Moses Sternstein on health care; Steven Pinker on common knowledge; Lionel Page on long time horizons and civilized behavior; J. Zachary Mazlish on AI and economic growth
I think it is a fallacy to assume AI being able to economically out-compete a lot of human labor means explosive growth, as opposed to a slightly higher rate of growth. There are still going to be scarcity of key inputs and various important bottlenecks, some material, some conceptual (discovery, innovation) and some regulatory (including important geopolitical considerations). Consider electricity, prices have recently gone way up, everything up and down the chain is tight, and production growth is paltry compared to demand. Yeah, "that's what's booms are always like", ok, but this boom is just going to keep booming now because power = production = money. Even with big efficiency gains, AI still can't explode without electricity exploding, and, domestically, it's not, it can't keep up. We need 1,000 more nuclear reactors in a decade - yes, one thousand, two new brought online per week - and we'd be lucky to get two -total- the normal way at the currently expected rate. Things are changing fast, but they can still only get physically built and operating in reality so fast. China builds biggest fastest, so in retrospect we'll probably say our stupid refusal to allow ourselves to keep pace is how they won the future.
Levine is a Professor at Berkeley and is working on using LLM like models for physical robots. His estimate is that in about 5 years there will be useful physical robots.
It doesn't really matter if these are AGI.
A robot that could be really useful in construction, agriculture, mining and other physical tasks would be transformative.
“The problem is that we associate health care with something given (the gift of health) to people who are suffering.”
Arnold doesn’t usually go for the simple/simplistic argument, but I think he has here.
Because by this logic, this would surely apply to food and shelter as well. But we don’t have 3rd party payers for those things. [Yes, we give some of each to the poor, but that is still a minority.]
I think the primary reason is that catastrophic care could wipe someone out financially, and “we” think both that catastrophic care should be “given” *and* that it should not bankrupt anyone.
And this argument has been allowed to infuse the *entirety* of the healthcare system.
It is unlikely to ever happen, but my solution [edit: really “proposal”] would be to socialize - at the national level re: the money, even if administered by states - catastrophic care while simultaneously deregulating *all of the rest* of healthcare and health insurance.
Yes, we would still have the issue of ordinary healthcare for the poor (my answer would be “crappy” service with small co-pays and long queues primarily provided by foreign doctors seeking to live in the U.S., paid for by government but not government run).
But we would otherwise return to market forces for 65% - 75% of healthcare services.
Your suggestion isn't any better. First there is no dividing line between catastrophic and not. Not only does everyone have a different ability to pay but the two biggest costs aren't addressed. End of life care would still be catastrophic and it would be difficult to impossible to categorize chronic conditions as catastrophic or not.
The Swiss may have the best practical healthcare system, at least that i am aware of. Hallmark of it is mandated coverage for all citizens who buy healthcare insurance through private health insurers who government requires to provide certain coverage for any max out of pocket costs for insureds borne by the healthcare providers.
For the poor, the gov provides assistance for them to pay premiums.
You can define “catastrophic” cost for a given year / “Incident” based on income (or assets). We do similar things with our tax code and welfare benefits right now. Surely that part would not be particularly difficult.
I did not claim my proposal would solve all problems.
Re: tradeoffs, I am merely claiming that this approach would be way mo’ bettah than our existing system.
The tradeoffs in our current system are particularly terrible, precisely because of the “people aren’t paying for health services with their own money” issue.
I do, however, apologize for having used the word “solution” above. You channeling Thomas Sowell are correct about that.
What you just proposed exists today. In my younger years I had it. It's called high deductible health insurance. The way it typically works is you spend the year hoping to avoid ANY medical bills. If you end up with a lot of small ones, or especially one big one, you hit you cap where insurance pays most or all the cost. Then you try to get all the care you've avoided. My experience was that it was awful. No thanks.
At root, the problem with your proposal is that routine care is a tiny part of the total. The vast majority is chronic care and catastrophic events that would almost all be covered by your insurance plan. Your proposal doesn't address any of the problems.
