SELF-RECOMMENDING!

The Amish Health Care System

I.

Amish people spend only a fifth as much as you do on health care, and their health is fine. What can we learn from them?

A reminder: the Amish are a German religious sect who immigrated to colonial America. Most of them live apart from ordinary Americans (who they call “the English”) in rural communities in Pennsylvania and Ohio. They’re famous for their low-tech way of life, generally avoiding anything invented after the 1700s. But this isn’t absolute; they are willing to accept technology they see as a net positive. Modern medicine is in this category. When the Amish get seriously ill, they will go to modern doctors and accept modern treatments.

The Muslims claim Mohammed was the last of the prophets, and that after his death God stopped advising earthly religions. But sometimes modern faiths will make a decision so inspired that it could only have come from divine revelation. This is how I feel about the Amish belief that health insurance companies are evil, and that good Christians must have no traffic with them.

And Deists believe that God is like a watchmaker, an artisan who built the world but does not act upon it. But by some miracle, the US government played along and granted the Amish exemptions from all the usual health care laws. They don’t have to pay Medicare taxes or social security. They aren’t included in the Obamacare mandate. They can share health care costs the way they want, ignoring any regulations to the contrary. They are genuinely on their own.

They’ve ended up with a simple system based on church aid. Everyone pays tithes to their congregation (though they don’t call it that). The churches meet in houses and have volunteer leaders, so expenses minimal. Most of the money goes to “alms” which the bishop distributes to members in need. This replaces the social safety net, including health insurance. Most Amish go their entire life without needing anything else.

About a third of Amish are part of a more formal insurance-like institution called Amish Hospital Aid. Individuals and families pay a fixed fee to the organization, which is not-for-profit and run by an unpaid board of all-male elders. If they need hospital care, AHA will pay for it. How does this interact with the church-based system? Rohrer and Dundes, my source for most of this post, say that it’s mostly better-off Amish who use AHA. Their wealth is tied up in their farmland, so it’s not like they can use it to pay hospital bills. But they would feel guilty asking their church to give them alms meant for the poor. AHA helps protect their dignity and keep church funds for those who need them most.

How well does this system work?

The Amish outperform the English on every measured health outcome. 65% of Amish rate their health as excellent or very good, compared to 58% of English. Diabetes rates are 2% vs. 8%, heart attack rates are 1% vs. 6%, high blood pressure is 11% vs. 31%. Amish people go to the hospital about a quarter as often as English people, and this difference is consistent across various categories of illness (the big exception is pregnancy-related issues – most Amish women have five to ten children). This is noticeable enough that lots of health magazines have articles on The Health Secrets of the Amish and Amish Secrets That Will Add Years To Your Life. As far as I can tell, most of the secret is spending your whole life outside doing strenuous agricultural labor, plus being at a tech level two centuries too early for fast food.

But Amish people also die earlier. Lots of old studies say the opposite – for example, this one finds Amish people live longer than matched Framingham Heart Study participants. But things have changed since Framingham. The Amish have had a life expectancy in the low 70s since colonial times, when the rest of us were dying at 40 or 50. Since then, Amish life expectancy has stayed the same, and English life expectancy has improved to the high 70s. The most recent Amish estimates I have still say low seventies, so I think we are beating them now.

If they’re healthier, why is their life expectancy lower? Possibly they are less interested in prolonging life than we are. R&D write:

Amish people are more willing to stop interventions earlier and resist invasive therapies than the general population because, while they long for healing, they also have a profound respect for God’s will. This means taking modest steps toward healing sick bodies, giving preference to natural remedies, setting common-sense limits, and believing that in the end their bodies are in God’s hands.

The Amish health care system has an easier job than ours does. It has to take care of people who are generally healthy and less interested in extreme end-of-life care. It also supports a younger population – because Amish families have five to ten children, the demographics are weighted to younger people. All of these make its job a little bit simpler, and we should keep that in mind for the following sections.

How much do the Amish pay for health care? This is easy to answer for Amish Hospital Aid, much harder for the church system.

Amish Hospital Aid charges $125 monthly per individual or $250 monthly per family (remember, Amish families can easily be ten people). Average US health insurance costs $411 monthly per individual (Obamacare policies) or $558 monthly per individual (employer sponsored plan; employers pay most of this). I’m not going to bother comparing family plans because the definition of “family” matters a lot here. On the surface, it looks like the English spend about 4x as much as the Amish do.

But US plans include many more services than AHA, which covers catastrophic hospital admissions only. The government bans most Americans from buying plans like this; they believe it’s not enough to count as real coverage. The cheapest legal US health plan varies by age and location, but when I take my real age and pretend that I live near Amish country, the government offers me a $219/month policy on Obamacare. This is only a little higher than what the Amish get, and probably includes more services. So here it seems like the Amish don’t have much of an efficiency advantage. They just make a different tradeoff. It’s probably the right tradeoff for them, given their healthier lifestyle.

But remember, only a third of Amish use AHA. The rest use a church-based system? How does that come out?

It’s hard to tell. Nobody agrees on how much Amish tithe their churches, maybe because different Amish churches have different practices. R&D suggest families tithe 10% of income, this article on church-based insurances says a flat $100/month fee, and this “Ask The Amish” column says that churches have twice-yearly occasions where they ask for donations in secret and nobody is obligated to give any particular amount (“often husbands and wives won’t even know how much the other is giving.”) So it’s a mess, and even knowing the exact per-Amish donation wouldn’t help, because church alms cover not just health insurance but the entire social safety net; the amount that goes to health care probably varies by congregation and circumstance.

A few people try to estimate Amish health spending directly. This ABC story says $5 million total for all 30,000 Amish in Lancaster County, but they give no source, and it’s absurdly low. This QZ story quotes Amish health elder Marvin Wengerd as saying $20 – $30 million total for Lancaster County, which would suggest health spending of between $600-$1000 per person. This sounds potentially in keeping with some of the other estimates. A $100 per month tithe would be $1200 per year – if half of that goes to non-health social services, that implies $600 for health. The average Amish family earns about $50K (the same as the average English family, somehow!) so a 10% tithe would be $5000 per year, but since the average Amish family size is seven children, that comes out to about $600 per person again. So several estimates seem to agree on between $600 and $1000 per person.

One possible issue with this number: does Wengerd know how much Amish spend out of pocket? Or does his number just represent the amount that the official communal Amish health system spends? I’m not sure, but taking his words literally it’s total Amish spending, so I am going to assume it’s the intended meaning. And since the Amish rarely see doctors for minor things, probably their communal spending is a big chunk of their total.

[Update: an SSC reader is able to contact his brother, a Mennonite deacon, for better numbers. He says that their church spends an average of $2000 per person (including out of pocket).]

How does this compare to the US as a whole? The National Center For Health Statistics says that the average American spends $11,000 on health care. This suggests that the average American spends between five and ten times more on health care than the average Amish person.

How do the Amish keep costs so low? R&D (plus a few other sources) identify some key strategies.

First, the Amish community bargains collectively with providers to keep prices low. This isn’t unusual – your insurance company does the same – but it nets them better prices than you would get if you tried to pay out of pocket at your local hospital. This article gives some examples of Amish getting sticker prices discounted from between 50% to 66% with this tactic alone; Medicare gets about the same.

Second, the Amish are honorable customers. This separates them from insurance companies, who are constantly trying to scam providers however they can. Much of the increase in health care costs is “administrative expenses”, and much of these administrative expenses is hiring an army of lawyers, clerks, and billing professionals to thwart insurance companies’ attempts to cheat their way out of paying. If you are an honorable Amish person and the hospital knows you will pay your bill on time with zero fuss, they can waive all this.

But can this really be the reason Amish healthcare is cheaper? When insurance companies negotiate with providers, patients are on the side of the insurances; when insurance companies get good deals (eg a deal of zero dollars because the insurance has scammed the hospital), the patient’s care is cheaper, and the insurance company can pass some of those savings down as lower prices. If occasionally scamming providers meant insurance companies had to pay more money total, then they would stop doing it. My impression is that the real losers here are uninsured patients; absent any pressure to do otherwise, hospitals will charge them the sticker price, which includes the dealing-with-insurance-scams fee. The Amish successfully pressure them to waive that fee, which gets them better prices than the average uninsured patient, but still doesn’t land them ahead of insured people.

Third, Amish don’t go to the doctor for little things. They either use folk medicine or chiropractors. Some of the folk medicine probably works. The chiropractors probably don’t, but they play a helpful role reassuring people and giving them the appropriate obvious advice while telling the really serious cases to seek outside care. With this help, Amish people mostly avoid primary care doctors. Holmes County health statistics find that only 16% of Amish have seen a doctor in the past year, compared to 54% of English.

