Random Critical Analysis on Health Care

The excellent Random Critical Analysis has a long blog post, really a short book, on why the conventional wisdom about health care, especially in the United States, is wrong. It’s a tour-de-force. Difficult to summarize but, as I see it, the key points are the following. (I also drawn on It’s still not the health care prices.)

1. Health care spending is well predicted, indeed caused, by income.

Notice that the United States doesn’t look unusual when income is measured at the household level, i.e. Actual Individual Consumption, which measures the value of the bundle consumed by households whether the bundle items are bought in the market or provided by governments or non-profits. (AIC also avoids some issues with GDP per capita when a country has lots of intellectual property and exports, e.g. Ireland).

2. The price of health care increases with income but at a slower rate than income.

As a result of the above:

3. The price of health care relative to income is lower in rich countries, including the United States.

Let that sink in, health care prices are lower relative to income in richer countries. Health care in the United States is cheaper relative to income than in Greece, for example.

Since spending is going up faster than income but prices are not it must be the case that quantities are also increasing with income.

4. The density of health care workers (number of workers) and intensity (what the workers do) increases with income.

RCA: Rich countries consume much more cutting edge health care technology (innovations).  For every 1% increase in real income, we find a 1-3% increase in organ transplant operations, a 1-2% increase in pacemaker and ICD implants, a 1-2% increase in the density of medical imaging/diagnostic technology, and likely similar patterns for all manner of other new technologies (e.g., insulin pumps, ADHD prescriptions, etc.).   Obviously, these indicators are just the tip of the iceberg. Still, where data of this sort are available, they tend to be highly consistent with extreme income elasticity (particularly newer, more expensive forms of health care).  In the main, costs rise because this technological change tends to requires a lot more people in hospitals and providers’ offices to deliver this increasingly complicated array of health care (surgical procedures, diagnostics, drugs, therapies, etc.).

A bottom line is that health care spending in the United States is not exceptional once we take US income into account.

RCA’s analysis is consistent with the Baumol effect and my analysis with Helland in Why Are the Prices So Damn High (we have some minor differences with RCA on physician incomes but neither of our analyses depend on that point). A big point is that RCA and Helland and I argue that the rising price sectors are not crowding out consumption of other goods. We can and are buying more of other goods even as we spend more on health care and education. Or, as RCA puts it:

…these trends indicate that the rising health share is robustly linked with a generally constant long-term increase in real consumption across essentially all other major consumption categories.

It is true that the United States has a convoluted payment system which results in absurd and enraging bills. Fixing the pricing system could generate more equity and efficiency but RCA’s analysis tells us that billions are at stake, not trillions. A corollary is that as other countries reach current US levels of income their health care spending will look more like the United States does today.

See RCA for much more.

Comments

This is a good analysis, but focused on inputs, not outputs. In other words, sure the US can afford to spend more money on health than other countries, but is it getting value for money for this extra expenditures? I have the suspicion that what actually matters for health are not medical things, but things like eating habits, exercise and so on.

Exactly. See Robin Hanson's analysis for how healthcare spending is wasteful, sometimes harmful, signaling: https://www.cato-unbound.org/2007/09/10/robin-hanson/cut-medicine-half

As you note, the real key to long lifespans is nutrition and lifestyle -- which is why Europe has higher race-adjusted lifespan even though they spend far less on "healthcare".

Interesting. You think it is the food but then must adjust for race to prove your point. Maybe Europe should feed their diversity better.

If you don't adjust for race, Europe does even better.

But by not adjusting for race, criticism of universal state health care can blame race for the poor results in the US. Conservatives need a free lunch when arguing that it's purely individual choice when health care is needed. Everyone know Stephen Hawking chose to be born black with ALS by eating a poor diet. But at least he chose to be born in the US so he would not be killed by UK death panels.

Your missing the point. Many of the health problems are genetic not diet based.

To be honest, as Asians and Hispanics make up a larger share of US population than US Blacks and are advantaged relative to non-Hispanic Whites, US probably gains a dividend over Northern Europe from its ethnic balance (if we assume that the ethnic gaps with the US have some fixed basis and if we thought that for Black disadvantage to Whites, it would be pretty inconsistent and arbitrary not to think for Hispanic and Asian advantage over Whites).

So it's not really an essential to the point.

(Clarifying, I'm not the same M as above, but I'm not really too bothered about anyone else using that pseudonym if they want).

I'd say more like on a national level, the real keys to high average lifespans is not dying from a vehicle accident, an overdose or murder. Eating to much is probably a small driver and exercise a moderate driver. Do the Europeans exersize more than USAers?

European cities tend to be older and therefore not so suited to automobiles, as well as being hillier. So Europeans spend more time hiking around on foot. Like San Francisco with narrower streets. Everyone gets more daily exercise just from walking around.

And like NYC, they tend to have good public transportation, leading to more people walking. And to the highest life expectancy, at least according to https://www.businessinsider.com/us-cities-where-people-live-the-longest-2018-1?r=DE&IR=T

"The paper suggests that low-income Americans live longest in big, dense cities, like New York; Santa Barbara, California; and Miami, Florida. High-income Americans live longest in Salt Lake City, Utah and Portland, Maine."

So poor people live longer when close to rich people.
When rich people keep poor people away, they live longer in smaller cities.

Alternatively, poor people live longer when their immediate environment compells them to exercise more. Rich people live longer when they live near more opportunites for outdoor exercise - and aand fewer opportunities to indulge in unhealthy lifestyles.

My guess is the latter is probably selection.

Big, dense cities generate higher incomes, so persons in a small city with a high income probably have stronger underlying traits than equivalent persons in a big city.

The former (poor live longer in large, dense cities) probably reflects denser urban communities having healthcare access advantages through salary premiums; people further away from hospitals don't go to hospitals.

Normalizing for local income distribution (income quartiles relative to city) would probably smooth allow us to look at these things.

I immediately went to life expectancy and found it increases about 1.5% per year, or half point less than real economic growth rate. If the quality was increasing the chart would not be linear, it would be convex up.

The life expectancy chart I used was he USA chart, an if you note the series, the correlation is mostly driven by the USA. What is really happening is the USA is fighting a law of diminishing returns. This verified by our lower growth rate, real growth dropping as federal medical debt piles up.

Hammering the point home. I picked off the most recent life expectancy. Looking at the percent growth in life expectancy we see ithe gain is dropping from the past. The dynamics are obvious, the USA is moving left on that chart. We get linearity in the chart above via the simple fact that diminishing returns is shrinking the US economy. It is further compounded because the lower income folks make more money sending medical personal to the USA in a losing battle. The result will be a clustering about the central tendency, and a massive reduction in the R squared. The other thing we notice is me, applying the Abstract Tree theory to correct a bunch of dufas statisticians.

1 75.52 0.683
2 74.55 1.284
3 73.58 1.301
4 72.59 1.345
5 71.6 1.363
6 70.62 1.368
7 69.63 1.401
8 68.64 1.421
9 67.64 1.456
10 66.65 1.463
11 65.66 1.485
12 64.66 1.522
13 63.67 1.531
14 62.68 1.554

Yeah, but our economy is built on this broken window fallacy. Americans get fat eating sugary, fatty foods, waste money on things that promise to lose the pounds from gym membership to supplements, spend more money on a nip and tuck, then buy insulin and other medicine to stave off chronic problems, some devices like a CPAP or motorized wheelchair to go along with that, and then to top it off buy expensive end of life health care. In the meanwhile, shareholders are enriched every step of the way from the fast food industry, weight loss supplements, medical devices, drug makers, etc. If everybody just ate healthy and exercised, then that's a lot of unspent dollars that failed to circulate in the economy.