“What you just proposed exists today. In my younger years I had it. It's called high deductible health insurance.”
No it is not, and no you did not (unless perhaps you are over 80, when it is at least plausible you had something somewhat similar for a few years).
High deductible is indeed a part of my proposal. But it’d be an even higher cap than all of today’s plans.
But you completely ignored the key part: deregulate ALL else.
When that happens, the cost of most other things will come way down.
See the Russ Roberts podcast with the OK doctor who doesn’t take insurance for an example.
You absolutely cannot claim it’s like anything we have today, given how highly regulated, cost-shifted and price opaque the current system of mostly 3rd party payers for everything is.
I had a high deductible plan. High deductible plans are available today.
Regulation is an ambiguous term. Based on your reference to the econtalk podcast, it seems you are mostly not talking about regulation. at least not the types of regulation I don't want to see go away. (example: I don't want prescription drugs to be deregulated to be more like supplements, which are essentially only regulated to limit gross exaggerations of what the supplement does. And there is virtually no scientific confirmation that the supplements do what they are supposed to.)
The surgical center referred to in the econtalk podcast has somewhat lower costs but it still isn't cheap.
While the surgery center doesn't accept "insurance," they accept third party payments. Many of their patients are employees of self-insured companies.
there is nothing stopping other doctors and medical providers from doing this. one has to ask why they haven't.
Another question has to be how these patients learned what surgery they needed.
And that leads to the question of why the doctors of this surgery center don't provide primary care or other non-surgical specialists. And what about prescription drugs?
And that still leaves the question of whether even this surgery center can be replicated. Is it successful because of its business plan or because its business plan attracts the best doctors? Or because it is managed far better than other medical provider companies?
According to the podcast, one of the problems is supplemental payments to providers based on list prices versus billed prices. That is not an insurance problem and it is not a regulation problem but it sounds like a HUGE problem.
You don't want regulation but I'd argue one of the best steps would be to require providers to post prices for everything they do, preferably also limiting the ability of providers to overly complicate the price list. That would be regulation.
We disagree on almost everything else - regulation is *not* actually an ambiguous term, e.g., and I didn’t say you didn’t have a high deductible plan, I said you did NOT have a plan anything like my proposal (where almost everything has been deregulated) - but I’m with you that requiring posting of prices would be a good thing.
I’m not an anarcho-capitalist; I’m a classical liberal. Disclosure / transparency laws I am generally either explicitly for, or at least ok with. And given anything like our *current* healthcare system, requiring posted prices is an excellent idea.
"I do not think we will get true AGI. I think that the combination of the physical world being difficult to model, complex tasks being difficult to accomplish, and the need to have memory of all sorts of human interactions will keep computers from becoming know-it-alls."
This seems like an argument against human-level intelligence existing. Since humans exist, these are clearly solvable problems, and they mostly sound like they get automatically solved if you just assume that the trend for memory and computer to get cheaper continues.
AGI is a myth and chasing it without economic discipline will bankrupt those involved. We know AGI is a myth because no matter how the software is designed, the best it can do is either imitate existing work or make "guesses" with no capacity to measure if what was guessed has validity.
If you give the software a lot of data and models of analysis the software can do wonders. I see this as taking computer processing to a much higher level, both in performance and in synthesis. What the software cannot do is make itself better. And that is the true measure of AGI.
If computers and software are able to self improve then that would be nirvana. Then humans could simply focus on producing electricity and feed the computers and we would realize all sorts of "perfect" algorithms and designs and creations, etc. Alas, the computers do not know what improvement is. They can only improve according to the factors provided external to the computer.
A computer has no conscience and thus cannot know what it wants and it cannot know whether it has obtained the thing it doesn't know it wants. The computer can still only do what it is told.
The other thing to keep in mind is that Medicaid isn't "charity." No one is being charitable. It's perhaps a charitable sentiment, but it's all the feels, none of the sacrifice. It's the junkfood equivalent of kindness.