Fourth, the Amish never sue doctors. Doctors around Amish country know this, and give them the medically indicated level of care instead of practicing “defensive medicine”. If Amish people ask their doctors to be financially considerate – for example, let them leave the hospital a little early – their doctors will usually say yes, whereas your doctor would say no because you could sue them if anything went wrong. In some cases, Amish elders formally promise that no member of their congregation will ever launch a malpractice lawsuit.

Fifth, the Amish don’t make a profit. Church aid is dispensed by ministers and bishops. Even Amish Hospital Aid is run by a volunteer board. None of these people draw a salary or take a cut. I don’t want to overemphasize this one – people constantly obsess over insurance company profits and attribute all health care pathologies to them, whereas in fact they’re a low single-digit percent of costs (did you know Kaiser Permanente is a nonprofit? Hard to tell, isn’t it?) But every little bit adds up, and this is one bit.

Sixth, the Amish don’t have administrative expenses. Since the minister knows and trusts everyone in his congregation, the “approval process” is just telling your minister what the problem is, and the minister agreeing that’s a problem and giving you money to solve it. This sidesteps a lot of horrible algorithms and review boards and appeal boards and lawyers. I don’t want to overemphasize this one either – insurance companies are legally required to keep administrative expenses low, and most of them succeed. But again, it all adds up.

Seventh, the Amish feel pressure to avoid taking risks with their health. If you live in a tiny community with the people who are your health insurance support system, you’re going to feel awkward smoking or drinking too much. Realistically this probably blends into a general insistence on godly living, but the health insurance aspect doesn’t hurt. And I’m talking like this is just informal pressure, but occasionally it can get very real. R&D discuss the case of some Amish teens who get injured riding a snowmobile – forbidden technology. Their church decided this was not the sort of problem that godly people would have gotten themselves into, and refused to help – their families were on the hook for the whole bill.

Eighth, for the same reason, Amish try not to overspend on health care. I realize this sounds insulting – other Americans aren’t trying? I think this is harsh but true. Lots of Americans get an insurance plan from their employer, and then consume health services in a price-insensitive way, knowing very well that their insurance will pay for it. Sometimes they will briefly be limited by deductibles or out-of-pocket charges, but after these are used up, they’ll go crazy. You wouldn’t believe how many patients I see who say things like “I’ve covered my deductible for the year, so you might as well give me the most expensive thing you’ve got”, or “I’m actually feeling fine, but let’s have another appointment next week because I like talking to you and my out-of-pocket charges are low.”

But it’s not just avoiding the obvious failure modes. Careful price-shopping can look very different from regular medical consumption. Several of the articles I read talked about Amish families traveling from Pennsylvania to Tijuana for medical treatment. One writer describes Tijuana clinics sending salespeople up to Amish Country to advertise their latest prices and services. For people who rarely leave their hometown and avoid modern technology, a train trip to Mexico must be a scary experience. But prices in Mexico are cheap enough to make it worthwhile.

Meanwhile, back in the modern world, I’ve written before about how a pharma company took clonidine, a workhorse older drug that costs $4.84 a month, transformed it into Lucemyra, a basically identical drug that costs $1,974.78 a month, then created a rebate plan so that patients wouldn’t have to pay any extra out-of-pocket. Then they told patients to ask their doctors for Lucemyra because it was newer and cooler. Patients sometimes went along with this, being indifferent between spending $4 of someone else’s money or $2000 of someone else’s money. Everything in the US health system is like this, and the Amish avoid all of it. They have a normal free market in medical care where people pay for a product with their own money (or their community’s money) and have incentives to check how much it costs before they buy it. I do want to over-emphasize this one, and honestly I am surprised Amish health care costs are only ten times cheaper than ours are.

I don’t know how important each of these factors is, or how they compare to more structural factors like younger populations, healthier lifestyles, and less end-of-life care. But taken together, they make it possible for the Amish to get health care without undue financial burden or government support.

II.

Why look into the Amish health system?

I’m fascinated by how many of today’s biggest economic problems just mysteriously failed to exist in the past. Our grandparents easily paid for college with summer jobs, raised three or four kids on a single income, and bought houses in their 20s or 30s and never worried about rent or eviction again. And yes, they got medical care without health insurance, and avoided the kind of medical bankruptcies we see too frequently today. How did this work so well? Are there ways to make it work today? The Amish are an extreme example of people who try to make traditional systems work in the modern world, which makes them a natural laboratory for this kind of question.

The Amish system seems to work well for the Amish. It’s hard to say this with confidence because of all the uncertainties. The Amish skew much younger than the “English”, and live much healthier lifestyles. Although a few vague estimates suggest health care spending far below the English average, they could be missing lots of under-the-table transactions. And again, I don’t want to ignore the fact that the Amish do live a little bit shorter lives. You could tell a story where all of these add up to explain 100% of the difference, and the Amish aren’t any more efficient in their spending at all. I don’t think this is right. I think the apparent 5x advantage, or something like it, is real. But right now this is just a guess, not a hard number.

What if it is? It’s hard to figure out exactly what it would take to apply the same principles to English society. Only about a quarter of Americans attend church regularly, so church-based aid is out. In theory, health insurance companies ought to fill the same niche, with maybe a 10% cost increase for profits and overhead. Instead we have a 1000% cost increase. Why?

Above, I said that the most important factor is that the Amish comparison shop. Everyone needs to use other people’s money to afford expensive procedures. But for the Amish, those other people are their fellow church members and they feel an obligation to spend it wisely. For the English, the “other people” are faceless insurance companies, and we treat people who don’t extract as much money as possible from them as insufficiently savvy. But there’s no easy way to solve this in an atomized system. If you don’t have a set of thirty close friends you can turn to for financial help, then the only institutions with enough coordination power to make risk pooling work are companies and the government. And they have no way of keeping you honest except the with byzantine rules about “prior authorizations” and “preferred alternatives” we’ve become all too familiar with.

(and as bad as these are, there’s something to be said for a faceless but impartial bureaucracy, compared to having all your neighbors judging your lifestyle all the time.)

This is a neat story, but I have two concerns about it.

First, when I think in terms of individual people I know who have had trouble paying for health care, it’s hard for me to imagine the Amish system working very well for them. Many have chronic diseases. Some have mysterious pain that they couldn’t identify for years before finally getting diagnosed with something obscure. Amish Hospital Aid’s catastrophic policy would be useless for this, and I feel like your fellow church members would get tired of you pretty quickly. I’m not sure how the Amish cope with this kind of thing, and maybe their system relies on a very low rate of mental illness and chronic disease. A lot of the original “hygiene hypothesis” work was done on the Amish, their autoimmune disease rates are amazing, and when you take out the stresses of modern life maybe a lot of the ailments the American system was set up to deal with just stop being problems. I guess my point is that the numbers seem to work out, and the Amish apparently remain alive, but when I imagine trying to apply the Amish system to real people, even assuming those real people have cooperative churches and all the other elements I’ve talked about, I can’t imagine it doing anything other than crashing and burning.

Second, I don’t think this is actually how our grandparents did things. I asked my literal grandmother, a 95 year old former nurse, how health care worked in her day. She said it just wasn’t a problem. Hospitals were supported by wealthy philanthropists and religious organizations. Poor people got treated for free. Middle class people paid as much as they could afford, which was often the whole bill, because bills were cheap. Rich people paid extra for fancy hospital suites and helped subsidize everyone else. Although most people went to church or synagogue, there wasn’t the same kind of Amish-style risk pooling.

This makes me think that the Amish method, even though it works, isn’t the method that worked for past generations. It’s an innovation intended to cover for health care prices being higher than anything that traditional societies had to deal with.

Why did health care prices start rising? I’ve wondered about this a lot before – see here, here, and here. Looking into this issue, I noticed glimpses of a different possibility. The increase started around the same time that health insurance began to spread. In one sense, this is unsurprising – of course health insurance would become a thing around the time care became unaffordable. But I’ve never seen someone really try to tease out causality here. Might the two trends have been mutually self-reinforcing? The price of care rises due to some original shock. Someone invents health insurance, which seems like a good idea. But this creates a series of perverse incentives, which other actors figure out how to exploit (eg the Lucemyra example above). Insurance-based-health-care becomes less efficient, but hospitals can’t or don’t internalize this to the insured patients – they just raise the price for everyone, insurance or no. That makes even more people need health insurance, and the cycle repeats as prices grow higher and higher and insurance becomes more and more necessary. This syncs well with some explanations I’ve heard of rising college prices, where once the government made easy loans and subsidies available to everyone, prices rose until they consumed all the resources available.