People would still spend the money, just on different stuff.

Yeah, I'm not sure I understand the point in bringing up a known fallacy, and then treating it like you have no idea whatsoever why it's a fallacy.

My argument is America is not unique in this behavior. Conditional on well-measured income levels, the typical OECD country spends about as much money on health care as us and seems to deploy it very similarly.

In a separate blog post, I point out that health care is subject to rapidly diminishing returns. Countries that started closer to the frontier of medicine (or well-measured income levels) see much less rapid gains in life expectancy. Beyond this, American health outcomes are severely hampered by behavioral/lifestyle issues, such as obesity, homicide, car accidents, and drug abuse.

https://randomcriticalanalysis.com/2019/11/07/a-tale-of-two-covariates-why-owid-and-company-are-wrong-about-us-healthcare/

Not related to this thread, and i will read the whole blog. But one quick question I have is whether this is controlled for household age. I wonder if these findings are a proxy for the fact that people in richer countries tend to live longer. More old people means more health care spending closer to end of life.

No, I didn't control for household age in that particular analysis. But I've looked into this question, as have others, and there's very little to suggest age structure or other indicators of population health (e.g., diabetes, obesity, etc) are significant determinants of spending at the national level. Yes, age, obesity, diabetes, and more determine are fairly strong determinants of spending *within* countries, but these clearly aren't significant determinants of aggregate spending (even if it determines the distribution within each country).

https://randomcriticalanalysis.com/wp-content/uploads/2020/01/rcafdm_elderly_share_elasticity_v2.png

https://randomcriticalanalysis.com/wp-content/uploads/2020/01/rcafdm_diabetes_elasticity_2015.png

In a world where health care isn't a luxury good you might expect objective indicators of need to predict, but it's likely that countries adjust their intensity of care according to their willingness to spend (ultimately income levels) and need. As in, a country with higher need likely spreads the resources a little more thinly, but that standard of care is still much, much more intensive than how lower-income countries would treat the same sort of patient/encounter.

The American health care system is what provided the basis for the American opiate crisis, isn't it? While making a lot of money at the same time. Other countries seem not to have used their health care systems to actually decrease life expectancy..

Canada has an opiod crisis as well.

Purdue Pharma took advantage of NAFTA, one assumes.

Yep, so it seems. "Purdue Pharma is seeking an injunction to temporarily halt all litigation against the company in Canada, as it seeks to settle thousands of lawsuits in the United States over a deadly opioid epidemic."

Canada just cannot escape its neighbor's ability to generate world leading pharma products. "The British Columbia government launched a class-action lawsuit in 2018 on behalf of the federal, provincial and territorial governments, suing dozens of players in the opioid industry, including Purdue, for their role in an overdose crisis that has devastated communities across Canada." https://www.theglobeandmail.com/canada/article-purdue-pharma-asks-court-to-halt-all-litigation-against-it-in-canada/

Where RCA is great is in showing that the US is not an outlier in prices or an outlier in total expenditure relative to income. These are the two usual claims made by reformers. He also shows it's not an outlier in the dispersal of care across income categories (not all or much "extra procedures for rich guys").

That leaves whether the US is still systematically less effective at improving population wide measures of health and disease for a given level of expenditure, even considering burden on health for pre-existing health behaviours, and so needs reform. I think it is plausible it may very well be the case (RCA may dispute this).

You can also argue (as Alex T does) that the US billing structure is bizarre and its low collection rates often land individuals with medical bankruptcies and ridiculous high bills in a sort of idiotic random bingo, and that needs reform (even if the total prices are not high).

But those two are not the basis most pro-reformers argue for; they argue that the prices are too damn high and that the US spends too much relative to income, and that there are absolutely vast savings to be made by delivering almost the exact same patterns of care that the US currently delivers under a different healthcare system. (They claim, extraction, profiteering, and inefficiency on a procedure for procedure basis). And that's what RCA seems to skewer very effectively.

Even RCA agrees that American pharma prices are not comparable, but then seems to exclude pharma from any further discussion of health care costs.

Does it even matter it pharmaceuticals are not really a big part of the composition of overall costs?

This is another example of who cares about saving billions, that is just chump change.

prior, if you want to argue about saving billions, sure, but make the stakes clear and don't allow it to be elided with arguments you're saving people trillions. If you are indeed saving anything when switching and restructuring costs are consider.

The topic under discussion is about the top line of whether US expenditure as a whole is out of whack with income, living standards and consumption levels. If you want to argue that the US can save relatively small costs as part of the overall composition then do so honestly, do not goalpost shift and elide.

Where's Switzerland in the 1st figure? Was the X axis extent capped for cherry-picking?

More than that, notice the gap in several of those charts. Then keep in mind that many people in the U.S., unlike in those other countries, basically receive no routine health care for years at a stretch, which seems a perverse way to describe a rich country, or 'real income.'

And in the fourth chart about health spending, Germany seems to be missing too (though possibly CHE is Switzerland in some of the charts).

One has to respect this sort of elision, "Save perhaps for comparisons of pharmaceutical prices" This one however, is worth something other than respect, "It excludes things like pharmacies, insurance companies, drug and device manufacturers, and assorted middlemen."

So much more to explore, but why not end here by concluding that rarely does one see such an exceedingly thorough job of cherry picking.

And you have to love how easily one can make the joke that a billion here, a billion there, and at some point it ends up being real money.

Prior, how much does Germany spend on treating your autism ?

Why don’t you pop some Ritalin, get your weighted blanket and read the actual linked post?

"many people in the U.S., unlike in those other countries, basically receive no routine health care for years at a stretch"

That seems like it would make the U.S. more efficient. Why would you see the doctor if you are healthy? I thought the U.S. had a lot of people going to the doctor all the time for colds and things like that.

"Then keep in mind that many people in the U.S., unlike in those other countries, basically receive no routine health care for years at a stretch, which seems a perverse way to describe a rich country"

I haven't received any "health care" in several years. I don't see the problem.

"So much more to explore, but why not end here by concluding that rarely does one see such an exceedingly thorough job of cherry picking."

Anytime you analyze data across countries you are going to struggle to find data points for every single one on every single measure. I doubt you'd have this objection if it were an analysis in favor of your views. So it's an isolated demand for rigor.

If you haven't been to a dentist in several years, you might want to consider it.

Switzerland's adjusted household disposable income is also much lower the United States. The specific SNA transaction for Switzerland just wasn't (probably still isn't) available in the non-financial accounts (SNA) data for 2017 and beyond.

https://twitter.com/RCAFDM/status/1228409070895366147

Also, since I have this data handy, I posted a plot for 2015 that includes Switzerland.

https://rpubs.com/random_critical_analysis/income_and_health_spending_2015_includes_CHE

OK, the figure with 2015 data clarifies the issue with CH.

But, all countries moved to higher health expenditures in only 2 years? For example, US from below 10k to above 10k ? Norway from below 6k to above 6k? Not sure due to log scale.

The 2017 and 2015 versions of those plots are current PPPs (not adjusted for inflation). Although I make use of the constant prices, constant PPP series when I want to run time series or panel data analysis, the current series is likely to be a little more accurate for cross-sectional analysis. When averaged over a several year period, it's readily apparent that health spending is highly elasticity with respect to well-measured income.

https://i1.wp.com/randomcriticalanalysis.com/wp-content/uploads/2019/12/rcafdm_oecd_facet_delta_hce_delta_income.png?ssl=1

It's also evident that relative changes in income predict relative changes in health spending. When other-high income countries show real convergence in real incomes their health spending convergence on the United States and vice versa when their income declines in relative terms.

https://twitter.com/RCAFDM/status/1228409070895366147

It's also evident in the time series analysis for each OECD country that (1) health spending is highly elasticity and (2) the household perspective measures (AHDI and AIC) are much stronger predictors of health spending growth than GDP and its other components when the data are disaggregated. If you actually read my blog posts you can find plenty more analyses that address these sorts of questions.

