the roads. The roads were improved by collective action imposing a political solution. The better roads made bicycles better and also - sigh - paved the way for automobiles.
You had this problem with electrical power. It was limited by the need to use batteries or having a local generator. It didn't make economic sense for a single factory to replace a steam powered prime mover with a generator, however, coordinated action to encourage and convince the public of the safety of electric power along public conduits made regional generation facilities practical.
Technology may provide a long term solution. (e.g. There was no way to implement congestion pricing without creating worse problems without recent technology.) However, coordinated action is frequently necessary to drive the technology.
My understanding is that you did not have this problem with electrical power. Factories did install generators and electrify long before there was a grid. The advantage of electricity was not efficiency, which might have required the scale of a grid, but the flexibility of having many independent motors. This required redesigning and rebuilding factories, which was slow, but did not require coordination.
Large factories could retool and new factories could be rebuilt around an onsite generating plant. I'm sure there were some factories built or rebuilt to take advantage of electrical power, but electrical power use in manufacturing didn't really take off until very late in the 19th century after power utilities made the decision easier. There were all sorts of barriers before then. If nothing else, you needed to hire an engineer who could keep the plant and the distribution system operating. The companies making machinery faced a who goes first situation. Why sell electric versions of one's machinery until there was enough demand with that demand being held back by the need for local generation and available equipment.
There were big payoffs from having independent motors and electrical control systems with things like thermostats, switch sensors, rotary timers and relay logic. Those had to be weighed against the cost and learning curve of operating one's own generator. Clearly, a lot of companies took the jump and did very well, but it didn't happen overnight.
Coal was competing with bottled gas, a specialty item suitable for residential and specialized industrial use. Coal had delivery infrastructure. Natural gas didn't.
Still, I'd be interested in seeing price comparisons from back in the day, before collective action started changing the incentives.
That is an interesting example. Definitely one where political coordination has failed and probably made things worse. And I do agree that technology has made things easier, but tech can't solve this one on its own since most roads are politically administered.
Your comment also calls to mins an anecdote I heard about Milton Friedman once. He was interested in congestion pricing but they didn't have automatic toll booths or GPS, so he recommended putting a radioactive chemical on toll roads and then testing people's tires for radioactivity to see how many miles they've driven!
But today we do have sensors that could register the proximity of one vehicle to other slow-moving vehicles. Parking meters could use dynamic pricing to ensure just less than 100% use.
London solved its truly apocalyptic late-90s traffic problem with congestion pricing, sinking the money raised into increasing buses. Last time I was there the only vehicles on the road were elite black cars, taxis, and red buses.
This was *definitely* not a technological solution.
Perhaps you are correct, but that is not what computer-based traffic simulations show. Human drivers behave fundamentally differently than other moving objects.
When cars driven by humans get close to capacity on a highway, the rate of travel decreases dramatically. That is why you can suddenly go from 60mph to stop-and-go.
Other objects like water keep moving at the same speed. Traffic simulations show self-driving cars keep moving at the same speed even when capacity has been reached. Self-driving cars can also drive much closer to each other so the capacity of the road increases dramatically.
Fortunately, it only takes a relatively small percentage of self-driving cars to make a huge difference in speed of through-put. So the current capacity of roads is not relevant.
In a world where late-abiding self-driving cars predominate, I'd feel safe to ride my bike at a sedate 20kmh in the middle of the lane, and so would many others. When on foot I'd feel safe to cross the road wherever I like, as leisurely as I want.
This would certainly increase the transport capacity of the road, but not for people in cars.
"The story continues that if only we could coordinate, we could fix the misallocation caused by externalities. We might get the government to tax and subsidize externalities"
Well, yes, but.....
The point of the tax - to make just the one point - is so that people *recognise* those external costs by being forced to include them in market prices. This then turns attention to, forces even, the design of technologies to alleviate these costs.
The taxation of externalities - Pigou Taxes - is not in opposition to technology solves externalities. It's a helping along, forcing, of technology solves externalities. By making them clear in market prices therefore effort is devoted to etc.....