I have no idea if this is true or not. If it is, the Amish succeed partly by successfully forcing providers to internalize the costs of insurance to insurance patients. Sometimes they do this by literally asking hospitals for better prices because they are not insured (eg the “honest customer” example above). Other times they flee the country entirely to reach a medical system that doesn’t deal with insured patients (eg Tijuana). This seems to work well for them. But their reliance on church alms and Amish Hospital Aid suggests that their care is still more expensive and burdensome for them than past generations’ care was for them. They’ve just learned ways to manage the expense successfully.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

561 Responses to The Amish Health Care System

  1. OutsideContextProblem says:

    According to Google, doctors in the US earn 10x the median salary – presumably because of an artificially restricted supply. I’ve always assumed similarly inflated salaries across the medical sector explain most of the excessive costs of medical care.

    I also think the costs still exist in the Amish system, but are implicit rather than explicit. For instance, having the minister handle the administrative affairs looks free, but his time is still costly; and no law suits means that Amish must absorb the costs of malpractice themselves (and are possibly at greater risk). And less directly, there’s all sorts of utility points being lost across the system (lack of choices etc) which would be more formalised and counted in the standard system.

    The standard system makes these costs more legible, and they fall on different parties, but many of them still exist. This is not to say that modern complex systems like healthcare are not poorly optimised for some policy outcomes.

    • gbear605 says:

      > doctors in the US earn 10x the median salary – presumably because of an artificially restricted supply

      The supply does seem to be a bit artificially restricted, but I doubt that doctors would make the median salary if the supply wasn’t restricted. To be a doctor that a patient would willingly go to, you have to have fairly advanced medical skills and be fairly intelligent. I’d think that these factors alone would push the salary up above the median, even if not to 10x.

      • OutsideContextProblem says:

        Undoubtedly, but assuming salaries have some sort of normal distribution 10x seems absurd.

        • gbear605 says:

          Even without credentialism, assuming a normal distribution is a big leap to take. In our current system, some people without a college degree (think programmers at Google) can make 5x the median income and that is only the rare cases who don’t go through college. I doubt that it’d be a 10x multiplier for every doctor without credentialism, but it’s also not a 10x multiplier for most doctors today. A median doctor makes 250k/year while the median working adult earns around 40k/year, working fewer hours than the average doctor. So the multiplier seems to be more like 5x, about the same as some (rare) uncredentialed engineers. It seems much more likely that salary has a very long tail.

          • OutsideContextProblem says:

            Without doing any research whatsoever, I’d bet that engineer salary at Google is actually much less excessive relative to the local median salary than the typical doctor. And of course, there are far more doctors in America than FAANG engineers, and FAANG engineers add economic value at a much greater scale. And the San Franscio tech scene is also very much an aberration.

            Lots of smart, hardworking people want to be doctors. Most of them don’t get into medical school, or can’t afford it, fail-out, can’t get a residency or rationally look at the (very arguably unnecessary) time and money commitment of two lengthy degrees and go be engineers or something. Those who make it into the profession can charge higher rates, and work to limit supply further, including by overworking residents. The result is that only a very few exceptionally amazing people get to be doctors, and I have to pay an extortionate amount to see them. I would quite happily pay a lot less to have someone merely in the top 10% prescribe me antibiotics, rather than the top 1%.

            If you look at Australia, or Europe doctors there are paid well and have comfortable middle class lives, but the multiple of median income is a lot more constrained.

            Of course, i’m singling out doctors here but most US medical professionals seem very well compensated.

            Everyone keeps talking about administrative bloat, or tort reform, or insurance games or whatever and ignoring the elephant in the room – medical professional compensation – which absolutely dwarfs malpractice insurance costs (~3% if I remember correctly), and is really high relative both to the western world and the rest of the US economy.

          • aphyer says:

            @OutsideContextProblem

            Quick research with data from here:

            The average salary under code 29-1000 (‘Healthcare Diagnosing or Treating Practitioners,’ which is the closest-looking thing to a broad category for ‘doctors’) is $102,470.

            The average salary under code 15-1200 (‘Computer Occupations’, which is the closest-looking thing to a broad category for ‘programmers’) is $93,620.

            In support of your point, the 29-1000 healthcare code includes some lower-paid people who drag down the average of doctors. For instance, ‘Family Medicine Physicians’ have an average pay of $213,270, while ‘Therapists’ average $80,850 (and appear to include a lot of odd specializations like ‘Recreational Therapists’, which don’t really sound like doctors).

            On the other hand, there is also a lot of salary dispersion among programmers. For instance, the category ‘Software Developers and Software Quality Assurance Analysts and Testers’ has an average pay of $111,620 – but this category appears to include both software developers (who I think make a lot of money) and QA people (who I think often make less).

            Additionally, it is possible that a lot of programmers, when they get promoted and make more money, are moved into jobs that get them classified as ‘managers’ rather than as ‘programmers’, while doctors usually remain classified as ‘doctors’.

            Finally, medicine is a more dualized field than programming. Implementing supply restrictions increases doctor pay for two reasons: first, as you say, it reduces supply, and so drives up prices; but secondly, it gets rid of the bottom X% of doctors (who would have been paid less) and leaves only the top X% of doctors (who would have been paid more anyway). If the ‘programmers’ salary numbers includes both competent and incompetent programmers averaged together, while medical licensing removes the incompetent doctors, then it is possible for the average doctor to be paid more than the average programmer while a competent programmer is paid as much as a competent doctor.

        • keaswaran says:

          Salaries are a classic example of something with a power law distribution rather than a normal distribution.

          With a normal distribution, being one standard deviation further from the mean gets less and less likely in a superexponential way. The average US male is 5’9″. A US male is about half as likely to be 6’0″. Being 6’3″ is about a quarter as likely as being 6’0″. Being 6’6″ is far less than a quarter as likely as being 6’3″. Being 6’9″ is far, far, far, less likely than being 6’6″.

          But with a power law distribution, these large deviations aren’t anywhere near as unlikely. The average US male makes something like $30,000. Maybe half as many people make $60,000. But $120,000 is not that rare (it’s top 5%). And $240,000 is also not that rare (it’s about top 2%). And $500,000 is far more common than being 6’9″.

          You get a normal distribution when the variable is the result of adding a whole bunch of completely independent factors (like height is with some genes and nutritional factors and whatever). You get a power law distribution when each further factor is more likely to go to the people that had the earlier factors (like income is – as everyone since the the book of Matthew in the Bible has noted, most factors mean that the rich get richer, since it’s much easier to make more money if you already have money to invest).

          • vaniver says:

            Salaries are a classic example of something with a power law distribution rather than a normal distribution.

            I think only the top ~3% of incomes are a power law, and that’s because they’re mostly driven by capital returns; see figure 6 of this paper.

          • Del Cotter says:

            You get a normal distribution when the variable is the result of adding a whole bunch of completely independent factors

            I hope you don’t if I nitpick your terminology a little. Per Gauss’s Central Limit Theorem, you get a normal distribution when the variable is the sum of adding a whole bunch of completely independent terms. When the variable is the product of multiplying a whole bunch of completely independent factors, you get a log-normal distribution.

      • North49 says:

        To be a doctor that a patient would willingly go to, you have to have fairly advanced medical skills and be fairly intelligent.

        I don’t think that this is true, or not true in a wide variety of situations. Maybe you want the best heart surgeon in all the land no matter the cost, but if we think about the folk medicine/ chiropractor function as a screening process, there’s no reason you’re first contact with the medical system for any ailment should have to be a fully credentialed doctor. I would be more than happy to solicit advice from a pharmacist or RN as a screening function, or even be treated by an RN for simple procedures and go to a doctor only if it is warranted. There’s probably an 80/20 rule that applies here with common procedures like getting stitches and antibiotics, or running annual check-up test could all easily be handled by someone with an RN’s level of training and skill.

        I also have spent quite a lot of years uninsured, and know first hand what paying out of pocket for the ‘doctor patients would willingly go to’ costs, which may account for my willingness to go to someone with less credentials (sorry, I mean ‘skill and intelligence’). And by the way, the mere existence of several nested layers of regulation stopping people from going to less credentialed care providers should probably be regarded as strong evidence of people’s willingness to do so if they could.

        • albatross11 says:

          Indeed, if you go to a primary care doctor around here nowadays, whether that’s a regular practice or a quick clinic, you almost always will be seeing a PA or NP, rather than an MD. Both of those involve someone with much less training than a doctor–which makes sense in a context where like 95% of what they see is routine low-complexity stuff.

        • Mabuse7 says:

          It is said that the reason that Homeopathy has remained so enduringly popular over the past century; despite the complete quackery of the remedies involved, is that Homeopaths do primary care and screening better than anyone, people seem to get much more out of the more personable, holistic manner of Homeopathic practitioners than the more clinical style of GPs.

          • albatross11 says:

            In a world where a large fraction of patients would do better with a placebo than an actual medicine, homeopathy is almost certainly a win.