Charity includes an embedded constraint on spending: the donor's willingness and ability to underwrite the project. There's no such thing with nationalized healthcare, especially when the "donor" in this case is future generations of taxpayers.
I mean, Imagine we get AGI, but the computers are really expensive to run at this level, and not fast, and the AGI has only average IQ.
You won't get any economic benefit from that.
The assumption is that AGI will be cheaper per "Thought" from a human at comparable level. We don't know. Of course, AGI are much more amenable to improvements.
"As the time horizon gets longer, people are more and more able to approximate full cooperation."
I've often heard that increased complexity has left more and more people behind but maybe the time horizon is a bigger factor. People who can think and act using long-term strategy have a huge advantage over those who don't.
“Medicare has been steadily disinflationary since January 2022, again, in contrast to Private prices, which have accelerated substantially of-late.
Why did this happen? “
Not being a subscriber, I would be interested in his explanation. But what I believe the standard explanation for the relationship between Medicare pricing cuts being correlated with private price increases is “cost shifting.” Not sure how cost shifting is considered dis-inflationary.
Private prices reflect the government’s setting of low reimbursement rates:
“Commercial prices reflect the combined effect of government underpayments, the inflationary impact of larger ecosystem factors beyond hospital’s control driving up input costs, and structural differences among hospitals and other care settings. The results of significant actions to reduce operating, clinical, and capital costs are often masked within the data by the larger ecosystem and policy forces driving increased input costs and the related cost shift that occurs.”
Part of the reason why people don’t just go out and pay the doctor directly is because the doctor charges the uninsured a higher price to make up for the lower amount he is receiving from the government payer. Do people think that the Emergency Medical Treatment and Active Labor Act (EMTALA) is disinflationary because people in a health care emergency just show up at the emergency room and don’t have to pay anything?
The short teaser doesn’t really explain what Sternstein means by “we’d be much better off if we just tried to pay for it ourselves.” One wants to assume that he doesn’t mean that he wants to outlaw health insurance. But at any rate he ought to be receiving a nice “Thank You” letter from Zohran Mamdani. Every right leaning pundit’s pontification on health reform strikes a blow for the cause of social democracy. And Trump’s preserve the status quo approach is definitely no way to MAGA.
If the right doesn’t want to be sitting on the sidelines kibitzing when real health reform goes down, they would do well to come up with something new that (1) provides universal coverage, (2) preserves a private option, (3) eliminates the federal tax provisions related to health care and consolidates and simplifies all existing federal health programs and transfers their control to the states, and (4) provides for federal funding directly to the states of say about 20% of total annual treasury receipts. Similar arrangements are successful all around the world, even in developing countries.
When Otto von Bismarck was cornered by the Socialist Worker’s Party, he didn’t sit around and mouth libertarian platitudes. The right in the US might learn from his example.
A big problem now is that is so difficult for anyone to even agree on the basic facts. We can only have reasonable discussions about solutions to problems if we can agree on the facts. If we can't agree on the facts, discussion is pointless.
Agreement on facts allows discussion to begin. But to reach conclusion, acceptable to all, there must be shared premises. For instance, if one person has premise "Property is theft", and another "property is fruit of my labor", they are not going to reach a conclusion.
Liberalism erodes the stock of shared premises in a society leading to a situation where you can not be sure whether you share premises with any other person. The society dissolves into an aggregate of strangers,
Discussion starts by understanding which "facts" we disagree. From there it is sometimes possible to work toward a plan of action. Discussion is only pointless when neither side tries to understand what the other is saying.
