Years ago a network scientist named Samuel Arbesman wrote a book about this kind of thing. How there are facts that change constantly -- the weather -- so we know we need to check the latest. And there are facts that never change -- the names of the planets -- so we know we don't need to bother checking on them. But there's another category of fact he called "mesofacts" (a term which obviously failed to catch on) for facts that change over the span of a few decades and people seem poorly suited cognitively to deal with because nobody puts a Google Calendar reminder for 2035 to check the current consensus on Industrial Policy.
This is interesting and it parallels another point in evolution. For some animals, their environment changes so slowly over evolutionary time that it’s best to just wire in optimal behavior genetically as instinct. For other animals, their environment changes so quickly over evolutionary time that it’s best for evolution to wire in a learning ability so that each individual learns what behavior works best in the environment as they experience it. But there are some animals for whom the environment changes at a medium speed, and thus it’s best not to wire in behavior, but instead to wire in a propensity to learn culturally, so that individuals can use the optimized behaviors of the generations before them, rather than having to reinvent it all from scratch, or know it instinctually. The anthropologist Joe Henrich says that humans have this feature in his book The Secret of Our Success. Cultural learning turns out to be really powerful at developing traditions that optimize for all environments on earth, including developing superstitions and cultural practices that happen to help against local dangers without anyone knowing that this is why they do them.
Yes, Heinrich’s “The Secret of Our Success” is a fantastic book that is full of interesting observations. For those who have not yet read it, you might want to read a summary here:
It’s not so much about not sharing knowledge - it’s rather that humans have a tendency to use social learning to blindly copy practices without understanding them, while many other species do social learning by seeing that another individual does something successful, and then figures out for themself how to be successful rather than blindly copying the model. Henrich gives some experimental examples, where a human child who sees an adult doing something weird before opening a box that has candy in it will often copy the weird thing as well as the action that opens the box, while a chimpanzee will just copy the parts that seem relevant to opening the box.
He argues that this blind copying is how humans learn to do long processes that are actually helpful even though no one knows why - you boil the cassava for many hours before eating it, because that’s what grandma always did, even though it seems soft enough to eat after a few minutes, but it turns out that if you only boil it a few minutes, then it has a low amount of cyanide that will build up in your body if you eat it this way for many years.
But I think there are also plenty of species like cats and squids that have lots of individual intelligence, but very little social learning.
I feel like this is this is the central insight of conservatism, and it’s genuinely a good insight. There is a class of things that work really well that we don’t fully understand *why* they work. Is this class is large and important, something like the conservative worldview emerges as an effective strategy.
You see a ton of friction around this with parenting where, to a first approximation, the current generation of parents are the first ones in millions of years to have any actual scientific/medical evidence on anything about child-rearing. And lots of grandparents react very poorly to things being done differently than they did it, as if "we've learnt better" means "you are a bad person".
(I just read a story this morning about grandparents who took a 21-month old toddler to the zoo without a car seat and didn't see anything wrong with that.)
One unclear aspect of this interesting essay is whether it is written from the point of view of the average American household income or the median American household income. In other words, are we talking about households making $106,270 or $80,610? There's a great gap in disposable income between those two figures.
Also, I see a recurring insensitivity to past inflation. The idea that prices haven't gone up in the last five years doesn't mean that the pain has gone away. it just means that the thumb screws are no longer being tightened, after reaching the level of torture in the past. Economists seem to think that the pain of past inflation and (a continuing inflated status) evaporates instantaneously.
OTOH, it is refreshing to learn that both American higher education and American health care are somewhat responsive to market forces.
A person making $42,000 a year doesn't sound like they're in a position to buy much in the way of higher education or healthcare.
I gave numbers for household income.
I got those figures from Bing copilot. Here is what Google has to say on the topic.
In 2023, the median household income in the United States was $80,610, and the average household income was $135,700 according to FRED, the Federal Reserve Bank of St. Louis. The median household income represents the middle value, with half of households earning more and half earning less. The average, also known as the mean, is calculated by adding up all household incomes and dividing by the number of households.
On behalf of all economists, I admit to really believing that people hate the first derivative of the price level. :) It is sort of too bad that the don't. There ARE ways to reduce _inflation_ at low cost in terms of creating unemployment. There really is not any good way to reduce the price level.