        • JPNunez says:

          I also have spent quite a lot of years uninsured, and know first hand what paying out of pocket for the ‘doctor patients would willingly go to’ costs, which may account for my willingness to go to someone with less credentials (sorry, I mean ‘skill and intelligence’). And by the way, the mere existence of several nested layers of regulation stopping people from going to less credentialed care providers should probably be regarded as strong evidence of people’s willingness to do so if they could.

          Yeah but people will go to shamans, homeopaths, crystal healers, etc, etc if allowed.

          • Garrett says:

            > Yeah but people will go to shamans, homeopaths, crystal healers, etc, etc if allowed.

            So? As long as they aren’t misrepresenting what they do, why should having that option be any worse?

          • albatross11 says:

            Note that in the US, despite piles and piles of regulations, you can go to naturopaths, homeopaths, chiropracters, herbalists, accupuncturists, etc. Nobody will stop you–those guys all have offices with big signs facing the street, listings on the internet, etc. Many of them even have their own regulations to keep out substandard less-trained practitioners of Reiki or treating your illness with water.

            So yeah, if you let people, some will go to quacks of various kinds. And we do, and they do. I am not at all convinced that we’d be better off banning those folks, even though I think they are mainly providing entertainment services and placebo packaging services instead of useful medical care.

      • Winja says:

        Once upon a time, someone told me that the supply of doctors was artificially restricted by the AMA; that they exercised a level of control over how many doctors were to be admitted to medical school in any given year, and that this number was kept low in order to ensure that doctors would have no trouble finding work and good salaries upon graduation.

        I have no idea if this is true or not.

        • Mary says:

          Residencies are more a sticking point.

          • Cliff says:

            The AMA won’t allow any more medical schools or slots on the basis that there are not enough residencies for any more graduates. The AMA insists that training a new doctor is completely impossible without the government paying for it. All residencies are government funded. The AMA says there is nothing they can do unless the government pays to train more doctors. Strangely, every other profession in the world manages to train new practitioners without any government funding at all. Meanwhile, new doctors in residencies are paid $60k/year and work 273 hours/wk on average.

          • John Schilling says:

            Meanwhile, new doctors in residencies are paid $60k/year and work 273 hours/wk on average.

            Something seems a bit off here, and I’m now unsure whether we’re supposed to take any part of your post seriously.

          • albatross11 says:

            We should definitely stop requiring the use of time turners to get through a residency.

    • Scott Alexander says:

      Disagree. First, doctors only make about 5x median salary (I think average doctor salary is about 175K, median salary is about 35K). Second, some of this is because doctors are higher-skilled labor than average. My guess is the artificial shortage only doubles their salary – without it, average doctors would probably make about the same as average lawyers (high 5 to low 6 digits).

      In any case, I don’t think this is especially relevant to high health care costs, since doctors only make up about 20% of health care expenses.

      (obviously I’m biased here)

      • OutsideContextProblem says:

        The mean is apparently $294,000, but I can’t find the median easily and I have no idea what the distribution looks like. The mean salary in 2015 was apparently $56,516. So i’ll take 5x as a good estimate.

        I think getting paid the same or a bit more than the average lawyer sounds about right.

        So, I agree with all your points. And I’m left with the fact that excess MD salary (on a very crude modelling assuming no further affects) seems to account for at least 10% of health care expenses. That seems enormous to me. Imagine the regained utility if every American had their (highest in the world) health care costs cut by 10%.

        And, I don’t think it’s just doctor salaries but all the other healthcare professionals. Doctor salaries are just the most visible.

      • viVI_IViv says:

        . First, doctors only make about 5x median salary (I think average doctor salary is about 175K, median salary is about 35K). Second, some of this is because doctors are higher-skilled labor than average.

        In Europe doctors make 2-3x the median salary.

        My guess is the artificial shortage only doubles their salary

        Which would indeed put their salaries in line with Europe.

        In any case, I don’t think this is especially relevant to high health care costs, since doctors only make up about 20% of health care expenses.

        What about nurses, technical staff and admin staff? Are their salaries similarly 2x of what they would be without artifical barriers?

        • Akhorahil says:

          Which would indeed put their salaries in line with Europe.

          Although there’s an amount of artificial shortage here as well. The medical schools and hospitals ensure it.

          • viVI_IViv says:

            The alternative is every barber and their dog performing open heart surgery.

            Because of information asymmetry, you can’t have high quality health care in a completely free market.

          • RicardoCruz says:

            The alternative is every barber and their dog performing open heart surgery.

            I am sure there are other alternatives. Here in Portugal, the “Order” controls the number of medicine schools and even how many students each one has. The power of the Order could be restricted so they cannot decide such things. Also, two or three Orders could be created so there is not a single monopoly deciding everything. Also, hospitals could have the right to hire doctors from other approved countries (say, EU countries) outside of the Order.

          • DeWitt says:

            In the Netherlands, at least, medicine in particular is doubly restricted. We’re not the United States; you can’t show up to a university with money in hand and expect to receive an education, you need to show that you have either finished a high school of the highest ‘tier’ or completed some prior higher education before university will let you in. This means every barber and their dog isn’t, in fact, already practicing law or building bridges: they simply don’t get let in.

            Medicine was special, in that it was (somehow) decided that we should restrict the supply – God forbid people study medicine instead of art history. To study medicine you either needed very good grades in high school, or to apply for a spot that got filled by lottery. I don’t know why this was done, though I suspect it’s just really expensive to educate doctors when you don’t shoulder people with American amounts of debt, but it was the double restriction that I spoke of.

            Of course, just a few years ago, it was decided that random chance is very unfair, and that people with high grades shouldn’t get preferential treatment. Instead, we went for the much fairer way of selecting students by manner of interview. Sure hope your parents aren’t very black, Moroccan, or anything, kid.

            Fortunately, people noticed this didn’t work, so with some luck we’ll get something even worse.

          • John Schilling says:

            In the Netherlands, at least, medicine in particular is doubly restricted. We’re not the United States; you can’t show up to a university with money in hand and expect to receive an education,

            You can’t show up with just money in hand and expect to receive an education in the United States either, unless you show up with so much money that it’s irrelevant which country you happen to be in because you’re part of the jet-set globalist elite (and you’re probably not studying to be anything as practical as a physician, either). Also, how confident are you that Dutch universities wouldn’t make an exception for someone who was offering ~10E6 cash, which is I understand the normal price for such things in the US.

            High school diploma or the equivalent is the bare minimum for not-filthy-rich in the United States. There may be a difference in that the US at least pretends that all of its high schools are college-prep schools, we don’t have a separate track for future tradesmen.

          • IrishDude says:

            Because of information asymmetry, you can’t have high quality health care in a completely free market.

            That doesn’t follow. Markets can solve information asymmetry. See Uber and Amazon ratings or private certification like kosher food and underwriters laboratory.

          • viVI_IViv says:

            Markets can solve information asymmetry. See Uber and Amazon ratings

            Nice surgery, five stars. Like and subscribe.

            What can possibly go wrong?

            private certification like kosher food and underwriters laboratory.

            Because private certification worked so well in the 2007 financial crash…

          • viVI_IViv says:

            High school diploma or the equivalent is the bare minimum for not-filthy-rich in the United States.

            If I understand correctly in the US you have to be literally retarded or a juvenile criminal in order to have a chance to fail high school.

          • Swami says:

            I am not being facetious, but the mentally retarded get diplomas in California too. Really.

          • DeWitt says:

            High school diploma or the equivalent is the bare minimum for not-filthy-rich in the United States. There may be a difference in that the US at least pretends that all of its high schools are college-prep schools, we don’t have a separate track for future tradesmen.

            The second part of your sentence is what my argument is based on, yes. We do have separate tracks for future tradesmen and progwyerdocgineers, and if you went to the former track, you flat-out don’t get to study physics or computer science unless you catch up.

          • albatross11 says:

            viVI_IViv:

            Private certification may have failed us in the 2008 financial crisis, but at least we had extensive government regulation to save us from any problems….

          • anonymousskimmer says:

            cough cough https://duckduckgo.com/?q=high+school+requirements+for+college&t=canonical&ia=web

            If you haven’t passed the right assortment of courses at your US high school, US universities (4-year schools) may not accept you, regardless of whether you have a high school diploma.

            If this is the case for you, and you do want a University degree, then you must resort to the same non-traditional path that’s available in Europe, which in the case of the US is an associate’s degree at a community college prior to transferring to a 4-year university.

            Yes. The US does have “tracks”, just like Europe. They simply aren’t publicized as well to the public. I.e. It’s quite possible to discover the fact that you’re on the wrong track, and require years more education just to qualify for your desired major, at the university application stage.

          • DeWitt says:

            That’s very nice, but it’s really not the same as what we have: it just means the top colleges have standards, and that the poor people ordinary ones just let in anyone. You can still finish high school and toss around money to join a university you have no business attending.

