(As a side note, the fact that France’s energy costs have gone up as much as the UK’s, despite France mostly running on nuclear power, should serve as a wake-up call for people who think nuclear power can solve the UK’s problems.)
This isn't true. The cost of energy is determined by the cost of the *marginal* kWh. At the time scales we are talking about, the marginal kWh is nearly always from gas, because it's the only source of energy that can be practically scaled up within minutes to years. Nuclear takes decades to build, and hydrothermal, wind or solar aren't much better, and none of them can increase power within minutes or hours, so they provide mostly inframarginal kWh. Even if your country's solar is producing power for $0.15/kWh, businesses and homes are paying $0.40/kWh because that's the price that the gas plant operators are charging to put out the last, marginal kWh the grid needs.
Nuclear can make a difference to energy costs, but only on much longer timescales. This is also why energy costs are so high even in many places which are heavily adopting solar. The marginal kWh is still from gas. As long as this is so, Europe will be saddled with high energy costs until they find a different source of gas than Russia, or they move to a different source of marginal kWh, e.g. grid-scale batteries charged by solar.
"Nuclear takes decades to build, and hydrothermal, wind or solar aren't much better, and none of them can increase power within minutes or hours"
The answer to the marginal kWh is storage. At present China is deploying all-in BESS systems for $66/kWh [1]. This is in China, but these systems are mostly deployed on shipping containers that are easy to assemble into full plants. It is very hard to imagine a world where this does not become the dominant factor in determining marginal energy costs, and that world is coming very quickly. (Much more quickly than any new nuclear or hydro can even make it through planning.) I keep seeing folks having forward-looking discussions on energy provisioning that just pretend sub-$100/kWh battery storage isn't a thing. At what all-in storage cost do we all agree that the world has actually changed?
I was thinking this is a feature of pricing in the energy market. Shouldn’t this mean massive profits for French nuclear plant owners and also renewables everywhere across Europe? Do we see this? I feel like I haven’t seen much of that in the news.
When we talk about "the price of electricity", we're usually talking about the spot price. The "average price of electricity" is usually the average spot price. The spot price is determined by the supply and demand at the current moment, and can even become negative if there is more electricity supply than demand. It is the marginal supplier (and demander) that largely determines the spot price.
Not all electricity users buy at the spot price. You can have longer term contracts between producers or suppliers, there are wholesale and retail electricity markets, forwards and other derivatives, etc. The structure of the electricity market is local and complex, and fortunes are made or lost trading in them (cf. Enron). Different users end up paying different prices to different providers, after different middlemen mediate between them, depending on how it all shakes out. However, when we talk about "the price of electricity" we aren't usually talking about all this complexity. We're just talking about the spot price.
Yeah all that extra complexity does make it hard to ‘follow the money’. So your original comment was basically explaining why the chart Noah included doesn’t say much about the cost of energy for retail or industrial consumers in UK vs in France.
"Eventually, as more investors learned to be cool and rational like Buffett"
Complete speculation on my part, but I don't think that anyone is any more cooler and rational than they were in the 60s/70s. My thesis is that technology improvements and regulatory changes have led to cheaper price discovery => more efficient markets.
Couldn't France's price surge be because they're all connected to the same energy grid? Which would be solved if all countries on said grid (such as Germany and the UK) stopped their ongoing energy-seppuku?
The Tech job bust is interesting - I would bet that there is still a lot of demand for tech roles outside of the IT sector. My hope is that deploying all those tech workers to other industries will help push them to improve their productivity...
I think that any model of a political economy that ignores political power is just an invalid model. There is a reason that many Democratic reforms have been stymied in the past, and there's a reason that Trump has the political freedom to propose price controls right now (irrespective of whether those are a good idea or not.) To a very great extent this is because corporate concentration makes these organizations (1) extremely politically powerful in Congress, but also (2) extremely vulnerable to extra-legal interventions by a vindictive executive branch, which means in turn that (3) they will deploy their power in ways that support the political goals of a vindictive executive, while blocking those of a more reasonable political party.
Democrats can try to address the power imbalances, or they can learn to govern like Trump. But there is no stable political world in which one political party has the political freedom to impose sweeping price controls, and the other party is unable to do anything similar. To the extent that "centrist" beat down "leftists" and prevent them from addressing concentration of wealth and power, you're just ensuring the only option is Trump-style executive governance, whether from the right or left.