I think it is a fallacy to assume AI being able to economically out-compete a lot of human labor means explosive growth, as opposed to a slightly higher rate of growth. There are still going to be scarcity of key inputs and various important bottlenecks, some material, some conceptual (discovery, innovation) and some regulatory (including important geopolitical considerations). Consider electricity, prices have recently gone way up, everything up and down the chain is tight, and production growth is paltry compared to demand. Yeah, "that's what's booms are always like", ok, but this boom is just going to keep booming now because power = production = money. Even with big efficiency gains, AI still can't explode without electricity exploding, and, domestically, it's not, it can't keep up. We need 1,000 more nuclear reactors in a decade - yes, one thousand, two new brought online per week - and we'd be lucky to get two -total- the normal way at the currently expected rate. Things are changing fast, but they can still only get physically built and operating in reality so fast. China builds biggest fastest, so in retrospect we'll probably say our stupid refusal to allow ourselves to keep pace is how they won the future.
Regarding AI, it is worth listening to the Dwarkesh interview with Sergey Levine .
https://www.dwarkesh.com/p/sergey-levine
Levine is a Professor at Berkeley and is working on using LLM like models for physical robots. His estimate is that in about 5 years there will be useful physical robots.
It doesn't really matter if these are AGI.
A robot that could be really useful in construction, agriculture, mining and other physical tasks would be transformative.
"common knowledge is necessary for coordination"
Arguments can only proceed to conclusion, if premises are shared, in addition to the facts.
The space of shared premises is what States provide to their citizens. This is what enables the citizens to peaceably live together.
“The problem is that we associate health care with something given (the gift of health) to people who are suffering.”
Arnold doesn’t usually go for the simple/simplistic argument, but I think he has here.
Because by this logic, this would surely apply to food and shelter as well. But we don’t have 3rd party payers for those things. [Yes, we give some of each to the poor, but that is still a minority.]
I think the primary reason is that catastrophic care could wipe someone out financially, and “we” think both that catastrophic care should be “given” *and* that it should not bankrupt anyone.
And this argument has been allowed to infuse the *entirety* of the healthcare system.
It is unlikely to ever happen, but my solution [edit: really “proposal”] would be to socialize - at the national level re: the money, even if administered by states - catastrophic care while simultaneously deregulating *all of the rest* of healthcare and health insurance.
Yes, we would still have the issue of ordinary healthcare for the poor (my answer would be “crappy” service with small co-pays and long queues primarily provided by foreign doctors seeking to live in the U.S., paid for by government but not government run).
But we would otherwise return to market forces for 65% - 75% of healthcare services.
Agreed Arnold's suggestion is overly simplistic.
Your suggestion isn't any better. First there is no dividing line between catastrophic and not. Not only does everyone have a different ability to pay but the two biggest costs aren't addressed. End of life care would still be catastrophic and it would be difficult to impossible to categorize chronic conditions as catastrophic or not.
My answer? There is none, only tradeoffs.
The Swiss may have the best practical healthcare system, at least that i am aware of. Hallmark of it is mandated coverage for all citizens who buy healthcare insurance through private health insurers who government requires to provide certain coverage for any max out of pocket costs for insureds borne by the healthcare providers.
For the poor, the gov provides assistance for them to pay premiums.
You can define “catastrophic” cost for a given year / “Incident” based on income (or assets). We do similar things with our tax code and welfare benefits right now. Surely that part would not be particularly difficult.
I did not claim my proposal would solve all problems.
Re: tradeoffs, I am merely claiming that this approach would be way mo’ bettah than our existing system.
The tradeoffs in our current system are particularly terrible, precisely because of the “people aren’t paying for health services with their own money” issue.
I do, however, apologize for having used the word “solution” above. You channeling Thomas Sowell are correct about that.
What you just proposed exists today. In my younger years I had it. It's called high deductible health insurance. The way it typically works is you spend the year hoping to avoid ANY medical bills. If you end up with a lot of small ones, or especially one big one, you hit you cap where insurance pays most or all the cost. Then you try to get all the care you've avoided. My experience was that it was awful. No thanks.
At root, the problem with your proposal is that routine care is a tiny part of the total. The vast majority is chronic care and catastrophic events that would almost all be covered by your insurance plan. Your proposal doesn't address any of the problems.
“What you just proposed exists today. In my younger years I had it. It's called high deductible health insurance.”