          • IrishDude says:

            @viVI_IViv

            Nice surgery, five stars. Like and subscribe.

            The point of mentioning Uber ratings was to rebut that ‘information asymmetry’ means markets can’t work. Markets solve information asymmetry when consumers demand info. With Uber, it’s why people get into the cars of strangers and feel safe.

            I wasn’t implying that an Uber/Amazon model would be best for free market medical care, though aspects of it would certainly help. How do you choose what doctors or surgeons you want to use? In part, consumer feedback from prior patients is certainly valuable to me, including word of mouth from friends and family and other private mechanisms like Healthgrade.

            Anyways, private agencies work well to certify Kosher food and underwriters lab works well to certify safety of products. You can have experts doing certification and don’t need to only rely on consumer feedback to resolve information asymmetries.

            Because private certification worked so well in the 2007 financial crash…

            Are you referring to the big three credit agencies? The only agencies sanctioned by the state and approved by the SEC to provide information that financial firms had to rely on for regulatory purposes?

            Even if the ‘private’ agencies certified junk as AAA and the financial firms knew it, the incentive to buy and sell junk remained as long as they thought profits were private and bailouts would be public. Fannie, Freddie, and the Federal Reserve also ‘worked so well in the 2007 financial crash.’

          • anonymousskimmer says:

            it just means the top colleges have standards, and that the poor people ordinary ones just let in anyone.

            A community college in the US is generally the equivalent of a trade school, with the exception that you can get an associate of arts degree in something like English.

            The “poor-person” 4-year schools, also known as the for-profit colleges, are also trade-school like. Their degrees aren’t prestigious, fairly limited in type to business or technical programs, or have extra admission requirements. E.g University of Phoenix’s B.S. Nursing program requires an associate’s degree or Nursing diploma, plus an RN license, in addition to a high school diploma or equivalent.

            The for-profit 4-year schools and the community colleges seem to be approximately equivalent to a German Berufsakademie, with the exception that a 2-year community college degree can, depending on grades, generally transfer to the equivalent academic 4-year program at a university (e.g. a Biotechnology A.S. will transfer to a Biotechnology B.S. or Biology B.S. program, though the general Biology B.S. program will likely require extra classes to fulfill all of the graduation requirements.

            Successful transfers from 2-year programs at community colleges to a Bachelor’s degree aren’t that frequent. Most people who go to a 2-year program finish at a 2-year program, not a 4-year program.

          • m.alex.matt says:

            Are you referring to the big three credit agencies? The only agencies sanctioned by the state and approved by the SEC to provide information that financial firms had to rely on for regulatory purposes?

            Even if the ‘private’ agencies certified junk as AAA and the financial firms knew it, the incentive to buy and sell junk remained as long as they thought profits were private and bailouts would be public. Fannie, Freddie, and the Federal Reserve also ‘worked so well in the 2007 financial crash.’

            It’s even worse than that: Capital regulations (specifically the US implementation of Basel I and II) allowed smaller amounts of capital to be held against securities rated higher by one of those sanctioned agencies, so there was literally a profit opportunity in continuing to go to them for ratings. Alternative ratings agencies not only had the normal barriers to competition (difficulty competing with a larger, more prestigious competitor, the expense of getting your own name out there, etc), but they had an insurmountable regulatory barrier that amounted to an implicit subsidy in that banks had a serious regulatory opportunity cost in using them instead of one of the big three.

          • viVI_IViv says:

            Markets solve information asymmetry when consumers demand info. With Uber, it’s why people get into the cars of strangers and feel safe.

            People already got into taxis and felt safe even before user ratings were a thing, people feel safe because they trust the state to punish those who commit assault or drive under influence. (Uber just does some background checks on the drivers, it has nothing to do with user ratings).

            User ratings only work to assess the perceived quality of the service, which is something that the user can assess immediately and at worst if the service is really terrible they will get out of the car and lose a small amount of money.

            With health care the typical user can’t tell if a diagnosis was accurate, if their doctor recommended an expensive procedure because it was really needed or because they stood to gain money, and whether a complex procedure was performed correctly. User rating would just correlate with superficial things like how charismatic the doctor was.

            Anyways, private agencies work well to certify Kosher food

            How do you know that Kosher certifications are accurate? If a jew eats some meat that wasn’t properly slaughtered according to Halakic principles it’s not like YHWH descends from Heaven to smite them.

            underwriters lab works well to certify safety of products.

            Underwriters incur a legal penalty if they hapen to certify something that turns out to be false.

            Are you referring to the big three credit agencies? The only agencies sanctioned by the state and approved by the SEC to provide information that financial firms had to rely on for regulatory purposes?

            If people didn’t trust the big three and only used them because of regulatory requirements, then why wasn’t there a parallel private circuit of reliable certification agencies?

          • IrishDude says:

            People already got into taxis and felt safe even before user ratings were a thing, people feel safe because they trust the state to punish those who commit assault or drive under influence.

            Taxis helped reduce information asymmetry too, with public certification from the state.

            Taxis and Uber have two different mechanisms for reducing information asymmetry. It’s the reason why people would feel safer using either of them than hitchhiking, where no filtering mechanism exists. As to which mechanism is better for ensuring quality and low prices between Taxis and Uber, see which service most consumers would rather use.

            (Uber just does some background checks on the drivers, it has nothing to do with user ratings).

            Uber’s background checks are another way a private company reduces information asymmetry. Uber also uses ratings (given by customers) to remove bad drivers from their service.

            With health care the typical user can’t tell if a diagnosis was accurate, if their doctor recommended an expensive procedure because it was really needed or because they stood to gain money, and whether a complex procedure was performed correctly. User rating would just correlate with superficial things like how charismatic the doctor was.

            User experience can measure the quality of doctors in many ways, including whether the doctor’s recommended treatment fixed their issue to their satisfaction.

            What information do you use to pick a doctor? Do you use any other information besides whether they’re certified by the state?

            How do you know that Kosher certifications are accurate? If a jew eats some meat that wasn’t properly slaughtered according to Halakic principles it’s not like YHWH descends from Heaven to smite them.

            How do you know whether a government certified product really matches its certification? There’s no guarantees in life, so it comes down to trusting the mechanism used. In the case of kosher, you should skim through the link I gave above as to how accuracy is incentivized among the competing certification companies. Having a reputation for inaccurate certifications hurts business.

            Underwriters incur a legal penalty if they hapen to certify something that turns out to be false.

            What does this have to do with whether markets can reduce information asymmetry? Underwriters is a private laboratory that certifies the safety of products for insurance purposes and lets businesses and consumers have more insight into the safety of products that they might not be qualified enough to determine on their own.

            In the absence of Underwriters Laboratory, there would be more information asymmetry, not less. It’s presence in the market therefore reduces information asymmetry.

            If people didn’t trust the big three and only used them because of regulatory requirements, then why wasn’t there a parallel private circuit of reliable certification agencies?

            It’s hard to compete against government granted monopolies. See m.alex.matt’s response: “Capital regulations (specifically the US implementation of Basel I and II) allowed smaller amounts of capital to be held against securities rated higher by one of those sanctioned agencies, so there was literally a profit opportunity in continuing to go to them for ratings. Alternative ratings agencies not only had the normal barriers to competition (difficulty competing with a larger, more prestigious competitor, the expense of getting your own name out there, etc), but they had an insurmountable regulatory barrier that amounted to an implicit subsidy in that banks had a serious regulatory opportunity cost in using them instead of one of the big three.”

            See also my prior response: “Even if the ‘private’ agencies certified junk as AAA and the financial firms knew it, the incentive to buy and sell junk remained as long as they thought profits were private and bailouts would be public.”

          • Edward Scizorhands says:

            I’m willing to believe that the health markets have so much information asymmetry that they can’t work.

            But not if the starting proposition is “well, people felt perfectly safe getting into taxis.” Taking a taxi was always an unknown factor and filled with trepidation and required me to actively watch for the driver taking an expensive route. Being able to complain to some bureaucracy was not a valid relief. And despite it being plainly against the rules to decline to let me get in based on my destination, I experienced drivers refuse to take me to the airport.

            As much as Uber sucks in so many ways, I never wonder as I get in if I am going to be scammed. The transaction has been negotiated.

            The ability to stop using Uber is way more powerful than my ability to complain to the livery commission.

        • Garrett says:

          > What about nurses, technical staff

          Most techs of some sort have 2-year education (associates degree). This is based on a quick search of radiology techs and respiratory therapists. Paramedics (who frequently go on to be ER techs) typically have roughly an associate’s degree.

          Nurses are an interesting case. You can become licensed as a nurse with an associate’s degree. But every hospital around here wants people to get their bachelor’s degree in nursing. Indeed, a lot of them will help pay for it. And they facilitate getting master’s degrees in nursing as well. I suspect that there’s a ranking/compliance/something in play where hospitals are encouraged to have nurses with higher credentials.