That’s interesting, Switzerland is supposed to be one of only like 3 or 4 countries with higher median (adult equivalent) income than the US. Is this not really true, because the official statistics underestimate difference in cost of living or something? I know Switzerland is notoriously expensive.

Switzerland has less disposable income because as you guessed, basic universal expenditures like housing are much more expensive relative to income than the in the U.S.

Sounds from this summary like they are giving the US a pass for consuming much more healthcare on the basis of the US being richer. But wouldn’t you need to measure the effectiveness of all that treatment, in the form of better outcomes? That’s where things seem to break down—US is spending so much more so should be seeing better survival rates than other rich countries, and we just aren’t.

In that case, we’d be better off spending down to the level of those other countries and accepting the same outcomes, at least we’d have more left over to buy jet skis and pickup trucks and whatnot.

We don't talk about declining life expectancy in such exhaustive examinations of why the U.S. is really just a typical rich country.

Lengthening life expectancy means what, exactly? Sitting in a warehouse for geriatrics? Having your diaper changed by a lady that speaks English as a second language during a commercial in a Seinfeld re-run? Mumbling to a room mate of three years that you don't recognize? Everyone dies, refusing to accept that fact is a denial of biology. All life comes to an end, even yours.

... not dying prematurely from a drug overdose (heckuva of a job Purdue) or diabetes?

All available evidence suggests (1) other developed countries spend on health very similarly and deploy it very similar conditional on income levels and (2) health care is subject to rapidly diminishing returns. Consider, for example, that the likes of Norway and Luxembourg spend ~2x as much as Italy and Spain, but don't seem to live any longer (probably somewhat worse outcomes on average). Beyond this, there are clearly some idiosyncratic behavioral/lifestyle factors involved that have an outsized influence on outcomes amongst countries near the frontier of medicine.

https://randomcriticalanalysis.com/2019/11/07/a-tale-of-two-covariates-why-owid-and-company-are-wrong-about-us-healthcare/

"Health care spending is well predicted, indeed caused, by income": irrespective of the population's age distribution? That's rather odd.

Odd? Of course not, this is simply an attempt to convince a certain segment of Americans that there is no reason to change the American health care system, as really, it is just like every other country's.

Just more so.

Maybe, but that's certainly not all. I think another significant conclusion we can draw is that the USA's superior economic performance compared to the sclerotic social democracies of Europe over the past few decades means that healthcare costs are virtually guaranteed to be higher here, due to the deployment of cutting edge therapies and treatments that socialized healthcare systems simply can't afford to employ.

But much of the "cutting edge" stuff will be bogus: over-treatment of over-diagnosed conditions.
https://blogs.scientificamerican.com/cross-check/the-cancer-industry-hype-vs-reality/

Though I suppose all the advanced countries spend on nonsense such as mass screening for breast cancer and prostate cancer.
https://onlinelibrary.wiley.com/doi/full/10.1111/eci.13062

Progress sometimes requires experimentation.

But experimentation clearly needn't guarantee progress.

Ouch!

Percentage of the population 65 YO and above: USA16%, Germany 21%, Italy 23%, Japan 28%. https://data.worldbank.org/indicator/SP.POP.65UP.TO.ZS

Japan almost has the double of old people compared to the US. Old people doesn't get sick in Japan?

They do, but the Japanese are great at suffering in silence. In contrast to the US, where they get pumped with expensive stuff that by the way is causing side effects galore - liver failure, greater vulnerability of the cardiovascular system... At least they get to spend their last few years high on opioids.

With a 3% obesity rate, heart the number with heart failure is about a third that of Americans on a per capita basis and hospitalization due to heart failure patients over 65 is a major health care expense in the U.S.

It may seem odd to you, but there's very little evidence to suggest age, obesity, diabetes, or other indicators of population health needs is a systematic determinant of health spending. Things that determine spending *within* countries are generally poor determinants of spending *between* countries (and vice versa).

This is very consistent with health care being a luxury good. The willingness of a country to spend on health is determined by its average income level in the long run. Countries that spend a lot more on health mostly spend it by adopting increasingly more intensive, cutting edge health care. The same sort of patient with the same sort of condition is likely to be treated very differently in a rich country than a poor country (with a lot of intermediate variation between when averaged across all diseases/conditions). Conditional on income or willingness to spend, countries with greater objective need for health care can balance out their spending by spreading their resources a little more thinly than they would otherwise (which is still much more than would be found in lower-income, lower-spending countries)

Health care is quite a broad phrase. Does the standard of maternity care found in Europe (France being a leading example) lead to the advantages of a European lifestyle? There is much more to a health care system than hospitals or doctors, at least from the perspective of any European country.

This is what makes makes many of this discussion so hard to follow, many countries invest in the sort of health care that lead to a longer life, starting before birth.

This is what France does. https://www.expatica.com/fr/healthcare/womens-health/having-baby-france-maternity-leave-107664/ Does the U.S. have any equivalent at all? And if not, it would seem that French monies are better spent than in the U.S.

The United States invests more in preventative care in relative and real terms than most other OECD countries. I very much doubt the differences between countries are explained by such expenditures though as these sorts of things rarely have large effects (and they certainly don't save money on net!). If you're arguing benefits provided by France outside the health care system (child subsidies, maternal leave, etc) actually matters, I'd like to see evidence for this proposition and it should be priced into the cost of health care before we talk about how well "monies" (writ large) are being spent. More generally, as with general population health outcomes, it's very difficult to compare infant and maternal mortality statistics without accounting for confounding factors like obesity.

https://randomcriticalanalysis.com/2019/11/07/a-tale-of-two-covariates-why-owid-and-company-are-wrong-about-us-healthcare/#rcatoc-obesity-also-explains-much-for-infant-child-and-maternal-mortality

And apparently gets negative value for that spending.

Maybe the dominant factor is that healthcare costs skyrocket in the last few years of life, regardless of when that is. If treating a 50 year old with terminal cancer is about as expensive as treating an 80 year old, and end of life spending dwarfs the ‘chronic’ spending difference across age groups, the effect of life expectancy itself may be modest.

Sorry but if we compare this to all other indutries (health sector is actualy an industrie) expense should be falling in absolut termes on the long run not increasing. The only way to explain that it is not is because of state meddling and régulation. a suprise? no this is always the case.

People have to spend their money on something.

I normally buy 100 guns and 100 butters. If the price of guns falls by 50%, I will end up buying slightly more guns, but also slightly more butters (because I have more income leftover) even though butters have not changed in price.

There's a good chance the US could save a lot of money with little-to-no impact on outcomes if we forced people to spend less on health care. Good luck suggesting that, though.

I see others beat me to the punch. When I visit the US, what I am struck by is not how much Americans pay for health care but rather how little bang they seem to get for their buck.

Some US company plans are great, with nearly no employee copays, and the added benefit of no waiting times at your doctors'. Those large group plans are often well-negotiated by competent purchasers, meaning not particularly expensive to the employer either. Screw ya if you work at a small company or in the low wage sector, though.

Like I had to recently, you can take an old family member in for a suspected stroke, hove all the tests done, including MRI, in less than 24 hrs, and stay in a private room. Care is expensive, but quite good. My relatives from the EU are always surprised and the quickness and relative luxury of care here.