“I feel your pain” :) Biden should have been pantomiming reducing the price level, like a tax increase. Almost anything good could be claimed to be to “control prices.”
But here I’m getting out of my lane. I officially know nothing about politics.
Though being responsive to market forces is not all upside. In this case it means that 2 million fewer Americans have benefitted from a college education than 20 years ago because they've been priced out of it. And if, as appears likely, the fundamental divide between Trump and non-Trump voters is whether or not someone has been to college, then this market adjustment has actually been a disaster for progressive (or even rational) American politics. Fewer people getting a college education = more Trump supporters.
This reasoning is why the democrats are going to keep losing. I AM a college graduate who voted for trump because the democrats refused to listen to my rational concerns about inflation and unchecked illegal immigration. They assumed that I couldn’t think for myself or that their trump hysteria would scare me. I was hoping to vote for a reasonable democrat until the party put Kamala Harris in, again assuming I would just blindly follow them. People without college degrees aren’t stupid or credulous; like me, they know their own interest, and it’s not the Democrats. And they love their country, which has not been my experience with college-educated democrats, who insist on emphasizing all our flaws to the point of wishing America didn’t exist. Most Americans look at elite democrats as sellouts at best and traitors at worst. That kind of self hatred does not win elections.
Who said anything about them being stupid? I said they're more likely to vote for Trump than people who go to college. I actually think it's probably because a college education is increasingly viewed as a wealth-based gating exercise which determines, from a fairly early age, what you'll end up earning. By the same token, just because you went to college doesn't mean you're not stupid. Clearly there's still a lot of work to do before the US becomes fluent in anti-elitism. Plutocracy masquerading as meritocracy has done enormous damage to social cohesion in both the US and the UK. But especially in the US.
Rich con men masquerading as "champions of the working class" have done equal, if not more damage. To the extent that you're an intelligent person angry at college-educated elites, that makes total sense to me, and I can even see voting for Trump as a way of burning that all down. But actually believing Trump has your interests at heart does, in fact, make you kind of stupid.
Yes, exactly my view. Believing Trump is going to help you is an insane leap of faith, unless you're a millionaire. All the evidence points in the opposite direction. A lot of Trump voters are going to lose healthcare cover, for example. And they voted for Trump knowing that would be the case. Why would someone do that?!
Do you feel good about your choice? That is, do you think the actual government you voted for is/is going to do a good job at addressing your concerns, or is that beside the point?
My opinion is that way too many kids go to college. It should be aimed at the top third or quarter of kids, as most kids are better off in less cerebral pursuits and other careers.
That said, the cash strapped kids can usually go to junior college for the first two years and then transfer to a local state school, preferably while living at home. There are then countless grants and assistance packages, loans and part time jobs that contribute to college costs.
I agree with all your points. What I meant by responsive to market forces is that the upward trend in college and medical costs seemed absolutely unstoppable, but the inability of the average person to afford them seems to have had at last a small moderating effect.
While I think a college education should be available to all who can benefit, I don't think it's the answer for everyone. I wrote a piece about the need for government commitment to free and supported lifelong job retraining here:
Yes, I agree. HE's sorting function needs to be moderated by genuine, and genuine respect and support for, forms of education and training available in later life.
It’s interesting to see a cognitive bias evidenced in the comments where nothing can ever be fixed or even get better. College had gotten significantly cheaper. No it hasn’t! No, here is the data, it has. No, I was told it was rising 20 years ago and nothing ever gets better to it’s still the way it was.
Well thanks to the Trump administration suspending student visas, and trying to expel current students from China, we can look forward to university’s tuition charges to start rising again, since the schools bring in much more revenue from international students who pay full tuition, don’t get financial aid, sports scholarships or other discounts and have extra fees tacked on as well. To make up that lost money, they will need to raise fees for other students.
Of course the bigger impact will be from the lost human capital and scientific (including medical research) that will be lost along with these students, who on average are better qualified by being selected from a pool bigger than the 4% of global population that is American. Most of my doctors have been immigrants, and they also serve rural areas with doctor shortages.
Of course the tariffs will also raise manufactured goods prices, so we can at least not blame the higher living costs only on services.