Trump is still an idiot, but would it make sense to insist that drug manufacturers don't charge higher prices in the United States than they do in, say, France, Japan, or any other country with a GDP per capita of over $30,000? (With one of the goals being *higher* drug prices in other countries, so that they pay their fair share of the R&D expenses.)
It’s time to face up to the fact that the Abundance program can only be achieved if the Democrats are in power. DOGE, Trump, and the entire Republican party are staging a full scale attack on state capacity, a vital part of the Abundance agenda. Democrats are generally fine with highly competent technical state employees. The annoying tendency of Dems to attack corporate power is best overlooked-consider it a necessary pushback against capitalist Bonopartism that we are experiencing now from Musk, Thiel, and others. I hope the Abundance program aim isn’t to trade democracy for Abundance.
I admit, I'm siloed enough that I have not seen this very reasonable critique of force-lowering drug costs before. While it's pretty convincing, I'm curious if anyone can explain what the alternative would be. Having to spend a huge proportion of your income on medicine is, to me, simply morally unacceptable in a rich society--is there a way to correct that which doesn't have the negative results Noah outlined here?
Most countries regulate the bulk purchasing of drugs. The US is the outlier. It’s reasonable to assert that US profits were buttressing the R&D that everyone else benefited from.
“Limiting these companies’ market size by limiting the prices they can charge will result in less research spending, and ultimately fewer new medicines and treatments.”
My question is what other incentives and mechanisms can be put into place to further drug innovation. More public R&D seems like a no-brainer.
This is another idea:
“Since reforming its system in the early 2010s, Germany has allowed manufacturers to freely set prices for a limited period when bringing new drugs to the market. It then uses the data available from that period for a nongovernmental and nonprofit research body to evaluate the benefit provided by the new drug, as compared to existing alternatives. This added benefit, or lack thereof, then serves as the foundation for price negotiations between drug manufacturers and health plans.”
“My question is what other incentives and mechanisms can be put into place to further drug innovation. More public R&D seems like a no-brainer. “
While probably a decent idea, I think the real expensive parts of drug development are actually finding the drugs that pan out once targets and research have been identified and doing the efficacy and safety trials. I doubt the public sector is going to finance all that.
Pharma's argument that they needed to sell drugs for high prices in America to fund research was always flawed. Most of the breakthroughs came through government funded research that was basically gifted to pharma.
It has been true for a long time that most medical research in the US is funded by the government through the National Institutes of Health (NIH), which is the largest public funder of biomedical research in the world.
But, trump just cut billions from NIH. Trump's price controls won't hurt innovation as much as his cuts to NIH will.
"Pharma's argument that they needed to sell drugs for high prices in America to fund research was always flawed. Most of the breakthroughs came through government funded research that was basically gifted to pharma."
The basic breakthroughs and target may come from basic research, but someone still needs to find the compounds, get everything working, get it through trials and regulatory clearances which is where the bulk of the expense is.
I'm certain that a comprehensive R&D plan could help with this. Unfortunately we're pairing this policy with a massive reduction in NIH funding, so it's hard to see good outcomes here. (Unfortunately they will be short-term *popular* outcomes, and the costs will be born by the next administration.)
As price controls go, it seems like drug prices are up there as an area where it may actually be a good idea, and you can't just resort to the Econ 101 counterargument. This is not exactly an area driven by market forces.
Not only do patents mean that drug makers have legal monopolies on many drugs, but even off-patent drugs are often effective monopolies due to the high capital costs and low marginal costs associated with producing many drugs, making market entry a risky prospect. This also means that, while price controls may drive down innovation, they are unlikely to lead to shortages in the same way that price controls on something like toilet paper might. Instead, in most situations, the manufacturers really can and will just take a smaller profit, because the alternative is usually no profit. It's comparable to utilities, another area where we have legal/natural monopolies combined with price controls.
Beyond that, insurance means that people often don't even know what they are paying for the drugs (how many think drug prices are the co-pay?). They also don't directly feel the price of the drug, as the only feedback they get is higher premiums, or even just lower wages due to their employer paying higher premiums, and you feel those indirect costs whether you chose to purchase the drug, or if someone else in your risk pool did so.
This is obviously a bad idea if it's just the camel's nose to other price controls, but there are non-trivial arguments that price controls on drugs make sense.
I would like my money back for my subscription to your substack, Noah. Implying that the availability of fossil fuels in the US is the only thing keeping energy prices down is nonsense. Also half the section on abundance was bordering on strawman arguments.