Can't speak for Europe but in Canada if you arrive at the hospital with a suspected stroke you will absolutely receive rapid care. Perhaps not a private room though.

My latest Canadian hospital experience was for a suspected broken finger. We saw a doctor, got x-rays, received our diagnosis, and left in about two hours total.

My first hospital experience as an adult in Canada involved going to the ER for shortness of breath and back pain. Again I saw a doctor within a couple hours, got x-rays, then was put under for surgery on the same day.

When I visit the US, what I am struck by is not how much Americans pay for health care but rather how little bang they seem to get for their buck.

How would you have an impression of that while visiting? I have visited many countries, but do not any firsthand impressions of their health care systems simply because neither I nor any of my traveling companions nor hosts have ever consumed any health-care while we were traveling. Are you especially sickly and/or accident prone?

The discussion about why US life expectancy is lower than in other developed countries is interesting. He blames homicide, car accidents, obesity, and drug overdoses. Clearly the first two are a matter of Americans disliking each other, and the last two of their disliking themselves. Does that mean that all the whining about drinking and smoking can now stop?

Put otherwise, why does the effect of so many people stopping smoking not show up in the figures? Or does abstention from smoking tend to make you fatter, or more drug-prone, or more suicidal, or more tetchy, or more impatient and disagreeable, thus cancelling the benefits?

> why does the effect of so many people stopping smoking not show up in the figures?

I mentioned this on my blog. The peak mortality effects of smoking lag by 2-3 decades. The United States is likely still suffering more than most from smoking because our population smoked much more intensely decades earlier. Those countries with higher *current* smoking rates will likely suffer higher smoking-related relative mortality than the US and other high-income countries that have moderated their smoking habits in another decade or two.

It means lung cancer that would have killed someone in three months is replaced, starting five years later, by dementia that kills someone over three years.

My guess is that obesity weighs more heavily on a population’s health than smoking.

Not yet, but give it time.

The US was the leading consumer of cigarettes per capita among developed nations up until 1983 (when Japan passed us). We dropped to third in 1989 (Switzerland) and dropped below much of Western Europe only in the 90s. Worse, we peaked much higher and had more pack-years going. Over the next few decades the burden of cigarettes on life expectancy, greater than any other activity in healthcare, will finally decline to levels more commonly seen in the rest of the world.

This will not be noticed because that is right about when the obesity wave in the US will be crossing over into prime cardiovascular death risk territory.

What I like about my last very high hospital bill that almost bankrupted me is that the free market will respond with new supply from eager entrants because the profits were so high (I should know!) which will bring the cost of the next bill down to a lower equilibrium price because all those providers will compete very hard for my business. I am so glad to be an American in this day in age.

No insurance?

He probably didn't fill out the charity care application they gave him.

Few people pay their hospital bill/charges in full. Insurers of course pay less. There is always negotiation room even for those uninsured. Hospitals actually collect about 25% of their billed charges, which of course are wildly inflated.

Right, and the vast majority of hospitals have what's called a charity care program, also, where, depending on your income, assets, household size, and the amount you were billed, you can get the hospital to write off somewhere usually between 20 and 100% of the charges. You have to fill out of a form, though, and present some pay stubs and whatnot, so a lot of people don't do it, and then complain when they start getting collection notices.

FYI- hospital debt goes to “soft collections” and they forgive the debt after a couple years if you continue to pay. Charity care is also used few and far between when income is way below federal poverty line.

Not always soft. My wife is in collection and reported to the scoring agencies for a few thousand on over $5M of billings. The hospital billed some charges wrong, and it was too late by the time they straightened it out. After over a year, we're still disputing it.

What a racket! You can also try the American Hospital Association for advocacy or your state’s Department of Managed Healthcare if insurance was involved in the billing mishap. So sorry

The new hospital probably couldn't get the certificate of need required by law.

This.
Among other artificial barriers to entry - regulation, occupational licensing.

Of our two hosts, I prefer how Mr Tabarrok presents studies. Much better than just an excerpt and link.

Never mind. This was a blog post not a study.

Lot's of questions about these charts. The comparison starts with "household disposable income", then adjusts for PPP. How did they calculate "disposable income"? I presume, this means after taxes.

Also, we've got, effectively, a one year comparison (at a time when the US dollar is at relatively high levels). How does this affect the PPP analysis? I'd like to see more years.

WIthout seeing all the underlying data and the methodology, this could be a case (or not) for (more) socialized health care. In other words, states with higher taxes have lower household disposable income. Those taxes, in part, are used to fund a greater portion of health care. If the spending is lower, does that mean those systems are more efficient? Or, does it mean they spend less and get less?

To merely say that the spending increases with "income" is a bit misleading and doesn't tell me much. And, as others have noted, spending more does not necessarily equate with getting more.

I've written several lengthy blog post wherein I dive into the details at length. I didn't create these measures out of whole cloth; they're standard NA-based measures proposed years ago by several notable left-leaning economists which have been adopted by the OECD, EuroStat, World Bank/ICP, and others.

https://ec.europa.eu/eurostat/documents/118025/118123/Fitoussi+Commission+report

Both the consumption (AIC) and income (AHDI) measures I use are after taxes and (comprehensive) transfers. These transfers very much include social transfers in-kind (STiK), like health, education, and social housing. These measures were specifically designed to satisfy the invariance principle so that the welfare states of Europe can be compared fairly with the likes of the United States. The PPPs also are also designed and weighted to reflect this broader consumption basket (AIC).

https://randomcriticalanalysis.com/2018/11/19/why-everything-you-know-about-healthcare-is-wrong-in-one-million-charts-a-response-to-noah-smith/

RCA, you are a hero. Keep being awsome

I agree.

Hey RCA, about the Fitoussi report table 1.1 and the X axis in figure 1. Take a high tax country like France that has a "free" health system. The disposable income in cash is corrected by adding health expenditures, right? Then, the Y axis is health expenditures and the X axis is disposable income + health expenditures. So, assume disposable income as D and health expeditures as E.

For high tax countries, the first figure is just a plot of D+E Vs E.

No, because I'm comparing the bottom line figures (AIC or AHDI), and these get netted out by transfers back to the household sector in the form of cash/near-cash transfers (which show up in regular disposable income) or social transfers in-kind (adjusted disposable income). The table is only an example to show how differences in health policy get balanced out, but similar adjustments are also applied for other STiK (e.g., education, social housing, etc.).

Taxes only matter here to the extent it reflects differences in fiscal policy or differences in collective consumption (military, transportation, etc.). France isn't appreciably more fiscally conservative than the United States, and these differences aren't driven by collective consumption either. Broadly comparable results are obtained if I regress this on final consumption (AIC+collective consumption) instead. Save perhaps for Norway, which is unusual with its temporary income flows (oil)/consumption smoothing, these differences have relatively little to do with the incremental contribution of government beyond that which we observe in the HH perspective (fiscal policy and collective consumption).

This has mostly to do with the relative size of the corporate sector (especially non-financial corporations), whose retained earnings are counted in GDP even when the corporations are ultimately owned by foreigners and the prices faced by residents of countries for their consumption expenditures. In an increasingly globalized world, where capital & corporate ownership is highly mobile, and NFCs have been differentially building up savings (mostly without investing it domestically), GDP is an increasingly weak proxy for the material circumstances of households.

https://twitter.com/RCAFDM/status/1229785366837760000

The United States is probably the ultimate beneficiary of a disproportionate share of these savings in NFC (tax avoidance behavior, in large part, I suspect) -- Ireland, Netherlands, and some others are notable examples of this. I'm not saying fiscal policy plays no role whatsoever for America, but it's likely to a modest one, and it doesn't much change the conclusion that our high health spending is driven by our high (adjusted) disposable incomes and is likely to have little to do with idiosyncratic features of our health care system.