The huge jump in tuition in the 2000s occurred as the percentage of foreign students rose dramatically. The data seems to show that increasing the percentage of foreign students increases the pace of tuition hikes. If we were to revert to 2003 tuition levels (CPI-adjusted) and 2003 percentages of foreign students (still significant), I’d be very happy!
There are two aspects that make up the perceived cost of a good or service. The price and how much I get for that price (quality). The rate on increase in healthcare spending is less than it was that is a numerical fact. My perception of what I get for the dollar of spending has definitely declined over the last 20 years. A standard half hour doctor visit has become 12 minutes, I used to be able to get an appointment to see a specialist in two weeks,now it can take six months and I got sent home from a hospital stay far sooner than I would in Czechia. On the other hand I just replaced my TV that was five years old. The new one is far superior and cost $200 less. There are quality adjustments in good CPI I’m not sure if how or if that is done with services. And don’t get me started when I’m paying for plumbers.
Re the healthcare spending chart: given the ACA had a large Medicaid expansion, and Medicaid has low payment rates to providers, could that be what's driving the chart? Like a composition effect, where the addition of these low-paying customers is keeping the average down ?
Also may reflect how “prices” are calculated vs actual costs.
Let’s say a hospitals “list price” for a procedure is $10k but its average reimbursement (the real price, tracked in the data series) for it between private insurance and Medicare is around $4k.
Used to be that an uninsured person came in and was billed the $10k list price. Many didn’t pay the bill and debt was sold to collectors or written off. A few people actually paid the $10k. Maybe the true average payment the hospital receives for all the $10k bills it sends to the uninsured was $2 or $3k (a bit lower than the $4k Medicare/private average):
Now hospitals sign up the uninsured for Medicaid at the front door if the person qualifies. Instead of a bill for the list price of $10k the person gets charged (rather, the government gets charged) $3k. You could have a situation where the same person buys the same service and the hospital receives $3k net on average after each, but pre AHA that was billed as $10k and post AHA $3m. Because of the expansion you’d also have more people consuming the product at the 3k price, bringing down the average price.
Just a theory, That would really depend on whether the healthcare cost data series is tracking billed prices to the patient or the net received by the hospital after write offs. I have no idea.
In any case, great point about the expansion (and also more regular Medicaid sign ups) driving weighted average prices down
I think it measures total payments for all medical services in the country, as a fraction of GDP. If all you do is add spending, it can only make this number grow. This is different from something like “average wages per employee”, which can go down when unemployed people become employees, or “medical spending per insured person”, which could go down when uninsured people become insured.
If the Medicaid expansion is reducing this, it would have to be because Medicaid is paying for cheap medical services that prevent conditions that used to lead to expensive medical problems that got treated by emergency rooms.
Hmm, I see your point re that chart. I was responding more to the narrative provided, like "What’s happening is that although health care prices have still been going up, they’ve been going up more slowly than before." The Medicaid costs are more tightly capped, so I was wondering if those cost controls were doing a lot of the work in this expense slowdown.
What is the evidence that the flattening of health care costs was not due to improved productivity? After all, recent years have seen a decline in coronary artery revascularization procedures due to better control of hypercholesterolemia and placement of coronary artery stents by catheterization, a reduction in cesarean section deliveries, more limited resections of breast cancers, and less invasive surgery of every sort due to the use of endoscopy for procedures rather than open surgery. It is true that such innovations as CAR-T cell infusions as a treatment for cancer are very costly, but the trend toward less invasive surgery may very well have reduced the intensity of medical care and, hence, its cost.
This is an interesting theory, but like all interesting theories my objection is that we shouldn't have to speculate about it. The data is out there. Also unrelated to this, but the flat trends of Noah's chart don't appear to show up in actual medical cost trends, which all seem to be above inflation levels (if only just a little in the super-high inflation years) [1].
Hmm... I am not sure services in general are persistently high. I can think of many services where "productization" or "franchising" models have taken cost down. Rather GOVERNMENT connected services have been persistently high (health, education being at the top of list). A great deal of the reason is that the velocity of innovation has been retarded due to the revenue models controlled by government. Also, inventive models do not reward cost optimization. Interestingly, the COVID pandemic created an emergency situation which unleashed a number of innovations, and I wonder if that is the reason for the muted recent rise. It is not clear to me that Obamacare, primarily an insurance program, has really attacked healthcare costs.