          • theredsheep says:

            While a two-year degree is acceptable for RTs, the AARC is pushing pretty hard for four year minimums, and a lot of people seem to think they’ll eventually get it. I don’t understand why this would be necessary; I’m going to school for five semesters and frankly everyone agrees that most of what I’ll actually learn about respiratory care will happen on the job after I certify. Inflated credentials will only drive up costs and restrict supply without significantly increasing quality of care. I just hope I can get grandfathered in.

          • Garrett says:

            > pushing pretty hard for four year minimums

            We’re running into the same thing in EMS. Basically nobody (at least around here) can afford a decent living in EMS working a single job. Everybody I know is either working a second job or going to school. The problem is that there just isn’t a good source of money available unless you can turn yourself into a taxpayer-funded organization directly. Medicare and medicaid pay very poorly.

            One suggestion has been to require a bachelor’s degree to become a paramedic. There’s value in that on some level. But at the same time, it would be done only to reduce the supply of people in the field and thus drive up costs due to artificial scarcity. Which is terrible. It’s literally adding waste to drive up costs. We’d be better off with a lottery system to get into paramedic school and then keeping the supply low.

      • MadRocketSci2 says:

        How much of doctors’ high salaries are mandated by the student loan debt trap? One of my relatives is trying to get into medical school. The GI bill won’t cover her completely for the cheapest tier of medical school! The debt in addition to the GI bill benefits is staggering.

        Doctor’s aren’t really free and independent professionals anymore, it seems: They need that income to service their debt.

        Artificial supply restrictions are evil. If you’re concerned about quality, grade harshly, don’t turn people away from even entering the school! You’re not throwing out the worst candidates for doctors, you’re throwing out a random sample of everyone. (Or think of what you’re selecting for! Determination to game a rigged admissions system.)

        And the doctors salary multiple isn’t anywhere near as important as the provider/everyone-else ratio. My primary care provider (orwellian bureauspeak, btw) might make a decent living, but he’s also ensconced in several layers of secretaries, administrators, and office staff who are all there to fight the insurance companies. I don’t mind paying him as much as I mind the necessity of paying them. (Though, at about ~10 minutes of distracted attention for hours of paperwork, I sort of mind that too.)

        How can educating a doctor be more expensive than educating any other profession? Chemical engineers need chem-labs too. Ag-bio students are dissecting cows one building over. The equipment is all the same: Classrooms, professors, godawful textbooks, all-nighters, exams.

        • rahien.din says:

          How much of doctors’ high salaries are mandated by the student loan debt trap?

          If you’re concerned about quality, why not just grade harshly?

          How can educating a doctor be more expensive than educating any other profession? The equipment is all the same

          Are these genuine questions?

        • Edward Scizorhands says:

          If doctors have a wage premium, and schools can be a bottleneck on becoming a doctor, schools have no reason not to capture as much of that wage premium as possible for themselves.

          In theory some schools could compete by offering low-cost education, but the entire environment is built for increasing prices since lots of your customers don’t see the price. See
          https://slatestarcodex.com/2015/06/06/against-tulip-subsidies/

      • Winja says:

        Reason, or some other libertarian outlet, once ran a diagram that showed the growth over time of the number of healthcare practitioners; doctors, nurses, etc. vs. the number of healthcare insurance company employees.

        And it was one of those graphs that makes you go “holy cow” to the point that I can still recall seeing it ten-odd years later.

        I think Scott is mostly right that the pay doctors get probably isn’t a huge contributor to overall cost, and I’m inclined to agree that insurance companies are eating up the lion’s share of the resources, even if it doesn’t show up on the bottom line of the insurance companies.

        • Matt M says:

          Agreed.

          Just because insurance companies aren’t raking in the net profits doesn’t mean that they can’t be responsible for a huge part of the cost increases.

          It could just mean that a lot of those cost increases are being passed on to the employees of insurance companies rather than the shareholders.

          How many middle managers making well over six figures exist at these companies? All of the money they are making is shown as an “expense” for the insurance company and therefore not reflected in profitability. But there’s still a real shift in resources going on from health care consumers to these people, who may very well be providing a negative-value product to society.

          • acymetric says:

            Right, I was going to make the same point after reading Scott’s “fifth” and “sixth” paragraphs about profits and admin costs.

            When most people talk about “health insurance profits” they really mean something like “profits and anyone getting significant wealth from employment at a health insurance company” which would obviously include CEOs but probably includes a lot of mid-level employees as well, even if the insurer is “non-profit”.

          • Matt M says:

            Indeed. If you suspect that these companies are providing no real value, then the amount they are siphoning away from consumers is far closer to “their entire operating budget” than it is to “their net profit”

          • etheric42 says:

            Since insurance providers are only legally allowed to make a maximum of X percent profit (and they have to return the excess if they make more), is there anything that stops them from increasing costs in order to make that maximum percent into a larger absolute number?

        • AKL says:

          Insurance companies collectively pass through 90% of their revenue (insurance premiums, copays, deductibles, etc.) to providers (i.e. by paying them for health care services), so there’s no way that “insurance companies are eating up the lion’s share of the resources.”

          source

          • acymetric says:

            Well that’s still 10%.

            In terms of cost increases, what if you factor in the administrative costs spent by providers to work with insurance companies?

          • AKL says:

            Sure, in a world with multiple insurance companies there will be higher administrative costs for providers (though any kind of reliable calculation would be really, really hard). And 10% is a big number but it’s not the lion’s share of the resources.

            Obviously the system we have is wildly suboptimal, and the existence of 500 +/- insurance companies is an element of that system, but the idea that the US has high costs because of the bad insurance companies isn’t accurate.

      • Charlie Lima says:

        Physicians have several other factors that drive up their salaries:

        1. Doctors work longer. Average is something like 50 hours per week (Merritt Hawkins puts the mean at 51.4). US average is only 34.4 hours worked per week. Even when you equalize everything else, physicians work 50% more on average than the general populace. And assuming there is diminishing tradeoff between time and money, we should expect that coaxing more hours each week out of physicians is going to cost much more for each additional hour.

        2. Doctors have to pay off their training. Yeah, lawyers have get through law school, but not only do physicians have to spend 25% more time after college out of the workforce, when they finally get jobs as residents they make less than entry level lawyers. Additionally physician training is pretty hellish. You give up four years of your life to go through medical school. Unless you are lucky or crush Step, you often have to relocate multiple times and may not even be able to live within 100 miles of your ideal location. All of this gets paid back with deferred compensation, and again money later is less valuable than money now.

        3. Doctors have fewer years to earn their salaries. The average physician starts medical school at 24. They enter residency at 28. They finish residency at 32. If we expect them to retire at 70, that leaves them only 38 years to earn all their compensation. Say they went into business straight out of college. Well that means retiring at 70 gives 49 years to earn everything.

        4. Physicians take real risks. We are seeing this now with an unfortunate number of docs not getting to see retirement due to Covid. We see it also with the stress that comes from working the ED, having to manage life and death surgery, or dealing with psychic toll of having all your patients die in Rad Onc. In other professions, these sorts of risks command premiums, like the wage difference between mall cops preventing thefts and actual cops on the beat (let alone security for actual warzones).

        Some professions have any one these. Exceedingly few have all of them. And then we want our physicians to among the best and brightest so we are adding implicit multipliers on top of a high salary in whatever else they do.

        My guess is the shortage has done nothing more than make a bunch of NPs wealthier. After all, you can easily earn $100,000 more per annum as a family physician by moving from a nice coastal metropolis to some rural area, yet the latter continue to have vastly more physician vacancies.

        • Edward Scizorhands says:

          Scott has talked about how the US is unusual in requiring more schooling and training and doesn’t seem to produce doctors any better than other countries.

          • Clutzy says:

            That doesn’t surprise people who follow signaling based theories of the American education system.

        • acymetric says:

          US average is only 34.4 hours worked per week.

          Seems like we should only be looking at full time employees, right? 34.4 hours/week is obviously including a bunch of part time labor. Maybe only full-time salaried workers. (Obvious counterpoint is that we should only be comparing wages for doctors with other full time workers, and I fully agree there).

          In my experience across a couple different careers, 50 hours/week would be a light workload.

          • Charlie Lima says:

            A better option would be to just look at average hourly rates. Say our average physician is making $300,000 per year and working 51 hours per week. That works out to around $112 per hour.

            Average American earns $43,585 per google. That comes out to be around $24 per hour. This gives on overall figure of around 4.5.

            This is, of course still an overestimate because we are not looking at health insurance which is going to easily be 10% of the average full time worker’s compensation, but I doubt it makes up even 5% of average physician’s compensation.