Incidentally, in case you're concerned about some sort of mechanical backward causation, it's clear health spending lags changes in disposable income, and that there's basically no correlation between changes in the two the current period (according to CMS it takes about 7 years for health spending to fully reflect income change). It's unlikely health spending is somehow inflating this disposable income levels, *especially* in real terms (adjusting for prices).....

RCA, I appreciate all the work you've put into this analysis, but it seems like one of the major points that needs to be understood here is that the US is much wealthier than other traditional comparisons seem to indicate. When we look at median income per household or per capita GDP, the US is not 30-50% wealthier than basically every other western European country. Even in your own charts, Norway is ahead of the US in real GDP/capita, but well behind on household consumption. Is this really all just due to differential capital accumulation in Norwegian vs US corporations?

Thanks for pointing that out.

I've explained this at some length on my blog already. But for the benefits of those that can't be bothered to read lengthy explanations......

Most of the systematic differences between OECD countries can be adequately explained by *observable* net primary and secondary income flows, as in, dividends, rent, employee compensation, and remittances, which are collectively reflected in its effects on gross *national* disposable income, and the share of disposable income attributable to the non-financial corporate (NFC) sector, which is functionally identical to gross savings or retained earnings. Basically, globalization has weakened the link between real domestic production and the real incomes ultimately attributable to the actual residents (households) of a country.

https://pbs.twimg.com/media/ERKEezdW4AAoDiw?format=jpg&name=medium

These globalization factors also influence the apparent price level of GDP since it gets weighted to reflect the GDP expenditures, as opposed to the (comprehensive) consumption basket consumers are actually interested in.

https://pbs.twimg.com/media/ERETRZDXsAIA4O-?format=png&name=900x900

The combined effects of the shares of GDP (net income flows x-border +NFC savings) and the effects on the price level can lead to large discrepancies between the two measures. Ireland and Luxembourg are two particularly pronounced example, but similar issues are found in other OECD countries, even if somewhat less severely (with highly elastic expenditure categories a few percentage points in the independent variable can lead to large differences in fitted/predicted values).

Norway is something of a special case because it's basically the only high-income country in OECD that's heavily reliant on oil -- a temporary income source subject to highly volatile commodity prices. Because of the temporary and uncertain nature of these income flows, Norway has decided to engage in large consumption smoothing (although several other petrostates outside of the OECD appear to be behaving similarly). Even if the theoretical income of Norweigan households could theoretically be higher if they were to spend it all immediately, their permanent income levels are likely substantially lower than we would normally expect for a country conditional on its income level. Regardless, the government of Norway has taken the decision out of the hands of Norweigan households. These "offshore" profits, as they call it, aren't ending up in the disposable income flows of households (instead it's in Norway's sovereign wealth fund and largely being invested outside the country).

https://randomcriticalanalysis.com/2018/11/19/why-everything-you-know-about-healthcare-is-wrong-in-one-million-charts-a-response-to-noah-smith/#rcatoc-health-spending-is-eating-all-oecd-countries

(scroll down a few paras to where Norway is discussed)

This is still highly relevant here because the households' comprehensive consumption preferences refect the comprehensive real incomes available to households in the near term. The price level of consumption also tends to track with these aggregates (AIC, AHDI, etc). In short, the real income available to Norweigan households is substantially less than American households. Our expenditures go further in *real* terms and our households even have higher savings on average (within the sector).

https://twitter.com/RCAFDM/status/1228409070895366147

You need to just go read the Random Critical Analysis post. There is like 400 pages of info or "underlying data and the methodology" there. Sorry but RTFM is the only response you will get on this one. RCA ran every lead to the ground go read it.

The big difference at the national level is that in the US, people without any health insurance with a developing medical condition starting at working age get fleeced (and society says, rightly so), while anywhere else, that segment is mostly covered by charity or socialized contributions.

Anywhere else? A while back on this blog Cebu was cited as the most median city in the world.

What happens to the median Filipino who presents a clear case of appendicitis to the hospital, but who (if I understand the economic situation of the median Filipino) doesn't have cash on hand to cover the surgery? Ray?

The data in this study is OECD. So yes sorry, should have said in every other OECD country. As to the rest of the world: poor country doctors are orders of magnitude less expensive than in the OECD.

I don't think anyone says "rightly so" except insurance adjusters. I think most of the rest of us think the US has a deeply broken healthcare system, we just can't do anything about it because the population is split on increase or decrease government involvement as the solution. And while the population is in gridlock, the insurance and medical lobbies get to do whatever they want.

What kind of developing medical condition? If they are impoverished they get free healthcare and if not, heavily subsidized by the government- right?

And if they managed to not take their free/heavily subsidized insurance, then they can get the care and then just not pay- right?

I was thinking of people who start out middle class and don't take up insurance because, why, I am healthy and it costs money every month. Most of those will stay healthy and do fine. Some will end up with ridiculous healthcare debt levels. They're out of network and going to the wrong hospital hits them hard (the Keith Smith EconTalk episode explains this well)

So if you're a complete idiot and you turn down your subsidized insurance and then you get sick and in the brief period between getting sick and getting insurance you incur massive debt, then if they try to collect it you just declare bankruptcy and start over, right? Most younger middle class people have little wealth.

And in which State can you get free health insurance if sick and thus unable to work? Texas? Texas where they refuse to allow even low income workers to get health insurance?

Massachusetts is not the United States.

"As of July 2019, Indiana is the only other state to have implemented a work requirement waiver."

Anyone who is sick and on disability qualifies for Medicaid in any case.

American politics has gotten too weird and the internet too self-referential for me to follow anymore

Thank you for sharing this excellent analysis.

One constantly hears about “healthcare inflation”, but what is inflating is the quantity of healthcare; not the per unit costs of procedures and medications. Several common procedures (cataract surgery, coronary bypass grafting, and total hip arthroplasty) are significantly less costly in real terms compared to 30 years ago. And the outcomes of these procedures have much improved over that time span due to improved technology and techniques. Surgeons and hospitals have become dramatically more efficient in performing these interventions resulting in large reductions in operating time and post-operative stays.

The rest of the world has benefitted greatly from the fact that a very large percent of healthcare innovation takes place in the US. No other country comes close to the US in biotechnology, medical devices, or surgical innovation. It would be a pity if misinformed politicians destroyed this incredibly important and dynamic sector of our economy.

I am not an economist but I see some problems with the terms being used here. The random critical analysis post claims that it is discussing “real income”, which is not net taxes, but the OECD data on household income that it uses on its graphs *is* net taxes.

So I don’t think the post can rule out the explanation that if the government pays for health care via taxation, thus lowering household disposable income, health care costs are lower than if it doesn’t.

I may be misunderstanding this but would that really lead to much change in the underlying income variable?

It seems like you wouldn't want to consider incomes before *all* tax, but only before that part of tax which relates to healthcare costs.

I can't source any numbers but then if you assume extreme high tax cultures about 20% higher tax as income than the USA, then about 25% of that on healthcare, then pre-health expenditure income only 5% higher relative to the US? Does that really change very much considering you're on a logarithmic scale and this hits all non-USA countries roughly equally.

Well, Norway, for example, has a much higher per capita GDP than the US but a significantly lower income on these graphs, and a lot of that is due to health care spending.

Eh? Not really. The amount of GDP that is healthcare is not supersized or anything and doesn't make up much of the gap with consumption. As I understand it, oil exports do, and associated sovereign wealth. Like Saudi Arabia.