Both healthcare and education have a long way to go in terms of optimizing cost while raising quality/service. The answer is likely to be a far bigger spectrum of services and a great deal more "productization." To do this, it is very likely that one has to unleash reforms to the revenue and accountability model.
Just trying to make sure I understand. So, the Perry graph shows that prices have skyrocketed. But the Yglesias charts show that even with that price increase, health care as a percentage of GDP (or what we spend as a nation) has leveled off ... does that just mean our spending on things like groceries and laptops and natural disaster response or whatever has increased even more?
I recognize this is a misunderstanding on my part, I'm just not seeing how the Perry graph skyrockets in the same years the Yglesias chart flatlines.
Regardless, this was a great, informative read. Thanks!
Likely you are aware and Senator Warren established her reputation with her thesis in 2007, The Coming Collapse of the Middle Class (Jefferson Prize). As you suggest also, she observed that costs that Middle Class Americans are able to control are getting less expensive and those that they cannot control (housing, healthcare, education) become more expensive. https://youtu.be/akVL7QY0S8A?si=6fxp_qC0I8axtv8M
Interesting she didn’t note that the government-dominated sectors seem to be the ones most out of whack!
The progressive solution- to take large donations from the SEIU and NEA and allow them to write provisions of “Build Back Better” so that the NEA and “licensed” teachers will be running preschools and the SEIU will be running home health care, paid for by the federal government, doesn’t seem like a good one to me.
I am ever hopeful for a tech revolution in education (still waiting), but the interest groups running that sector aren’t so interested in cost efficiency, productivity and outcomes.
There have been a dozen tech revolutions in education. You can take advantage of any of them. Plus many old-fashioned inexpensive education options, such as community colleges and cheaper non-brand-name schools. The problem is that people don't want those things, they want the credential.
And the reason they want the credential is because we have a society that has made the credential into a requirement for vetting high-earning new knowledge employees. That isn't a problem that a tech revolution is necessarily going to fix.
I have no problem at all with kids going to school. I have a problem with the quality of the teaching, the poor use of tech and measurement tools and the lack of accountability.
You can teach a different way with different tools and still offer whatever credential you like.
Really appreciate the opportunity to update my priors!
There's definitely been some belt tightening at my hospital but we'll still survive. And we are seeing a number of hospitals close. It's good to control the costs though I'm worried whether high-consumers are scaling back or low-consumers are losing much of their access. Or are we just seeing better efficiency out of necessity?
I honestly don't know the answer but it's a good time to look into it. FWIW, I've long had an "abundance mindset" about the healthcare sector that's out me somewhat out of step with many of my colleagues. I don't think we've come close to supply meeting demand given the insane wait times I'm seeing. Maybe supply has become SO constrained that people are just losing access regardless of their insurance? I dunno.
It's impossible for me to look at these kinds of data and see a future in which these trends continue. The political and economic landscape of the U.S. has been irrevocably devastated; do past trends have any real use in predicting the future of a country in which our leaders are actively destroying the economy?
Right on. I started med school in '63. ( makes me a certifird "old fart" lol!) The medical technology explosion (think mri, new drugs and the rapid integration of genomics etc) make my med school days look very primitive indeed.
I'd like to see this logic applied to other labor intensive industries - childcare and construction for instance - that do not seem to be moderating their rate of cost increase. What makes these industries different from healthcare and education. It is hard to separate cost disease from increases in consumer expectations and quality of provision in all these spaces, but it doesnt seem likely that is making all the difference.
Years ago a network scientist named Samuel Arbesman wrote a book about this kind of thing. How there are facts that change constantly -- the weather -- so we know we need to check the latest. And there are facts that never change -- the names of the planets -- so we know we don't need to bother checking on them. But there's another category of fact he called "mesofacts" (a term which obviously failed to catch on) for facts that change over the span of a few decades and people seem poorly suited cognitively to deal with because nobody puts a Google Calendar reminder for 2035 to check the current consensus on Industrial Policy.