A carbon tax doesn't tax CO2-e emissions because people are naughty boys. A carbon tax taxes CO2-e emissions to encourage the technological advances which reduce CO2-e emissions by making those costs, thus possible savings, clear in market prices....
I don't understand the implied trade-off of technology v incentives. A tax on coal pollution would have made the change to gas more lucrative sooner. Incentive are a spur to technology
No implied tradeoff, just observing that technology is far more important for solving externalities compared to coordination than is let on in economics courses.
I agree. I would even go further by saying that every solution that works involves an application of technology. Without technology, there are no solutions.
What's missing here is that technology didn't solve any of these problems, it provided a solution. The solution was via the use of technology in a particular way. It seems natural looking back at history that it was used in the way it was, but it wasn't always so natural. Some of these cases, I doubt that simply availability and deployment of the technology would have reached the necessary endpoint alone. Whaling is the first example here. Yes, coordination would have been very difficult if there weren't alternatives to whale oil, but I also think the whaling industry's remaining momentum would have continued until extinction without the coordination efforts.
Technology can often look like the only cause because it has a clear turning point, where political coordination is more progressive. Yes, there's the turning point of a law being passed, but in many cases the real change is opinion, which both motivates what is researched, how research is used (or not used), how quickly it's used, and how the remainder and laggards are dealt with.
I will go contra you on this long form later, because I believe technology (historically) shifts instead of solving, this ultimately (I) cumulatively leading to much worse externalities, EG nuclear weapons, Zuboffian problems, and robot militaries; (II) therefore accumulating unprecedented & insurmountable conceraction & solidification of power that; (III) leads probably to mass humans suffering & death; (IV) that technology cannot solve. This is because the universe historically happened to evolve without letting us to adapt sufficiently to new technology socially or otherwise developed resistance, because there happened to not to be single all-powerful-empire that would have made serious attempts to globally control technology at all costs, and because power vacuums must always be filled. There's no one to blame for this.
very nice piece. one cavil: "Leech" refers to a blood-sucking worm or a person who exploits others, while "leach" means to drain or remove something gradually, often through a liquid.
I disagree with the conclusion here. Technology didn't actually "solve" any externality. When someone talks about solving an extremely, they mean the externality is internalized, not that the effect goes away. Technology does not and can not internalize an externality, it can only charge how much externalities doing something has. This is an important distinction, because new technologies can just as easily have more externalities than the technology it replaces as it can less. Without an actual correction of the externalities, there is no reason to expect technology to reduce externalities on average.
Also, Malthus's theory has been thoroughly discredited. Its quite clear in modern times that we don't always live at subsistence. I would guess your chart has more to do with land values than malthusian population growth. Ie more productive land costs more and land lords can charge rent that takes a substantial amount of people's working product
Good essay pointing out the role of tech in improving our social problems: important but not always the final or best solution, but more important than often realized. Thanks to emphasizing that insight. And a hat tip to Dr. Magoon for bringing this subject into the progress related discussions.
Some comments below about economists needing to factor in technology in their analyses and thinking: it is a shame they went down the math intensive path and wandered away from "political economy" and the messy social and political aspects of their discipline. Millions of man hours spent on potentially valid ideas but still woefully incomplete.... and therefore likely incorrect.
"Our firefighters are barely ever called to actual fires. "
That suggests that when we see fire trucks accompanying an ambulance to a site, the fire trucks are sometimes really just using the "emergency" as a rationale to provide another training exercise?
Then again, when you want that talent and training, say to fight forest fires, it is quite essential to have such a trained cadre available, with supporting equipment and infrastructure.
I am a consultant who specializes in identifying (or ruling out!) authentic demand, for clients with startups or new products.
In general, the hardest part of driving innovation is finding and removing the blockers that prevent people from making progress on a goal important to them. These blockers are highly-context specific and are not obvious, since if they were the blocker would have been solved already. Technology often plays a role in innovation adoption, but usually in context and amplified by some other factor. The problem with historical examples is that in my experience, a deep dive shows far stronger contextual reasons for the adoption or change. We simply forget.