Very nonplussed about several of these personally, because I have a safe tech job in the US - but I'm kind of stuck at it because nobody /else/ is hiring. Similarly I have dual citizenship, but the other one is the UK so it basically gets more useless every day.
We have the reverse issue where we are hiring, but nobody wants to live in Silicon Valley anymore (it's terminally boring and expensive) so we can't get anyone new.
"Last year, I argued that this was because the internet has been mostly completed, leaving much less work for software folks to do"
Dubious - most of the software out there has little to nothing to do with the "internet" other than using it. IF Noah is arguing that most companies have already got functional websites and apps - maybe. Except that most of those are buggy as hell and lacking security. So, yes, there is a valid argument that a lot of very bad software has already been written.
So, if we are all willing to accept crappy software, then Noah might be right. Although most large companies have basic internal software that is outdated and full of problems - someday someone will need to rewrite and replace it - "Tech Debt" is massive out there and growing daily.
AI is useful for "coding" - basically a powerful auto-complete. However "coding" is about 10% of what goes into making real software.
So, yes, I do expect "programming" jobs to no longer be so plentiful - whether due to AI productivity gains or just companies buying the BS that AI can fully replace software engineers. Real software engineers will still be in demand - just maybe not so many "programmers".
And yes, I might be quite wrong if no one cares that any software they interact with works and is secure and continues to work over time (and that might unfortunately be the case).
And yes, "elite overproduction" is part of this story. Although I would use "elite" very loosely. There was a big push to software automate a lot in normal businesses and that led to hiring a lot of "programmers" - many who were not even close to skilled software engineers. And lots of college grads told to learn "computers". So, yes, an overproduction of people who weren't particularly good at their jobs due to a big jump in demand without much understanding of what was to be produced.....
(As a side note, the fact that France’s energy costs have gone up as much as the UK’s, despite France mostly running on nuclear power, should serve as a wake-up call for people who think nuclear power can solve the UK’s problems.)
This isn't true. The cost of energy is determined by the cost of the *marginal* kWh. At the time scales we are talking about, the marginal kWh is nearly always from gas, because it's the only source of energy that can be practically scaled up within minutes to years. Nuclear takes decades to build, and hydrothermal, wind or solar aren't much better, and none of them can increase power within minutes or hours, so they provide mostly inframarginal kWh. Even if your country's solar is producing power for $0.15/kWh, businesses and homes are paying $0.40/kWh because that's the price that the gas plant operators are charging to put out the last, marginal kWh the grid needs.
Nuclear can make a difference to energy costs, but only on much longer timescales. This is also why energy costs are so high even in many places which are heavily adopting solar. The marginal kWh is still from gas. As long as this is so, Europe will be saddled with high energy costs until they find a different source of gas than Russia, or they move to a different source of marginal kWh, e.g. grid-scale batteries charged by solar.
"Nuclear takes decades to build, and hydrothermal, wind or solar aren't much better, and none of them can increase power within minutes or hours"
The answer to the marginal kWh is storage. At present China is deploying all-in BESS systems for $66/kWh [1]. This is in China, but these systems are mostly deployed on shipping containers that are easy to assemble into full plants. It is very hard to imagine a world where this does not become the dominant factor in determining marginal energy costs, and that world is coming very quickly. (Much more quickly than any new nuclear or hydro can even make it through planning.) I keep seeing folks having forward-looking discussions on energy provisioning that just pretend sub-$100/kWh battery storage isn't a thing. At what all-in storage cost do we all agree that the world has actually changed?
[1] https://archive.is/UXcdL
I was thinking this is a feature of pricing in the energy market. Shouldn’t this mean massive profits for French nuclear plant owners and also renewables everywhere across Europe? Do we see this? I feel like I haven’t seen much of that in the news.
When we talk about "the price of electricity", we're usually talking about the spot price. The "average price of electricity" is usually the average spot price. The spot price is determined by the supply and demand at the current moment, and can even become negative if there is more electricity supply than demand. It is the marginal supplier (and demander) that largely determines the spot price.
Not all electricity users buy at the spot price. You can have longer term contracts between producers or suppliers, there are wholesale and retail electricity markets, forwards and other derivatives, etc. The structure of the electricity market is local and complex, and fortunes are made or lost trading in them (cf. Enron). Different users end up paying different prices to different providers, after different middlemen mediate between them, depending on how it all shakes out. However, when we talk about "the price of electricity" we aren't usually talking about all this complexity. We're just talking about the spot price.