This is interesting and it parallels another point in evolution. For some animals, their environment changes so slowly over evolutionary time that it’s best to just wire in optimal behavior genetically as instinct. For other animals, their environment changes so quickly over evolutionary time that it’s best for evolution to wire in a learning ability so that each individual learns what behavior works best in the environment as they experience it. But there are some animals for whom the environment changes at a medium speed, and thus it’s best not to wire in behavior, but instead to wire in a propensity to learn culturally, so that individuals can use the optimized behaviors of the generations before them, rather than having to reinvent it all from scratch, or know it instinctually. The anthropologist Joe Henrich says that humans have this feature in his book The Secret of Our Success. Cultural learning turns out to be really powerful at developing traditions that optimize for all environments on earth, including developing superstitions and cultural practices that happen to help against local dangers without anyone knowing that this is why they do them.
Yes, Heinrich’s “The Secret of Our Success” is a fantastic book that is full of interesting observations. For those who have not yet read it, you might want to read a summary here:
https://techratchet.com/2020/01/02/book-review-the-secret-of-our-success-by-joseph-henrich/
Kenny, can you give an example of an animal that learns quicker than a human but does not share its knowledge with others of its species?
It’s not so much about not sharing knowledge - it’s rather that humans have a tendency to use social learning to blindly copy practices without understanding them, while many other species do social learning by seeing that another individual does something successful, and then figures out for themself how to be successful rather than blindly copying the model. Henrich gives some experimental examples, where a human child who sees an adult doing something weird before opening a box that has candy in it will often copy the weird thing as well as the action that opens the box, while a chimpanzee will just copy the parts that seem relevant to opening the box.
He argues that this blind copying is how humans learn to do long processes that are actually helpful even though no one knows why - you boil the cassava for many hours before eating it, because that’s what grandma always did, even though it seems soft enough to eat after a few minutes, but it turns out that if you only boil it a few minutes, then it has a low amount of cyanide that will build up in your body if you eat it this way for many years.
But I think there are also plenty of species like cats and squids that have lots of individual intelligence, but very little social learning.
I feel like this is this is the central insight of conservatism, and it’s genuinely a good insight. There is a class of things that work really well that we don’t fully understand *why* they work. Is this class is large and important, something like the conservative worldview emerges as an effective strategy.
Yes, Henrich’s work has helped me understand that!
You see a ton of friction around this with parenting where, to a first approximation, the current generation of parents are the first ones in millions of years to have any actual scientific/medical evidence on anything about child-rearing. And lots of grandparents react very poorly to things being done differently than they did it, as if "we've learnt better" means "you are a bad person".
(I just read a story this morning about grandparents who took a 21-month old toddler to the zoo without a car seat and didn't see anything wrong with that.)
Thanks for clarifying.
One unclear aspect of this interesting essay is whether it is written from the point of view of the average American household income or the median American household income. In other words, are we talking about households making $106,270 or $80,610? There's a great gap in disposable income between those two figures.
Also, I see a recurring insensitivity to past inflation. The idea that prices haven't gone up in the last five years doesn't mean that the pain has gone away. it just means that the thumb screws are no longer being tightened, after reaching the level of torture in the past. Economists seem to think that the pain of past inflation and (a continuing inflated status) evaporates instantaneously.
OTOH, it is refreshing to learn that both American higher education and American health care are somewhat responsive to market forces.
The median personal income is only around $42,000!
A person making $42,000 a year doesn't sound like they're in a position to buy much in the way of higher education or healthcare.
I gave numbers for household income.
I got those figures from Bing copilot. Here is what Google has to say on the topic.
In 2023, the median household income in the United States was $80,610, and the average household income was $135,700 according to FRED, the Federal Reserve Bank of St. Louis. The median household income represents the middle value, with half of households earning more and half earning less. The average, also known as the mean, is calculated by adding up all household incomes and dividing by the number of households.
Household vs personal.
On behalf of all economists, I admit to really believing that people hate the first derivative of the price level. :) It is sort of too bad that the don't. There ARE ways to reduce _inflation_ at low cost in terms of creating unemployment. There really is not any good way to reduce the price level.
Agreed, there is no way to good way to reduce the price level, but at least we can acknowledge that the shock and the pain doesn't go away quickly.