No it is not, and no you did not (unless perhaps you are over 80, when it is at least plausible you had something somewhat similar for a few years).
High deductible is indeed a part of my proposal. But it’d be an even higher cap than all of today’s plans.
But you completely ignored the key part: deregulate ALL else.
When that happens, the cost of most other things will come way down.
See the Russ Roberts podcast with the OK doctor who doesn’t take insurance for an example.
You absolutely cannot claim it’s like anything we have today, given how highly regulated, cost-shifted and price opaque the current system of mostly 3rd party payers for everything is.
https://www.econtalk.org/keith-smith-on-free-market-health-care/
I had a high deductible plan. High deductible plans are available today.
Regulation is an ambiguous term. Based on your reference to the econtalk podcast, it seems you are mostly not talking about regulation. at least not the types of regulation I don't want to see go away. (example: I don't want prescription drugs to be deregulated to be more like supplements, which are essentially only regulated to limit gross exaggerations of what the supplement does. And there is virtually no scientific confirmation that the supplements do what they are supposed to.)
The surgical center referred to in the econtalk podcast has somewhat lower costs but it still isn't cheap.
While the surgery center doesn't accept "insurance," they accept third party payments. Many of their patients are employees of self-insured companies.
there is nothing stopping other doctors and medical providers from doing this. one has to ask why they haven't.
Another question has to be how these patients learned what surgery they needed.
And that leads to the question of why the doctors of this surgery center don't provide primary care or other non-surgical specialists. And what about prescription drugs?
And that still leaves the question of whether even this surgery center can be replicated. Is it successful because of its business plan or because its business plan attracts the best doctors? Or because it is managed far better than other medical provider companies?
According to the podcast, one of the problems is supplemental payments to providers based on list prices versus billed prices. That is not an insurance problem and it is not a regulation problem but it sounds like a HUGE problem.
You don't want regulation but I'd argue one of the best steps would be to require providers to post prices for everything they do, preferably also limiting the ability of providers to overly complicate the price list. That would be regulation.
We disagree on almost everything else - regulation is *not* actually an ambiguous term, e.g., and I didn’t say you didn’t have a high deductible plan, I said you did NOT have a plan anything like my proposal (where almost everything has been deregulated) - but I’m with you that requiring posting of prices would be a good thing.
I’m not an anarcho-capitalist; I’m a classical liberal. Disclosure / transparency laws I am generally either explicitly for, or at least ok with. And given anything like our *current* healthcare system, requiring posted prices is an excellent idea.
"I do not think we will get true AGI. I think that the combination of the physical world being difficult to model, complex tasks being difficult to accomplish, and the need to have memory of all sorts of human interactions will keep computers from becoming know-it-alls."
This seems like an argument against human-level intelligence existing. Since humans exist, these are clearly solvable problems, and they mostly sound like they get automatically solved if you just assume that the trend for memory and computer to get cheaper continues.
AGI is a myth and chasing it without economic discipline will bankrupt those involved. We know AGI is a myth because no matter how the software is designed, the best it can do is either imitate existing work or make "guesses" with no capacity to measure if what was guessed has validity.
If you give the software a lot of data and models of analysis the software can do wonders. I see this as taking computer processing to a much higher level, both in performance and in synthesis. What the software cannot do is make itself better. And that is the true measure of AGI.
If computers and software are able to self improve then that would be nirvana. Then humans could simply focus on producing electricity and feed the computers and we would realize all sorts of "perfect" algorithms and designs and creations, etc. Alas, the computers do not know what improvement is. They can only improve according to the factors provided external to the computer.
A computer has no conscience and thus cannot know what it wants and it cannot know whether it has obtained the thing it doesn't know it wants. The computer can still only do what it is told.
The other thing to keep in mind is that Medicaid isn't "charity." No one is being charitable. It's perhaps a charitable sentiment, but it's all the feels, none of the sacrifice. It's the junkfood equivalent of kindness.