On the subject of why all the extra expenditure doesn't seem to be well rewarded, considering this diatribe about the cancer industry.

https://blogs.scientificamerican.com/cross-check/the-cancer-industry-hype-vs-reality/

The writer, John Horgan, quotes a 2017 study that only 10% of cancer patients can benefit from immunotherapy but that has climbed to 15% to 20% and one of the Nobel Laureates has said this treatment will expand to solid tumors this decade while increasing the percentage of those with a blood cancer who can benefit. Of course, it doesn't suit Horgan's polemic to include the positives in cancer research.

You weaken your interesting point by including "this treatment will expand to solid tumors this decade". Journalism in the future tense is worthless, even if it's a Nobelist indulging in the journalism.

Right, why should we care what Nobel Laureate and immunotherapy developer James Allison says about immunotherapy advances when we have John Horgan to tell us the entire enterprise is hype?

Horgan cites a 2006 study that concludes: "concluded that without reductions in smoking "there would have been virtually no reduction in overall cancer mortality in either men or women since the early 1990s." That sounds right except that was for an analysis of cancer from 1991 to 2003 and apparently Horgan couldn't be bothered to note that the NCI estimates that if you analyze cancer deaths from 1991 to 2016 instead of stopping at 2003, the decline in tobacco use accounts for half of the decline in deaths.

Horgan's article was published February 12, only three weeks after a potential breakthrough in imunotherapy to treat all cancers was reported on January 20, so maybe he missed that.

https://www.bbc.com/news/health-51182451

There are definitely plenty of overhyped areas in cancer research, and immunotherapy hasn’t yet lived up to its hype. But Horfan’s post seems rather out of date. Just in the pas 15-20 years there have been significant improvements for some cancers - especially blood cancers - just from developing new small molecule inhibitors, and many new treatments have far fewer side effects than traditional chemo agents.

Horgan seems to be committed to scientific pessimism across the board since the 90s, so I think he’s underselling what progress has been made.

Immunotherapy is a recent treatment, so I'm not sure who's hyped standard this treatment was supposed to live up to. I've read cancer researchers criticize 60 Minutes, The New York Times and BBC News as media outlets that have hyped cancer therapies but the researchers aren't responsible for that.

Mike's moved on from buying cool memes to buying boring comments? I guess his campaign is doing worse than I thought.

Yes, health care is a luxury good with massive diminishing returns.

Ok, billions, not trillions. Fair point. The richer we get the more we spend on health care, even if it's not that effective. Ok.

But why do we have to drive 600k+ people reach year into bankruptcy over health care costs? The argument over health care isn't really about cost, it's about how we pool risk.

"drive 600k+ people reach year into bankruptcy over health care costs"

I'm skeptical of that statistic

Make it 10x less, and it is still likely higher than the number of medical based bankruptcies in the entire EU, Japan, Canada, and Australia combined.

If you have a serious medical event and end up in the hospital for a year, you are still at significant economic risk anywhere in the world, even if your health care costs are completely covered -- because you probably are not working and bringing in income.

ᴀꜰʟᴀᴄ!

Not true:

https://www.fraserinstitute.org/article/medical-bankruptcy-myth

Of the bankruptcies I have seen among my patients, the median bill that drove them to bankruptcy was under $10,000. I suspect that their median bill was under $5,000.

Which tracks with studies showing that the average bill in those bankruptcies is typically in the low thousands, that most people have other debts greater than their medical debts, and that most of these folks were unlikely to be solvent in the long run even without large medical expenses.

A certain percentage of Americans live with little or no margin. When a bill, even something that is below the recommend level of savings by most financial experts, comes unexpectedly, we see disproportionate rates of bankruptcy.

The other dirty secret is that having a healthcare bill makes it a great time to declare an inevitable bankruptcy. Certainly many of my patients believe that judges are more lenient when you have medical bills and are sick.

Nothing substantive to add, just want to thank RCA for all the work he's done. I previously believed that the US was an outlier in terms of health expenditures, and, mostly due to his analyses, now tentatively lean in the other direction.

I still think our health care system is dystopian and awful, but focus should be more on its capriciousness and opacity and less on cost savings per se.

I second your thanks and also thank him for his comments here. Ample food for thought.

In summary: do I have this right? US health expenditure is a luxury good - it may be that much of it does little good, or even negative good, but the total reflects mainly the high average household income of the population.

(i) I'm not sure, though, that I yet know why Americans have shorter lifespans than comparable populations abroad.

(ii) Nor do I know the medical effect of the capricious nature of the system for meeting the costs of healthcare.

For national health outcome/mortality data to be very useful, it seems to me that it needs to be adjusted to reflect demographics. For example, how do the health outcomes of Minnesota Norwegian-Americans or California Japanese-Americans compare with outcomes in Norway and Japan? Is there a good source for this kind of data?

U.S. life expectancy increased in 2018 to 78.7 years, which is close to where it was in 2014 at 78.9 years.

Russia's life expectancy in 1990 was 68.9 years and fell to 64.7 years in 1995 and is now about 72.0 years.

https://jamanetwork.com/journals/jama/article-abstract/2756187

Importance US life expectancy has not kept pace with that of other wealthy countries and is now decreasing.

The U.S. has not beem simply getting less result for its money, its health care system is leading to a reduction in life expectancy through its opiate dispensing practices.

The article does not include the announcement by the CDC last month that life expectancy increased in 2018.

"The jump in longevity comes as deaths from opioid overdoses dropped for the first time in 28 years, as did deaths from six of the 10 leading causes." (U.S. News)

You forgot a bit of context. "U.S. life expectancy reached 78.7 years in 2018, which is an increase of 0.1 years (about a month on average) compared to the life expectancy of 78.6 years in 2017. That means an infant born in 2018 is expected to live 78 years and eight months, factoring in current death trends and other issues. Despite the uptick, U.S. life expectancy is still lower compared to the peak of 78.9 years that Americans reached in 2014, however."

That's right, we still are not back to the 2014 number, so an increase in relative, not absolute terms.

Huh? I stated this was still a little lower than 2014.

I'm not at all worried about the longer trend in life expectancy. Harvard CRISPR geneticist George Church was interviewed on 60 Minutes a few weeks ago:

Scott Pelley: So is this gene editing to achieve age reversal?

George Church: This is adding genes. So, it's not really editing genes. It's, the gene function is going down, and so we're boosting it back up by putting in extra copies of the genes.

Scott Pelley: What's the time horizon on age reversal in humans?

George Church: That's in clinical trials right now in dogs. And so, that veterinary product might be a couple years away and then that takes another ten years to get through the human clinical trials.

You were going on for months about some trials of NR that were supposed to conclude in 2019. Any word on those?

In addition to 1,000 mg a day of NR lowering systolic blood pressure by 10 points after 6 weeks in those with pre-hypertension (no decline if notrmal)...

An accidental discovery was that 500 mg of NR reduced fatty liver by 10% to 20% after several weeks of use so better trials are already underway for that.

1000 mg of NR + 200 mg of pterostilbine slightly improved the condition of ALS patients after 4 months and at at the end of the 12 month trial, thereby outperforming the best drug that only halts the slide 30% during the same time period.

Elysium released its favorable trial results of the same dose of NR/pterostilbine on a group of patients with acute kidney injury to the FDA and so have been granted permission to apply that as a drug. As far as I know, that still hasn't been published.

Either this month or next month two tiny trials of heart failure patients should be published and in late summer/early fall a larger trial of 30 heart failure patients taking 2,000 mg of NR a day over a few weeks should be published.