“I feel your pain” :) Biden should have been pantomiming reducing the price level, like a tax increase. Almost anything good could be claimed to be to “control prices.”
But here I’m getting out of my lane. I officially know nothing about politics.
Though being responsive to market forces is not all upside. In this case it means that 2 million fewer Americans have benefitted from a college education than 20 years ago because they've been priced out of it. And if, as appears likely, the fundamental divide between Trump and non-Trump voters is whether or not someone has been to college, then this market adjustment has actually been a disaster for progressive (or even rational) American politics. Fewer people getting a college education = more Trump supporters.
https://www.wsj.com/opinion/donald-trump-and-the-college-degree-divide-education-level-voting-patterns-political-parties-c26c4b02
This reasoning is why the democrats are going to keep losing. I AM a college graduate who voted for trump because the democrats refused to listen to my rational concerns about inflation and unchecked illegal immigration. They assumed that I couldn’t think for myself or that their trump hysteria would scare me. I was hoping to vote for a reasonable democrat until the party put Kamala Harris in, again assuming I would just blindly follow them. People without college degrees aren’t stupid or credulous; like me, they know their own interest, and it’s not the Democrats. And they love their country, which has not been my experience with college-educated democrats, who insist on emphasizing all our flaws to the point of wishing America didn’t exist. Most Americans look at elite democrats as sellouts at best and traitors at worst. That kind of self hatred does not win elections.
Who said anything about them being stupid? I said they're more likely to vote for Trump than people who go to college. I actually think it's probably because a college education is increasingly viewed as a wealth-based gating exercise which determines, from a fairly early age, what you'll end up earning. By the same token, just because you went to college doesn't mean you're not stupid. Clearly there's still a lot of work to do before the US becomes fluent in anti-elitism. Plutocracy masquerading as meritocracy has done enormous damage to social cohesion in both the US and the UK. But especially in the US.
Rich con men masquerading as "champions of the working class" have done equal, if not more damage. To the extent that you're an intelligent person angry at college-educated elites, that makes total sense to me, and I can even see voting for Trump as a way of burning that all down. But actually believing Trump has your interests at heart does, in fact, make you kind of stupid.
Yes, exactly my view. Believing Trump is going to help you is an insane leap of faith, unless you're a millionaire. All the evidence points in the opposite direction. A lot of Trump voters are going to lose healthcare cover, for example. And they voted for Trump knowing that would be the case. Why would someone do that?!
Do you feel good about your choice? That is, do you think the actual government you voted for is/is going to do a good job at addressing your concerns, or is that beside the point?
My opinion is that way too many kids go to college. It should be aimed at the top third or quarter of kids, as most kids are better off in less cerebral pursuits and other careers.
That said, the cash strapped kids can usually go to junior college for the first two years and then transfer to a local state school, preferably while living at home. There are then countless grants and assistance packages, loans and part time jobs that contribute to college costs.
I agree with all your points. What I meant by responsive to market forces is that the upward trend in college and medical costs seemed absolutely unstoppable, but the inability of the average person to afford them seems to have had at last a small moderating effect.
While I think a college education should be available to all who can benefit, I don't think it's the answer for everyone. I wrote a piece about the need for government commitment to free and supported lifelong job retraining here:
https://kathleenweber.substack.com/p/how-to-fight-trump-part-one
Yes, I agree. HE's sorting function needs to be moderated by genuine, and genuine respect and support for, forms of education and training available in later life.
It’s interesting to see a cognitive bias evidenced in the comments where nothing can ever be fixed or even get better. College had gotten significantly cheaper. No it hasn’t! No, here is the data, it has. No, I was told it was rising 20 years ago and nothing ever gets better to it’s still the way it was.
Well thanks to the Trump administration suspending student visas, and trying to expel current students from China, we can look forward to university’s tuition charges to start rising again, since the schools bring in much more revenue from international students who pay full tuition, don’t get financial aid, sports scholarships or other discounts and have extra fees tacked on as well. To make up that lost money, they will need to raise fees for other students.
Of course the bigger impact will be from the lost human capital and scientific (including medical research) that will be lost along with these students, who on average are better qualified by being selected from a pool bigger than the 4% of global population that is American. Most of my doctors have been immigrants, and they also serve rural areas with doctor shortages.