Charity includes an embedded constraint on spending: the donor's willingness and ability to underwrite the project. There's no such thing with nationalized healthcare, especially when the "donor" in this case is future generations of taxpayers.
I mean, Imagine we get AGI, but the computers are really expensive to run at this level, and not fast, and the AGI has only average IQ.
You won't get any economic benefit from that.
The assumption is that AGI will be cheaper per "Thought" from a human at comparable level. We don't know. Of course, AGI are much more amenable to improvements.
"and we’d be much better off if we just tried to pay for it ourselves"
Sure. But what percent of people would be able to do that?
Just as importantly, how much would prices really come down? How many of these new drugs would exist? Would preventative care get better or worse?
"As the time horizon gets longer, people are more and more able to approximate full cooperation."
I've often heard that increased complexity has left more and more people behind but maybe the time horizon is a bigger factor. People who can think and act using long-term strategy have a huge advantage over those who don't.
Moses Sternstein writes:
“Medicare has been steadily disinflationary since January 2022, again, in contrast to Private prices, which have accelerated substantially of-late.
Why did this happen? “
Not being a subscriber, I would be interested in his explanation. But what I believe the standard explanation for the relationship between Medicare pricing cuts being correlated with private price increases is “cost shifting.” Not sure how cost shifting is considered dis-inflationary.
Private prices reflect the government’s setting of low reimbursement rates:
“Commercial prices reflect the combined effect of government underpayments, the inflationary impact of larger ecosystem factors beyond hospital’s control driving up input costs, and structural differences among hospitals and other care settings. The results of significant actions to reduce operating, clinical, and capital costs are often masked within the data by the larger ecosystem and policy forces driving increased input costs and the related cost shift that occurs.”
(https://www.healthaffairs.org/content/forefront/cost-shift-health-care-ecosystem-and-commercial-prices )
Part of the reason why people don’t just go out and pay the doctor directly is because the doctor charges the uninsured a higher price to make up for the lower amount he is receiving from the government payer. Do people think that the Emergency Medical Treatment and Active Labor Act (EMTALA) is disinflationary because people in a health care emergency just show up at the emergency room and don’t have to pay anything?
The short teaser doesn’t really explain what Sternstein means by “we’d be much better off if we just tried to pay for it ourselves.” One wants to assume that he doesn’t mean that he wants to outlaw health insurance. But at any rate he ought to be receiving a nice “Thank You” letter from Zohran Mamdani. Every right leaning pundit’s pontification on health reform strikes a blow for the cause of social democracy. And Trump’s preserve the status quo approach is definitely no way to MAGA.
If the right doesn’t want to be sitting on the sidelines kibitzing when real health reform goes down, they would do well to come up with something new that (1) provides universal coverage, (2) preserves a private option, (3) eliminates the federal tax provisions related to health care and consolidates and simplifies all existing federal health programs and transfers their control to the states, and (4) provides for federal funding directly to the states of say about 20% of total annual treasury receipts. Similar arrangements are successful all around the world, even in developing countries.
When Otto von Bismarck was cornered by the Socialist Worker’s Party, he didn’t sit around and mouth libertarian platitudes. The right in the US might learn from his example.
I won't argue for or against your proposal but everything before it seems on track.
Re: Common Knowledge
A big problem now is that is so difficult for anyone to even agree on the basic facts. We can only have reasonable discussions about solutions to problems if we can agree on the facts. If we can't agree on the facts, discussion is pointless.
Agreement on facts allows discussion to begin. But to reach conclusion, acceptable to all, there must be shared premises. For instance, if one person has premise "Property is theft", and another "property is fruit of my labor", they are not going to reach a conclusion.
Liberalism erodes the stock of shared premises in a society leading to a situation where you can not be sure whether you share premises with any other person. The society dissolves into an aggregate of strangers,
Discussion starts by understanding which "facts" we disagree. From there it is sometimes possible to work toward a plan of action. Discussion is only pointless when neither side tries to understand what the other is saying.