A cognitive NR study is likely to be published late this year or more likely early next year.

its been like independence day the movie, i'm the alien controlling the president and your guy is still shooting me trying to free the presidents, release me, no peace (i'm just kidding), the power centers force the presidents on me, a give it to greenspan kind of thing, a good economy almost fixes everything.

I liked the fed raising interest rates to force the govt to cut spending or raise taxes, hitting 5 pct growth would've been a good precedent, then we could've expected 3pct growth for at least 5 10 years, most of the growth is govt minimum wage increases, and the increases come from the defense state a Americas following china nationalizing hiring and using advanced weapons to force wage increases, it looks ok for now, but it could be really bad in 5 10 years, so the 5 pct growth and fed forcing congress to raise taxes or cut spending would've been like being Switzerland instead of Venezuela, but I guess japan is always the debt rollover democrats model, a plateaued economy, Americas 3rd lost decade started if you look at median household income, more singles but the singles could be a symptom of something too

don't forget the stock markets are dependent on forcing money via satellite into the markets and sabotaging other places govts and economies, and the wages and hiring have been forced up from the state, America didn't become great America became china.

yes an optimist, but it wasn't bad in seattle in 2005 or nyc or san Francisco. remember when the minimum wage was 5 dollars an hour free market fundamentalism was the right way and the American blue city in the American blue state was still a nice and affordable place to live? wasn't long ago and it was arguably a better country to live in and a easier rise to middle class and above socioeconomic status, I doubt the attack asia for money way is the right way to go from best place to live to stay one of the best places to live, feels like soviets trying soviet union and Chinese communism

now tell cass and the neocons to go fix gersons lobotomy

Have always loved this series of posts. Also incredibly depressing though. RCA makes it about as obvious as one can in a blog post that several major points about US health-care are completely wrong, but these points continue to dominate national debates. On a more micro-level, RCA went back and forth with Noah Smith on some of these points once (Why Noah Smith is wrong about health-care in 1,000,000 charts), and Noah basically rejected it all in one tweet that amounted to "No. I think something different". It's an irritating world.

The beauty of the American health care system is that we do have significant free market forces in our system. And I think it makes sense that productive Americans should get higher quality health care as long as everyone has access to basic health care. So for Americans 18-65 years of age we pay for health care with what amounts to a VAT that is controlled by employers and their respective states. So every product or service purchased in America includes the costs of someone else’s health care. And the prices of consumer good and services is not more expensive in America so that means the cost of health care is not too high.

Cause or effect?

Might this be explained by providers charging what they can get away with, in a not very competitive, quasi-captive market for non-discretionary services?

Under that explanation, I would expect cost growth to track income growth, perhaps with a modest lag.

The curve not only flattens with diminishing returns, but it particularly flatters in less corrupt political environments

oops that should be to asdf

So we've got a graph with a cluster of dots where the developing countries are, a cluster of dots where the developed countries are, and then one single credible dot way out in the upper right corner called the USA.

And we draw a straight line through these three and say "its inevitable that people spend more on healthcare as AIC goes up.

But is it?

I'll take it for granted that the line between the developing and developed clusters is probably legit. It makes sense too.

But the extension of the line to the USA is speculation based on a single data point. Are we sure that when Germans or French will spend as much as the USA when its AIC is as high? I don't think there is enough data to make that prediction.

If the additional healthcare being purchased between where Germany is and where the USA is is subject to "highly diminishing returns", why should we believe that it will go that way (or has to go that way). Why would these other countries dump all this extra wealth into the equivalent of an incinerator called "virtually worthless additional healthcare intensity".

Couldn't they end up spending the wealth in other ways. They could retire early. Spend it on the environment. Spend it on consumer goods. You name it. I can think of a lot of things I would like to buy if my healthcare budget were cut in half.

Handwaving by saying its a "luxury good" is lazy. When food, flatscreens, or whatever else got cheaper we bought less of it.

Is there a particular reason that money HAD to go into healthcare? Of course it didn't have to. And its debatable if it should have, given that it doesn't appear to have done any good. This is like saying that as a society gets richer the money HAS to get spent bidding up fine art. No it doesn't.

Given that its not a free market and most people seem to be price and utilization "takers" rather then active participants in a market expressing preferences, whose to say its what they want or need.

Short Version:

There is no reason to believe that straight line couldn't or shouldn't start to curve flatter when you start to hit large diminishing returns to additional health expenditure.

That’s a gorgeous correlation!

You can't predict the future; it's entirely possible that should German household income grow to US sizes they'll notice long before then that additional spending provides decreasing margin returns and there are large residuals of spending to life expectancy. It's also possible that some smart government comes into the USA and radically imposes cost cutting measures that restrain the market and government spending on healthcare.

The question is whether the US has behaved and whether it allocates healthcare in a way that basically fits international patterns. Not whether a sufficiently smart regime could not deviate from those patterns in the future without really incurring much of a real cost in terms of mortality.

However, all the political and social pressures to me seem to indicate that as economies grow, others probably won't behave much differently.

Anecdotally, every time I bring up in UK politics discussions that additional spending beyond that was "lost" to the NHS by the Tories restraining the pre-Recession Era probably would lead to very little reduction in mortality or extension of life expectancy, it basically gets zero traction. No one cares that additional spending gives little benefit in life expectancy - the thrust of the counter argument from virtually everyone is we should do it anyway, and if we don't do it anyway, we are responsible for "murdering" the people who would have shorter lives if we didn't, no matter how high the cost:benefit.

I can't imagine it's too much different in Germany, France, etc.

For the USA to be "in line" we would need other significant (large non-unique countries) data points with the same AIC and health expenditure as the USA. We don't. We have a single dot, and we assume the line stays straight even as AIC goes up almost double based on that one dot.

Not good enough. And if there were other dots, it could just prove that we have an "inevitable" inefficient sociopolitical equilibrium, not that its a good thing.

The US is in line with the trend which exists in the range Norway to Latvia.

The trend is not dependent on one data point at the frontier.

It's more parsimonious to conclude that the relationship generally would extend beyond Latvia->Norway in essentially the same pattern than to assume that it would change with far more datapoints, and that therefore the US is still an outlier.

He gives time series figures that answers your question directly. It shows that as incomes grow, countries track over exactly the same trajectory over time (with a few bumps along the way that average out.)

It's obviously quite dumb for an Amerikan not to seek health care in Mexico, especially for citizens near the border.

In this case, wouldn't increasing inequality have a meaningful impact on how we think about healthcare - i.e. it stays relatively affordable for the modal worker but the left tail of the income distribution suffers worse as it is dragged along into higher costs.

I believe RCA tries some methods to look at distributional inequality of healthcare and doesn't find much of an indication that it is much greater than elsewhere.

Measures of US life expectancy inequality do seem modestly higher than other economies - https://ourworldindata.org/health-inequality

But it's not clear this is necessarily due to the healthcare prices - the gap goes back to the '50s and the US has long had more diversity of incomes and lifestyles than is the case in Europe.

Sorry, "than is the case in Europe" = "than is the case in any European country". Obviously Romania to Poland to Spain to Scotland to Norway to Sicily could be said to have at least as much or more diversity in lifestyles and incomes than the USA.

The left tail of the income distribution gets free healthcare though, right?

America's high prices makes sense for Doctor/nurse-provided services (because Baumol), but not for manufactured medical goods (drugs, bags of saline, etc)--we don't pay more for the same car, iPhone, or pair of Jeans than they do in Europe.