Of course the tariffs will also raise manufactured goods prices, so we can at least not blame the higher living costs only on services.
The huge jump in tuition in the 2000s occurred as the percentage of foreign students rose dramatically. The data seems to show that increasing the percentage of foreign students increases the pace of tuition hikes. If we were to revert to 2003 tuition levels (CPI-adjusted) and 2003 percentages of foreign students (still significant), I’d be very happy!
I hope these changes don’t actually last into the new academic year!
Great article. I really like when you focus on economic issues, as this is your strength.
There are two aspects that make up the perceived cost of a good or service. The price and how much I get for that price (quality). The rate on increase in healthcare spending is less than it was that is a numerical fact. My perception of what I get for the dollar of spending has definitely declined over the last 20 years. A standard half hour doctor visit has become 12 minutes, I used to be able to get an appointment to see a specialist in two weeks,now it can take six months and I got sent home from a hospital stay far sooner than I would in Czechia. On the other hand I just replaced my TV that was five years old. The new one is far superior and cost $200 less. There are quality adjustments in good CPI I’m not sure if how or if that is done with services. And don’t get me started when I’m paying for plumbers.
Yes- like making the package weight smaller!
That's been given the name shrinkflation.
Re the healthcare spending chart: given the ACA had a large Medicaid expansion, and Medicaid has low payment rates to providers, could that be what's driving the chart? Like a composition effect, where the addition of these low-paying customers is keeping the average down ?
Also may reflect how “prices” are calculated vs actual costs.
Let’s say a hospitals “list price” for a procedure is $10k but its average reimbursement (the real price, tracked in the data series) for it between private insurance and Medicare is around $4k.
Used to be that an uninsured person came in and was billed the $10k list price. Many didn’t pay the bill and debt was sold to collectors or written off. A few people actually paid the $10k. Maybe the true average payment the hospital receives for all the $10k bills it sends to the uninsured was $2 or $3k (a bit lower than the $4k Medicare/private average):
Now hospitals sign up the uninsured for Medicaid at the front door if the person qualifies. Instead of a bill for the list price of $10k the person gets charged (rather, the government gets charged) $3k. You could have a situation where the same person buys the same service and the hospital receives $3k net on average after each, but pre AHA that was billed as $10k and post AHA $3m. Because of the expansion you’d also have more people consuming the product at the 3k price, bringing down the average price.
Just a theory, That would really depend on whether the healthcare cost data series is tracking billed prices to the patient or the net received by the hospital after write offs. I have no idea.
In any case, great point about the expansion (and also more regular Medicaid sign ups) driving weighted average prices down
I think it measures total payments for all medical services in the country, as a fraction of GDP. If all you do is add spending, it can only make this number grow. This is different from something like “average wages per employee”, which can go down when unemployed people become employees, or “medical spending per insured person”, which could go down when uninsured people become insured.
If the Medicaid expansion is reducing this, it would have to be because Medicaid is paying for cheap medical services that prevent conditions that used to lead to expensive medical problems that got treated by emergency rooms.
Hmm, I see your point re that chart. I was responding more to the narrative provided, like "What’s happening is that although health care prices have still been going up, they’ve been going up more slowly than before." The Medicaid costs are more tightly capped, so I was wondering if those cost controls were doing a lot of the work in this expense slowdown.
What is the evidence that the flattening of health care costs was not due to improved productivity? After all, recent years have seen a decline in coronary artery revascularization procedures due to better control of hypercholesterolemia and placement of coronary artery stents by catheterization, a reduction in cesarean section deliveries, more limited resections of breast cancers, and less invasive surgery of every sort due to the use of endoscopy for procedures rather than open surgery. It is true that such innovations as CAR-T cell infusions as a treatment for cancer are very costly, but the trend toward less invasive surgery may very well have reduced the intensity of medical care and, hence, its cost.
This is an interesting theory, but like all interesting theories my objection is that we shouldn't have to speculate about it. The data is out there. Also unrelated to this, but the flat trends of Noah's chart don't appear to show up in actual medical cost trends, which all seem to be above inflation levels (if only just a little in the super-high inflation years) [1].