The United States sent 1,000,000 horses to France during WWI; 200 of these were returned. The animals were disproportionately work and carriage horses, capable of pulling equipment, and it turned out that they were better-than-average warhorses for troops. The loss of so many horses decimated breeds and breeding programs after the war; it's likely customers didn't want a faster horse, but *any* horse. Ford's cheap Model T hit authentic demand that was not being supplied elsewhere. In 1920, a carriage horse cost $100-200. A Model T cost $395. (It had also already been on the market for 12 years, don't forget.)
I like the suggestion that "Economics should emphasize the importance of technology..." One of the definitions of economics that I learned was something like, "economics is the study of how limited resources are distributed in society". This definition, at least on the surface, points to social coordination issues. But I think you and the progress studies movement in general are right to think that technology, by increasing the supply of resources, can obviate those problems; and therefore that it would be very good for economics as a discipline to study technology more intensively.
That's on the academic level. On the social/political level, I feel like "technology will solve it, just stop whining" is sometimes a tool for blocking progress; in particular, I think social equality is a very big positive factor in technological advancement, and social equality is inherently a social coordination question. So it's not like social issues should be ignored.
But yeah, in terms of how betterment happens, more understanding of how technology evolves would always be a good thing.
That's reasonable enough, but it was coordination that drove the technology. It was the restrictions on coal use that led people to shift to alternate energy sources. It was that coordinated push that made bringing in natural gas and the infrastructure for it to make economic sense. If it was just about technology without coordination, we'd have seen better air filtration systems for homes and workplaces slowly accelerate, not fuel replacement and emissions control technology.
There's something naive about the way many economists address problems that involve collective action. Collective action requires political and legal mechanisms, and economists don't like that, so they insist that economic forces alone are sufficient.
A good example was the bicyclist problem. People liked bicycles and the mobility they provided, but the roads were awful, cobblestones and mud. There was no way making economic decisions about bicycle technology were going to improve the roads. The roads were improved by collective action imposing a political solution. The better roads made bicycles better and also - sigh - paved the way for automobiles.
You had this problem with electrical power. It was limited by the need to use batteries or having a local generator. It didn't make economic sense for a single factory to replace a steam powered prime mover with a generator, however, coordinated action to encourage and convince the public of the safety of electric power along public conduits made regional generation facilities practical.
Technology may provide a long term solution. (e.g. There was no way to implement congestion pricing without creating worse problems without recent technology.) However, coordinated action is frequently necessary to drive the technology.
My understanding is that you did not have this problem with electrical power. Factories did install generators and electrify long before there was a grid. The advantage of electricity was not efficiency, which might have required the scale of a grid, but the flexibility of having many independent motors. This required redesigning and rebuilding factories, which was slow, but did not require coordination.
You are right.
Large factories could retool and new factories could be rebuilt around an onsite generating plant. I'm sure there were some factories built or rebuilt to take advantage of electrical power, but electrical power use in manufacturing didn't really take off until very late in the 19th century after power utilities made the decision easier. There were all sorts of barriers before then. If nothing else, you needed to hire an engineer who could keep the plant and the distribution system operating. The companies making machinery faced a who goes first situation. Why sell electric versions of one's machinery until there was enough demand with that demand being held back by the need for local generation and available equipment.
There were big payoffs from having independent motors and electrical control systems with things like thermostats, switch sensors, rotary timers and relay logic. Those had to be weighed against the cost and learning curve of operating one's own generator. Clearly, a lot of companies took the jump and did very well, but it didn't happen overnight.
> It was the restrictions on coal use that led people to shift to alternate energy sources.
No, it was the lower price and greater cleanliness.
Coal was competing with bottled gas, a specialty item suitable for residential and specialized industrial use. Coal had delivery infrastructure. Natural gas didn't.
Still, I'd be interested in seeing price comparisons from back in the day, before collective action started changing the incentives.
Yes, as someone is fond of pointing out, incentives matter.
What about congestion? Seems to me congestion pricing is only solution here and technology actually enables us to do a good job.
That is an interesting example. Definitely one where political coordination has failed and probably made things worse. And I do agree that technology has made things easier, but tech can't solve this one on its own since most roads are politically administered.