            At the end of the day, doctors are paid well, but it should not surprise us as they have fewer years of remuneration, higher entry costs, and truly abysmal working conditions for some of those (e.g. I think the limit is still 104 hours per week in some residencies and “most” are in compliance).

    • Aransentin says:

      > I’ve always assumed similarly inflated salaries across the medical sector explain most of the excessive costs of medical care.

      One of the reasons that healthcare in Europe is more affordable than the US might be that people are paid significantly less – e.g. here in Sweden, the median wage for nurses is 25,000 SEK/month, or about $30,000/year. For a “specialistläkare” (a “specialist doctor”, like a pathologist or oncologist) it’s 63,921 SEK/month, or $76,318/year. If salaries can be many times lower than in the US it makes sense that costs can be significantly reduced as well.

      • Akhorahil says:

        Note that it’s important to add 30% to Swedish salaries right away if we’re to compare salary costs – there’s an Employer Tax of 30% that is deducted before the salary even shows up to the employee.

        • OutsideContextProblem says:

          Eh, it’s always hard to compare. Are we allowed to add typical student loans repayments and health insurance costs to Swedish salaries because those are publicly funded?

          • 10240 says:

            When it comes to explaining the cost to people who use the service, no. What we need is the cost of employing someone for the employer.

      • Akhorahil says:

        That said, U.S. salaries for upper/upper-middle-class jobs are absurd compared to Sweden. It takes a lot – a lot – to earn even three times the median wage. And then you’re heavily progressively taxed on top. I’m a software guy with 20 years of experience and reasonable paid, and I don’t earn twice the median salary.

      • Winja says:

        Do you have comparatively lower cost of living there as well?

        • Akhorahil says:

          No, higher as a rule, at least as far as consumption goes (Sweden performs worse on PPP BNP than nominal, and our Big Mac index is appalling). 6-25% VAT on everything increases this further. On the other hand, there’s little to no direct cost to healthcare and education.

          Very high minimum wages also drive up service work costs. Basically no-one has a nanny, for instance (not that they’re particularly needed with cheap and easily accessible kindergartens).

    • rahien.din says:

      artificially restricted supply

      You say that like it’s a bad thing.

      I would quite happily pay a lot less to have someone merely in the top 10% prescribe me antibiotics, rather than the top 1%

      In our current situation, medical errors kill a lot of people. This is both in very directly-attributable ways, but also in more insidious and far-reaching ways.

      For instance, lots of people die from MRSA infections. An important way we fight MRSA (and other resistant bugs) is good stewardship of antibiotics. If one top-1% doctor has already done this poor a job of antibiotic stewardship, what will her nine top-10% colleagues do? Is life going to be better, or worse?

      We already fuck up all the time. Why do you want to give me a bunch of colleagues who are potentially worse than I am.

      Supply restriction is how you pay for working health care.

      • block_of_nihilism says:

        Yes, this is a good point. Even with the high standards for becoming a physician, there are still a lot of mediocre to poor practitioners out there. I think there is simply a limited supply of people who are smart AND don’t faint at the sight of blood AND want to work crazy hours AND… etc.

        • Cliff says:

          And YET, somehow everyone else in the world manages and gets healthcare about as good as ours… head-scratcher

      • Winja says:

        I find this argument unconvincing.

        Most medical issues are going to be quite routine, and could be done by someone who’s in that notional 10% rather than the top 1% (this of course assumes that the system is artifically restricting the top 1% exactly right, and that they aren’t cutting out a bunch of people who are just as smart for reasons that are purely economic or egotistical.)

        I don’t think I need to go see the top 1% doctor to diagnose a skin condition that can be treated with a common antifungal or whatever.

        Frankly, nurses could likely do most of that basic level work at an acceptably adequate level.

        • Edward Scizorhands says:

          I’d like see a break-out of how many jobs in health care are “O-ring jobs,” where what you need is every single person to not screw up.

        • rahien.din says:

          Heck. If I gave you a derm textbook and an array of topical treatments, you could probably do it for yourself.

          The increase in error rate will not show up in routine visits. It will show up in the visits where there is more potential for harm.

          If I misdiagnose your fungal dermatitis, it’s not really a big deal. If I wrongly give you albuterol and send you to the ICU with V-Q mismatch, or I give you the wrong drugs and you’re stuck with tardive dyskinesia, or I put a Dravet patient into status epilepticus with phenytoin, it’s a big deal.

          • Charlie Lima says:

            Ehh, having seen one case of Steven Johnson Syndrome due to topical antifungals I would amend that it is usually not a big deal.

            The big thing physicians can do that nurses have more trouble with is know when to get worried when the unexpected happens. Most of the time the patient gets better regardless of what you do. It is the rare times when things go sideways that you want a physician who can spot the problem and take care of things before it becomes an ICU stay.

          • rahien.din says:

            Oh you’re absolutely correct. Just trying to abide by Winja’s groundrules.

            But yeah. “Routine” is a retrospective description. And medicine is all about forecasting.

      • Garrett says:

        > Why do you want to give me a bunch of colleagues who are potentially worse than I am.

        We already do that in the United States when admissions decides to care about background, race, whatever, or anything other than competency. Moreover, I’m not certain that there’s any reason to believe that there’s an obvious capabilities cliff where provider level drops, and even if it exists, that it matches up with the current cutoff for medical school and residency admissions.

      • Scott Alexander says:

        Disagree, you need to consider elasticity of quality with respect to supply (I may not be using that term right).

        I used to do interviews for a low-prestige psych residency. Everyone was amazing. The failed candidates were amazing. The worst candidate we got was still amazing. There were no non-amazing people anywhere near the system. If we double the residency pool, are we really going to be letting in dumdums who can’t read medication labels correctly? Or does it mean we can finally take that Bangladeshi MD PhD who’s been driving taxis for the past five years because he can’t get a US medical position?

        I know I got into medical school because I had a philosophy degree and my medical school thought it was cool to have some humanities people in it. I mean, I was qualified and I had good grades just like everyone else, but the thing that got me accepted over the next candidate was the philosophy degree and the administrator thinking that was cool. Once you are making decisions along those lines, I think you no longer get to say that any loosening of supply restrictions will kill people.

        • rahien.din says:

          I mean, sure. Take the extreme case : only one applicant per year is accepted to medical school in the entire USA. That person would have to be the next Osler. But a great many near-Oslers would have been rejected.

          But it’s not as if the top 10% of applicants are homegeneously Oslerian in their intellect and conscientiousness, and separable only by the charmingness of their degrees.

          Take me : I got into med school because I was an engineer and slayed the MCAT. This allowed me to be accepted despite a crummy semester that ruined my GPA (IE, evidence of poor conscientiousness). And I was in the top 1%.

          What would it take for me to have been a top-10% applicant, rather than a top-1%? A worse MCAT? Two bad semesters? An uninteresting degree? Would I have been as safe of a bet? As good of a doctor (presuming I even am a good doctor)?

    • block_of_nihilism says:

      I think the most complete resource for physician salaries is the annual Medscape survey. Based on this, the average primary care physician (PCP) makes $237,000. Averages are tough though, since salaries vary in ways that one might not expect. For example, the highest-paying states for physicians are not the highest cost-of-living areas- Oklahoma, Alabama, Arkansas and Nevada are the top four. Anecdotally, my friends who are residents have told me that, due to much higher demand in rural areas, one can make substantially more money working in those areas as a PCP.
      So yeah, I think 5x is more typical, particularly for primary care physicians. Some specialists make crazy amounts of money though; some surgical subspecialists (orthopedics, plastics) apparently make an average of close to $500,000 a year, which is pretty crazy.
      One thing to add here is that the opportunity cost for becoming a physician is higher than most other professional careers. Not only does it entail a minimum of 8 years of post-secondary education, it also means a minimum of 3 years working at a resident’s wage (typically $50-60,000/year), which per hour often works out to close to minimum wage. Add in the longer hours, higher stress and unique responsibilities of being a physician and I don’t think 5x median wages is too out of line…

      • Frederic Mari says:

        I’ve heard US medical personnel say that with regularity but why are the European doctors, whose residency and post-secondary education are just as long and probably as tough, happy to do the work for 2X rather than 5X.

        In France, you become a doctor for the prestige, the near certainty to end up comfortably bourgeois/upper middle class and being instantly well respected in the community. But it’s well understood that, short of starting your own clinic or doing something unusual, you’re not going to end up rich.

        Now, it’s true we have to “borrow” doctors from our ex-colonies (since being a doctor/nurse in France beats being a doctor/nurse in a developing country apparently) so maybe we should review our system a bit but, overall, it works at 2X, without going 5X.

        • rahien.din says:

          We may not be comparing the same things via median salary. As others have pointed out, salary != disposable income != standard of living != satisfaction.