If one looks at percent of GDP spent on health care, US is at a bit over 17%. Its main rivals in this measure are Marshall Islands (actually ahead of US, only one), Sierra Leone, Tuvalu, and Micronesia. The highest percent for another high income nation is a bit over 12% for Switzerland. I think this makes mincemeat of the argument in this paper, not to mention the awful outcomes US gets for its healthcare in terms of being in the 30s on life expectancy and infant mortality rates.

More than 100 comments before someone pointed out this glaring blind spot. Many European countries simply spend 50% less on health care, without being 50% poorer.

This has got to be a parody. The whole point the first 5 seconds of reading RCA's post or Alex T's coverage is that GDP/capita relates poorly to Healthcare Expenditure per capita, and that relating to consumption / household income both A) logically relates more clearly to resources which can be diverted to healthcare expenditure, B) is actually a much stronger statistical relationship. No one has that little reading comprehension or intelligence.

Barkley, not a criticism of just you, but many in this discussion: "not to mention the awful outcomes US gets for its healthcare in terms of being in the 30s on life expectancy and infant mortality rates."

Life expectancy is not a measure of your health system. This has been pointed out ever since arguments on healthcare have existed. Americans die because a number of risky behaviors that have nothing to do with health care, and, when removed, boost our life expectancy above all the other countries. The infant mortality numbers also have problems, namely we try to save A LOT more preemies and we define live births differently.

Anyone making these arguments just does not have an idea what the argument even is.

TMC,

Just checked several sources. Your argument on infant mortality more relevant (and I was aware of it and almost did not mention it because I knew some jerk would show up making an assinine argument like I "have no what the argument even is"). But even for infant mortality, the higher rare of premies in the US and some well-known differences in reporting do not make the poor performance of the US go away.

Also, behavior explains part of the US story (especially our recent actual declines in life expectancy, clearly tied to our opioid epidemic). But that does not explain all of it.

Other explanations include widely varying availability of quality care tied to our lack of universal health insurance and our high (and rising) inequality.

Note that in 1980 our life expectancy numbers were about the same as the other high income nations, and our health care spending was also about the same as theirs, yet another problem with this idiot study posted here.

Do you also include the lower rate of abortion for fetal anomolies? Down's, Edward's, Patau's, and all the other chromosomal abnormalities all have drastically higher infant mortality rates even at term. In the US we abort maybe half of these kids. In Iceland it is basically 100% and in many other European countries, it is over 90%. Terminating unhealthy kids before birth does a remarkable job at lower infant mortality but is not an option for America unless you plan to force women to get abortions who choose not to have them for religious or ethical reasons.

Then of course, there is the high risk behavior of Americans. The most common cause of maternal death during the partum window is homicide. Beyond that car accidents, obesity, and drug use all contribute to both maternal and neonatal demise.

As far as your life expectancy data. Ehh whatever. Yeah, some places have seen the gap grow (or even reverse) since 1980. On the other hand, the single biggest factor in that is that until 1980 the US population was the greatest consumer of cigarettes per capita (and was only surpassed by some European countries in the 90s), almost like a bunch of people in Europe grew up too poor to smoke in the 40s and 50s. In any event the gap has not been consistent and in many cases (e.g. Canada) preceded the adoption of single payer.

Frankly, at the end of the day, Kaiser and every other reputable source, estimates that health care determines about 25% of life expectancy. If we wish to ascribe 100% of the changes in life expectancy to healthcare, the US system is not just worse, it is not just growing worse, but it is an order of magnitude worse and doing so at an accelerating rate. It also has continued to get worse post-ACA.

Funnily enough, this sort of analysis suggests that whatever has happened in the Trump years has slowed the degradation of American healthcare. Me, I think it kinda indicts the whole silliness of doing an apples to oranges comparison and then declaring that 100% of the difference is due to policy controlling 25% of the inputs, policies that have grown closer over the decades.

Because, ultimately, the US has moved closer to single payer healthcare with each new policy iteration, particularly when you include changes in state level policy. Medicare. EMTALA. CHIP. Medicare part D. Romneycare. Community health ratings. Medicaid expansion. Obamacare. Looking solely at policy changes here and abroad since 1980 suggest that the amount of policy difference has decreased, so why exactly does the smaller, remaining difference explain an increased difference in outcomes?

A much simpler answer is that the things we know drive decrease life expectancy, like obesity, abortion rates, etc. have been growing more divergent for 40 years and that is the primary move for changes in mortality.

Barkley, check out the 'Infant Mortality and Life Expectancy asIndicators of Health Care Quality' section here: https://energycommerce.house.gov/sites/democrats.energycommerce.house.gov/files/documents/Testimony-Atlas-UniversalCoverage_121019.pdf

Also, your statement '(and I was aware of it and almost did not mention it because I knew some jerk would show up making an assinine argument like I "have no what the argument even is").' is nonsensical. If you knew you were writing BS why did you write it? And why am I a jerk for calling you out on your lie?

Look, any serious discussion of these issues involves many caveats abd lots of details. You showed you are a jerk by oversimplifying going the other way, as if all this poor apparent health outcomes in the US are simply entirely due to the couple of issues you brought up. They are not. You are wrong, as well as a jerk.

Charles Phelps has been publishing similar estimates for several decades. His estimates on total expenditures show an elasticity of about 1.3 and the US slightly above predicted. Here the elasticity is 1.8 and the US not particularly high relative to prediction.

"The price of health care increases with income but at a slower rate than income."

So, the price of health care to Rush Limbaugh is much higher than the price for a Floridian earning food service wages with 1% of Limbaugh's income

Even when the delivery cost of doctors, hospitals, radiation, drugs, etc. are identical?

The price in the UK, EU, Japan, Canada, et al is tied to income which is tied to consumption, because the price is heavily driven by VAT (consumption/income), income taxes, means tested premiums/copays.

Costs are not tied to price for most of health care, just as the costs of road are not tied to income. Eg, road costs for rural America are much higher per capita than for urban America. Big government leftist socialism reduced the price of traveling in rural America way below costs. This is not true in most of Africa, for example, so, the price of transportation in Africa to at least half the population is so high they are cut off from the global economy.

"Free markets" require prices mostly equal labor costs, labor costs being the only true cost.

Opportunity cost like dying from infection because the price of health care access is the far higher than your income, a bit higher than your income and credit. In the US, EMTALA dictates credit far above what income allows in a free market for a subset of health care costs, ensuring socialist wealth redistribution. Ie, the price is certainly to be far below costs for most patients.

I grew up when economists knew costs were not price were not value.

The movement to try to force them to be the same for select cases in order to enrich the rich even more has forced economic growth to depend on exponentially increase in debt that will always require big government socialist wealth redistribution:
The United States Constitution (Article 1, Section 8, Clause 4) authorizes Congress to enact "uniform Laws on the subject of Bankruptcies throughout the United States".

Bankruptcy takes money from those who had it and lent it in some fashion, and gives it to those who borrowed spent other people's money - wealth redistribution.

Interesting which conclusion one draws. 1) US is not an outlier so no need to increase subsidies to health care insurance (and maybe reductions wouldn't hurt). 2) US is not an outlier so subsidies to health care insurance have not been harmful so no need to reduce them (and maybe increasing them wouldn't hurt).

Lunatic analysis. Doesn't in any way pass the smell test. It stinks and is smoke (toxic smoke, in fact) and mirrors.
Median income anihilates the entire nonsense of this. Result and reality amply indicate that the USA is off the charts in its extremely high healthcare cost level.

Alex always seems to be justifying health care spending, without looking at the inefficiencies in the system. Russ Roberts on EconTalk has had several guests who point out just how inefficient the medical care system is and how there is no proper market in health care.

Comments for this post are closed