Your comment also calls to mins an anecdote I heard about Milton Friedman once. He was interested in congestion pricing but they didn't have automatic toll booths or GPS, so he recommended putting a radioactive chemical on toll roads and then testing people's tires for radioactivity to see how many miles they've driven!
But today we do have sensors that could register the proximity of one vehicle to other slow-moving vehicles. Parking meters could use dynamic pricing to ensure just less than 100% use.
Is there any city in the world that is actually trying this?
London solved its truly apocalyptic late-90s traffic problem with congestion pricing, sinking the money raised into increasing buses. Last time I was there the only vehicles on the road were elite black cars, taxis, and red buses.
This was *definitely* not a technological solution.
My guess is that the technological innovation of self-driving cars will be the most effective solution to that problem.
It won't be, because of induced demand.
Not likely. Traffic congestion is caused by lack of coordination among cars, not by the number of cars.
Nope, better coordination may increase road capacity, but the capacity is still limited and there will be traffic if it is exceeded.
Perhaps you are correct, but that is not what computer-based traffic simulations show. Human drivers behave fundamentally differently than other moving objects.
When cars driven by humans get close to capacity on a highway, the rate of travel decreases dramatically. That is why you can suddenly go from 60mph to stop-and-go.
Other objects like water keep moving at the same speed. Traffic simulations show self-driving cars keep moving at the same speed even when capacity has been reached. Self-driving cars can also drive much closer to each other so the capacity of the road increases dramatically.
Fortunately, it only takes a relatively small percentage of self-driving cars to make a huge difference in speed of through-put. So the current capacity of roads is not relevant.
Has San Francisco witnessed a change now that there are waymos everywhere?
In a world where late-abiding self-driving cars predominate, I'd feel safe to ride my bike at a sedate 20kmh in the middle of the lane, and so would many others. When on foot I'd feel safe to cross the road wherever I like, as leisurely as I want.
This would certainly increase the transport capacity of the road, but not for people in cars.
Maybe simply better internet reducing the need for so many people to go to the city.
"The story continues that if only we could coordinate, we could fix the misallocation caused by externalities. We might get the government to tax and subsidize externalities"
Well, yes, but.....
The point of the tax - to make just the one point - is so that people *recognise* those external costs by being forced to include them in market prices. This then turns attention to, forces even, the design of technologies to alleviate these costs.
The taxation of externalities - Pigou Taxes - is not in opposition to technology solves externalities. It's a helping along, forcing, of technology solves externalities. By making them clear in market prices therefore effort is devoted to etc.....
A carbon tax doesn't tax CO2-e emissions because people are naughty boys. A carbon tax taxes CO2-e emissions to encourage the technological advances which reduce CO2-e emissions by making those costs, thus possible savings, clear in market prices....
This is one of my favorite substack blogs ive ever read. Its so simple and yet seems to address some of the most pressing problems throughout history
I don't understand the implied trade-off of technology v incentives. A tax on coal pollution would have made the change to gas more lucrative sooner. Incentive are a spur to technology
No implied tradeoff, just observing that technology is far more important for solving externalities compared to coordination than is let on in economics courses.
I agree. I would even go further by saying that every solution that works involves an application of technology. Without technology, there are no solutions.
excellent post!
What's missing here is that technology didn't solve any of these problems, it provided a solution. The solution was via the use of technology in a particular way. It seems natural looking back at history that it was used in the way it was, but it wasn't always so natural. Some of these cases, I doubt that simply availability and deployment of the technology would have reached the necessary endpoint alone. Whaling is the first example here. Yes, coordination would have been very difficult if there weren't alternatives to whale oil, but I also think the whaling industry's remaining momentum would have continued until extinction without the coordination efforts.
Technology can often look like the only cause because it has a clear turning point, where political coordination is more progressive. Yes, there's the turning point of a law being passed, but in many cases the real change is opinion, which both motivates what is researched, how research is used (or not used), how quickly it's used, and how the remainder and laggards are dealt with.