Yeah all that extra complexity does make it hard to ‘follow the money’. So your original comment was basically explaining why the chart Noah included doesn’t say much about the cost of energy for retail or industrial consumers in UK vs in France.
"Warren Buffet is dead" made me do a double take!
"Eventually, as more investors learned to be cool and rational like Buffett"
Complete speculation on my part, but I don't think that anyone is any more cooler and rational than they were in the 60s/70s. My thesis is that technology improvements and regulatory changes have led to cheaper price discovery => more efficient markets.
Couldn't France's price surge be because they're all connected to the same energy grid? Which would be solved if all countries on said grid (such as Germany and the UK) stopped their ongoing energy-seppuku?
The Tech job bust is interesting - I would bet that there is still a lot of demand for tech roles outside of the IT sector. My hope is that deploying all those tech workers to other industries will help push them to improve their productivity...
I think that any model of a political economy that ignores political power is just an invalid model. There is a reason that many Democratic reforms have been stymied in the past, and there's a reason that Trump has the political freedom to propose price controls right now (irrespective of whether those are a good idea or not.) To a very great extent this is because corporate concentration makes these organizations (1) extremely politically powerful in Congress, but also (2) extremely vulnerable to extra-legal interventions by a vindictive executive branch, which means in turn that (3) they will deploy their power in ways that support the political goals of a vindictive executive, while blocking those of a more reasonable political party.
Democrats can try to address the power imbalances, or they can learn to govern like Trump. But there is no stable political world in which one political party has the political freedom to impose sweeping price controls, and the other party is unable to do anything similar. To the extent that "centrist" beat down "leftists" and prevent them from addressing concentration of wealth and power, you're just ensuring the only option is Trump-style executive governance, whether from the right or left.
Trump is still an idiot, but would it make sense to insist that drug manufacturers don't charge higher prices in the United States than they do in, say, France, Japan, or any other country with a GDP per capita of over $30,000? (With one of the goals being *higher* drug prices in other countries, so that they pay their fair share of the R&D expenses.)
It’s time to face up to the fact that the Abundance program can only be achieved if the Democrats are in power. DOGE, Trump, and the entire Republican party are staging a full scale attack on state capacity, a vital part of the Abundance agenda. Democrats are generally fine with highly competent technical state employees. The annoying tendency of Dems to attack corporate power is best overlooked-consider it a necessary pushback against capitalist Bonopartism that we are experiencing now from Musk, Thiel, and others. I hope the Abundance program aim isn’t to trade democracy for Abundance.
I admit, I'm siloed enough that I have not seen this very reasonable critique of force-lowering drug costs before. While it's pretty convincing, I'm curious if anyone can explain what the alternative would be. Having to spend a huge proportion of your income on medicine is, to me, simply morally unacceptable in a rich society--is there a way to correct that which doesn't have the negative results Noah outlined here?
Most countries regulate the bulk purchasing of drugs. The US is the outlier. It’s reasonable to assert that US profits were buttressing the R&D that everyone else benefited from.
“Limiting these companies’ market size by limiting the prices they can charge will result in less research spending, and ultimately fewer new medicines and treatments.”
My question is what other incentives and mechanisms can be put into place to further drug innovation. More public R&D seems like a no-brainer.
This is another idea:
“Since reforming its system in the early 2010s, Germany has allowed manufacturers to freely set prices for a limited period when bringing new drugs to the market. It then uses the data available from that period for a nongovernmental and nonprofit research body to evaluate the benefit provided by the new drug, as compared to existing alternatives. This added benefit, or lack thereof, then serves as the foundation for price negotiations between drug manufacturers and health plans.”
https://theconversation.com/why-the-us-has-higher-drug-prices-than-other-countries-111256
“My question is what other incentives and mechanisms can be put into place to further drug innovation. More public R&D seems like a no-brainer. “
While probably a decent idea, I think the real expensive parts of drug development are actually finding the drugs that pan out once targets and research have been identified and doing the efficacy and safety trials. I doubt the public sector is going to finance all that.
Pharma's argument that they needed to sell drugs for high prices in America to fund research was always flawed. Most of the breakthroughs came through government funded research that was basically gifted to pharma.