[1] https://www.pwc.com/us/en/industries/health-industries/library/behind-the-numbers.html
Hmm... I am not sure services in general are persistently high. I can think of many services where "productization" or "franchising" models have taken cost down. Rather GOVERNMENT connected services have been persistently high (health, education being at the top of list). A great deal of the reason is that the velocity of innovation has been retarded due to the revenue models controlled by government. Also, inventive models do not reward cost optimization. Interestingly, the COVID pandemic created an emergency situation which unleashed a number of innovations, and I wonder if that is the reason for the muted recent rise. It is not clear to me that Obamacare, primarily an insurance program, has really attacked healthcare costs.
Both healthcare and education have a long way to go in terms of optimizing cost while raising quality/service. The answer is likely to be a far bigger spectrum of services and a great deal more "productization." To do this, it is very likely that one has to unleash reforms to the revenue and accountability model.
Just trying to make sure I understand. So, the Perry graph shows that prices have skyrocketed. But the Yglesias charts show that even with that price increase, health care as a percentage of GDP (or what we spend as a nation) has leveled off ... does that just mean our spending on things like groceries and laptops and natural disaster response or whatever has increased even more?
I recognize this is a misunderstanding on my part, I'm just not seeing how the Perry graph skyrockets in the same years the Yglesias chart flatlines.
Regardless, this was a great, informative read. Thanks!
Likely you are aware and Senator Warren established her reputation with her thesis in 2007, The Coming Collapse of the Middle Class (Jefferson Prize). As you suggest also, she observed that costs that Middle Class Americans are able to control are getting less expensive and those that they cannot control (housing, healthcare, education) become more expensive. https://youtu.be/akVL7QY0S8A?si=6fxp_qC0I8axtv8M
Interesting she didn’t note that the government-dominated sectors seem to be the ones most out of whack!
The progressive solution- to take large donations from the SEIU and NEA and allow them to write provisions of “Build Back Better” so that the NEA and “licensed” teachers will be running preschools and the SEIU will be running home health care, paid for by the federal government, doesn’t seem like a good one to me.
I am ever hopeful for a tech revolution in education (still waiting), but the interest groups running that sector aren’t so interested in cost efficiency, productivity and outcomes.
There have been a dozen tech revolutions in education. You can take advantage of any of them. Plus many old-fashioned inexpensive education options, such as community colleges and cheaper non-brand-name schools. The problem is that people don't want those things, they want the credential.
And the reason they want the credential is because we have a society that has made the credential into a requirement for vetting high-earning new knowledge employees. That isn't a problem that a tech revolution is necessarily going to fix.
I have no problem at all with kids going to school. I have a problem with the quality of the teaching, the poor use of tech and measurement tools and the lack of accountability.
You can teach a different way with different tools and still offer whatever credential you like.
Really appreciate the opportunity to update my priors!
There's definitely been some belt tightening at my hospital but we'll still survive. And we are seeing a number of hospitals close. It's good to control the costs though I'm worried whether high-consumers are scaling back or low-consumers are losing much of their access. Or are we just seeing better efficiency out of necessity?
I honestly don't know the answer but it's a good time to look into it. FWIW, I've long had an "abundance mindset" about the healthcare sector that's out me somewhat out of step with many of my colleagues. I don't think we've come close to supply meeting demand given the insane wait times I'm seeing. Maybe supply has become SO constrained that people are just losing access regardless of their insurance? I dunno.
It's impossible for me to look at these kinds of data and see a future in which these trends continue. The political and economic landscape of the U.S. has been irrevocably devastated; do past trends have any real use in predicting the future of a country in which our leaders are actively destroying the economy?
Right on. I started med school in '63. ( makes me a certifird "old fart" lol!) The medical technology explosion (think mri, new drugs and the rapid integration of genomics etc) make my med school days look very primitive indeed.
I'd like to see this logic applied to other labor intensive industries - childcare and construction for instance - that do not seem to be moderating their rate of cost increase. What makes these industries different from healthcare and education. It is hard to separate cost disease from increases in consumer expectations and quality of provision in all these spaces, but it doesnt seem likely that is making all the difference.
High costs of goods Gas food rent housing and services how its it possible on 42ks