I will go contra you on this long form later, because I believe technology (historically) shifts instead of solving, this ultimately (I) cumulatively leading to much worse externalities, EG nuclear weapons, Zuboffian problems, and robot militaries; (II) therefore accumulating unprecedented & insurmountable conceraction & solidification of power that; (III) leads probably to mass humans suffering & death; (IV) that technology cannot solve. This is because the universe historically happened to evolve without letting us to adapt sufficiently to new technology socially or otherwise developed resistance, because there happened to not to be single all-powerful-empire that would have made serious attempts to globally control technology at all costs, and because power vacuums must always be filled. There's no one to blame for this.
very nice piece. one cavil: "Leech" refers to a blood-sucking worm or a person who exploits others, while "leach" means to drain or remove something gradually, often through a liquid.
I disagree with the conclusion here. Technology didn't actually "solve" any externality. When someone talks about solving an extremely, they mean the externality is internalized, not that the effect goes away. Technology does not and can not internalize an externality, it can only charge how much externalities doing something has. This is an important distinction, because new technologies can just as easily have more externalities than the technology it replaces as it can less. Without an actual correction of the externalities, there is no reason to expect technology to reduce externalities on average.
Also, Malthus's theory has been thoroughly discredited. Its quite clear in modern times that we don't always live at subsistence. I would guess your chart has more to do with land values than malthusian population growth. Ie more productive land costs more and land lords can charge rent that takes a substantial amount of people's working product
Vaccines, at least with the rollout of modRNA went from bad to worse. That said perhaps other goals were in mind like as viewed in Baliwick News
Good essay pointing out the role of tech in improving our social problems: important but not always the final or best solution, but more important than often realized. Thanks to emphasizing that insight. And a hat tip to Dr. Magoon for bringing this subject into the progress related discussions.
Some comments below about economists needing to factor in technology in their analyses and thinking: it is a shame they went down the math intensive path and wandered away from "political economy" and the messy social and political aspects of their discipline. Millions of man hours spent on potentially valid ideas but still woefully incomplete.... and therefore likely incorrect.
"Our firefighters are barely ever called to actual fires. "
That suggests that when we see fire trucks accompanying an ambulance to a site, the fire trucks are sometimes really just using the "emergency" as a rationale to provide another training exercise?
Then again, when you want that talent and training, say to fight forest fires, it is quite essential to have such a trained cadre available, with supporting equipment and infrastructure.
I am a consultant who specializes in identifying (or ruling out!) authentic demand, for clients with startups or new products.
In general, the hardest part of driving innovation is finding and removing the blockers that prevent people from making progress on a goal important to them. These blockers are highly-context specific and are not obvious, since if they were the blocker would have been solved already. Technology often plays a role in innovation adoption, but usually in context and amplified by some other factor. The problem with historical examples is that in my experience, a deep dive shows far stronger contextual reasons for the adoption or change. We simply forget.
The United States sent 1,000,000 horses to France during WWI; 200 of these were returned. The animals were disproportionately work and carriage horses, capable of pulling equipment, and it turned out that they were better-than-average warhorses for troops. The loss of so many horses decimated breeds and breeding programs after the war; it's likely customers didn't want a faster horse, but *any* horse. Ford's cheap Model T hit authentic demand that was not being supplied elsewhere. In 1920, a carriage horse cost $100-200. A Model T cost $395. (It had also already been on the market for 12 years, don't forget.)
I enjoyed this article but I disagree. I would love to chat more if you have any time and interest: https://bsky.app/profile/andrewsudmant.bsky.social/post/3lkseyuuscc27
I like the suggestion that "Economics should emphasize the importance of technology..." One of the definitions of economics that I learned was something like, "economics is the study of how limited resources are distributed in society". This definition, at least on the surface, points to social coordination issues. But I think you and the progress studies movement in general are right to think that technology, by increasing the supply of resources, can obviate those problems; and therefore that it would be very good for economics as a discipline to study technology more intensively.
That's on the academic level. On the social/political level, I feel like "technology will solve it, just stop whining" is sometimes a tool for blocking progress; in particular, I think social equality is a very big positive factor in technological advancement, and social equality is inherently a social coordination question. So it's not like social issues should be ignored.
But yeah, in terms of how betterment happens, more understanding of how technology evolves would always be a good thing.