It has been true for a long time that most medical research in the US is funded by the government through the National Institutes of Health (NIH), which is the largest public funder of biomedical research in the world.
But, trump just cut billions from NIH. Trump's price controls won't hurt innovation as much as his cuts to NIH will.
"Pharma's argument that they needed to sell drugs for high prices in America to fund research was always flawed. Most of the breakthroughs came through government funded research that was basically gifted to pharma."
The basic breakthroughs and target may come from basic research, but someone still needs to find the compounds, get everything working, get it through trials and regulatory clearances which is where the bulk of the expense is.
I'm certain that a comprehensive R&D plan could help with this. Unfortunately we're pairing this policy with a massive reduction in NIH funding, so it's hard to see good outcomes here. (Unfortunately they will be short-term *popular* outcomes, and the costs will be born by the next administration.)
"Having to spend a huge proportion of your income on medicine is, to me, simply morally unacceptable in a rich society."
That's really dumb.
As price controls go, it seems like drug prices are up there as an area where it may actually be a good idea, and you can't just resort to the Econ 101 counterargument. This is not exactly an area driven by market forces.
Not only do patents mean that drug makers have legal monopolies on many drugs, but even off-patent drugs are often effective monopolies due to the high capital costs and low marginal costs associated with producing many drugs, making market entry a risky prospect. This also means that, while price controls may drive down innovation, they are unlikely to lead to shortages in the same way that price controls on something like toilet paper might. Instead, in most situations, the manufacturers really can and will just take a smaller profit, because the alternative is usually no profit. It's comparable to utilities, another area where we have legal/natural monopolies combined with price controls.
Beyond that, insurance means that people often don't even know what they are paying for the drugs (how many think drug prices are the co-pay?). They also don't directly feel the price of the drug, as the only feedback they get is higher premiums, or even just lower wages due to their employer paying higher premiums, and you feel those indirect costs whether you chose to purchase the drug, or if someone else in your risk pool did so.
This is obviously a bad idea if it's just the camel's nose to other price controls, but there are non-trivial arguments that price controls on drugs make sense.
Was kind of hoping you would answer your question "What’s a capitalist to do in times like these?"!
I would like my money back for my subscription to your substack, Noah. Implying that the availability of fossil fuels in the US is the only thing keeping energy prices down is nonsense. Also half the section on abundance was bordering on strawman arguments.
I do like this series, Noah. Very useful.
Abundance liberalism - here's another example, https://www.rollingstone.com/politics/political-commentary/abundance-discourse-ezra-klein-trump-musk-democrats-1235310224/ .
Very nonplussed about several of these personally, because I have a safe tech job in the US - but I'm kind of stuck at it because nobody /else/ is hiring. Similarly I have dual citizenship, but the other one is the UK so it basically gets more useless every day.
We have the reverse issue where we are hiring, but nobody wants to live in Silicon Valley anymore (it's terminally boring and expensive) so we can't get anyone new.
Yogi Berra would be proud! ;)
"Last year, I argued that this was because the internet has been mostly completed, leaving much less work for software folks to do"
Dubious - most of the software out there has little to nothing to do with the "internet" other than using it. IF Noah is arguing that most companies have already got functional websites and apps - maybe. Except that most of those are buggy as hell and lacking security. So, yes, there is a valid argument that a lot of very bad software has already been written.
So, if we are all willing to accept crappy software, then Noah might be right. Although most large companies have basic internal software that is outdated and full of problems - someday someone will need to rewrite and replace it - "Tech Debt" is massive out there and growing daily.
AI is useful for "coding" - basically a powerful auto-complete. However "coding" is about 10% of what goes into making real software.
So, yes, I do expect "programming" jobs to no longer be so plentiful - whether due to AI productivity gains or just companies buying the BS that AI can fully replace software engineers. Real software engineers will still be in demand - just maybe not so many "programmers".
And yes, I might be quite wrong if no one cares that any software they interact with works and is secure and continues to work over time (and that might unfortunately be the case).
And yes, "elite overproduction" is part of this story. Although I would use "elite" very loosely. There was a big push to software automate a lot in normal businesses and that led to hiring a lot of "programmers" - many who were not even close to skilled software engineers. And lots of college grads told to learn "computers". So, yes, an overproduction of people who weren't particularly good at their jobs due to a big jump in demand without much understanding of what was to be produced.....
Small nitpick in item three where you say "that does not phase them much", you want the homophone 'faze' there instead of phase.