          Just as money has to reduce down to stuff, compensation has to reduce down to satisfaction. A good example thereof :

          In America, physicians get paid based on productivity (time, risk, complexity). So in America, the most prestigious specialties are things like neurosurgery, interventional radiology, orthopedics, and cardiology. Those specialties bring in a lot of money, so that’s where all the smart people go.

          In India, all physicians get paid basically the same salary. So in India, the most prestigious specialties are things like general pediatrics. Those specialties provide the best lifestyle and satisfaction, so that’s where all the smart people go.

          So instead of assuming American doctors are just greedy… you might ask instead : what sucks about being a doctor in America, compared to being a doctor in France?

          • Frederic Mari says:

            I’m not sure how your point undermines the fact that US doctors are “greedy” (TBH, I don’t like that term. It’s about structural forces and institutional power being wielded. It’s not about morality).

            I’m sure each system has its subtleties and, within a given system, some choices make more sense than others if you’re trying to maximize money.

            But one system still ends up paying its doctors significantly more than the average earner. That… can’t be explained by productivity differences. It’s about the ability to extract surplus.

          • rahien.din says:

            within a given system, some choices make more sense than others if you’re trying to maximize money.

            There are systems in which it is impossible to maximize money, but doctors are still satisfied. There are systems in which it is less possible to maximize money, but doctors are still satisfied.

            This demonstrates that :
            1. maximization of money is not always necessary for maximization of satisfaction.
            2. doctors are not really trying to maximize money as an end in itself

            one system still ends up paying its doctors significantly more than the average earner. That… can’t be explained by productivity differences.

            Of course not. What makes you think there are productivity differences?

        • disluckyperson says:

          Instead of comparing doctor’s incomes to the median income, it might be better to compare them to the incomes of other high earners. By this comparison, American doctor incomes are not unusually high, and the international discrepancy (5x vs. 2x) can be explained by the fact that the gap between highest earners in the US (including non-physicians) and average earners in the US is much higher than the gap in other countries. So rather than being just a discrepancy between doctors in the US and other countries, it’s a discrepancy between all high earners in the US and other countries.

    • edmundgennings says:

      One reason why this comparison is harder is that doctors spend time and money in schooling(and sort of residencies) early in their career. Because of the time value of money, this is more extreme. Thus the present discounted value of $150,000 a year for the rest of one’s working life straight out of college as an engineer might be equal to the present discounted value of going to med school doing residencies and then making $250,000 a year. The exact comparison is heavily dependent on the discount rate and how much help one gets with schooling.

    • Clutzy says:

      High doctor salaries probably drives things like high hospital admin salaries (they often have MDs as well).

    • AJD says:

      In essence it seems like a combination of transparency, civic mindedness, good living and social accountability are mostly responsible for the Amish advantage. Not a bad lesson for us English.

      WRT pricing, it seems like the lie of “non-profit” hospital status, the insertion of the profiteering middleman and various artifacts of “pure capitalism” (pharma pricing, patent shenanigans, lack of regulatory vigilance, frivolous and/or exorbitant malpractice suits, inapt distribution of medical specialists because of skewed incentives) all work together to keep pricing high and escalating. In all, a tightly coupled and complex system… it feels very insoluble without wholesale reinvention.

      Now I feel encroaching despair. Better call a shrink! 😉

    • John Richards says:

      “no law suits means that Amish must absorb the costs of malpractice themselves (and are possibly at greater risk)”

      It seems to me that the Amish wouldn’t really be at much greater risk of malpractice than anyone else. Does a typical doctor, upon learning that they can perform a surgery guaranteed that they won’t be sued for malpractice, breath a sigh of relief and say to himself “well, this one I can do kinda sloppy”?

      • Alkatyn says:

        Maybe not the typical doctor, but if you were a doctor with a history of getting sued for malpractice, or were corrupt in some way, you’d love to have a community of patients who were guaranteed to never sue you

        • The Nybbler says:

          Indeed, but I suspect if you commit malpractice or are corrupt in some detectable way, word will spread around the Amish community and you’ll find you have no patients pretty quick.

    • Alkatyn says:

      having the minister handle the administrative affairs looks free, but his time is still costly;

      Also, the costs that happen when you’re ill in the kind of way that the minister doesn’t approve of. I strongly suspect its not just snowmobile accidents, but things like mental health where the traditional attitude is “tough it out”.

      Another implicit cost is that people with chronic conditions, the disabled, and the elderly will be having care from their immediate family presumably, rather than paid care.

    • Etoile says:

      So my sense is, the other countries which pay doctors less have to import theirs from other countries continuously, in part because they don’t pay doctors enough. If we try to address the doctor pay in a regulatory manner, we’re probably going to overshoot or distort in some even worse way.

      Honestly, this “THOSE DOCTORS MAKE TOO MUCH MONEY”, from a visceral perspective, is tiresome; there are lots of doctors who are not the most competent, sure, but honestly — is there a more important job out there? Do you really think a doctor should be paid, say, only three times than, say, a “retail salesperson” (the most common job in the US in May 2019)?

      • Del Cotter says:

        Other way round, I think. They don’t have to import doctors because they pay doctors less: they can pay doctors less because they import doctors. The wanting to pay doctors less came first, and the importation was an opportunity, not a subsequent and unexpected forced consequence.

        I suspect this because it’s what they did with bus drivers and cab drivers. First came the complaint that the drivers were being paid too much (for work the complaining class hired, but would neither care to do, nor care to encourage their children to seek a career doing), then came the importation. Sometimes it was official (one of Enoch Powell’s jobs was going round the Caribbean drumming up interest) and sometimes it was batched (a batch of workers would come over in one boat).

  2. havel says:

    This is happening with veterinary services in Australia with the introduction of pet insurance. Nil deductible is reducing the scrutiny that owners place on accepting services for their pets, premiums are rising, and suburban vets now do many more surgical procedures and mostly drive German cars.

    There is something important about the construction of insurance (or any contract that separates the payer from the beneficiary) in setting the deductible and limits relative to the personal risk. The right level of deductible is a major contributor to avoiding perverse incentives, and allow the risk pool to avoid complex conditions and expensive claim adjusting.

    We need some incentive to avoid risk.

    • Jeff R says:

      I think that’s good point. I have a high deductible plan paired with an HSA. My annual healthcare spending, on average, is probably in the tens of dollars. Of course, I’m still relatively young, though, so that could obviously change.

  3. Sniffnoy says:

    ooking into this issue, I noticed glimpses of a different possibility. The increase started around the same time that health insurance began to spread. In one sense, this is unsurprising – of course health insurance would become a thing around the time care became unaffordable. But I’ve never seen someone really try to tease out causality here. Might the two trends have been mutually self-reinforcing? The price of care rises due to some original shock. Someone invents health insurance, which seems like a good idea. But this creates a series of perverse incentives, which other actors figure out how to exploit (eg the Lucemyra example above). Insurance-based-health-care becomes less efficient, but hospitals can’t or don’t internalize this to the insured patients – they just raise the price for everyone, insurance or no. That makes even more people need health insurance, and the cycle repeats as prices grow higher and higher and insurance becomes more and more necessary.

    I’m surprised you hadn’t considered this possibility before; it seems to be brought up here a bunch. Indeed, I seem to recall someone once linking a science fiction story that predicted this mechanism (note: not actually clear on the timing here, maybe more discussed than predicted), though it was about a hypothetical form of car insurance (one that worked more like how health insurance works, rather than one that works like how real-world car insurance works), and I can’t find it at the moment.

    Mind you, I have no idea to what extent this is actually true, but I do want to make some notes expanding the hypothesis all the same: (note, none of this should be construed as original to me and all of it could likely be stated better by other people)

    1. Hospitals can’t or don’t price discriminate regarding insured vs uninsured — my understanding is that when they don’t it’s because they can’t; they have to quote one price, see. They can’t openly state the real price and then raise it for the insurance companies, they won’t go for that. So instead they have to openly state the insurance price as the real price. My understanding is that they then try to practice price discrimination by offering discounts to the uninsured, but this won’t consistently work (and you have to know to ask for it). But meanwhile the upfront price they quote has to be the inflated one, or no insurance company would pay it.

    And then part of this of course is that prices just get kind of obfuscated and hardly anyone can tell you what anything costs; patients are making decisions without knowing prices at all. So that adds to all of this. Like the simple sort of fee-for-service model, with clearly stated prices so that the patient can make an informed decision, that you might see in, IDK, Kenya or somewhere, doesn’t really exist here (though it does seem some people are trying to bring it back?). And so because of that you basically have to be covered instead! And when someone has a hospital bill they can’t pay for, it’s not because they foolishly paid for a procedure they couldn’t afford, but because, well, they honestly had no idea what it would cost; because even if you ask, likely noone can tell you.

    2. A big part of the problem seems to be that, as I mentioned, health insurance does not work like car insurance — it’s more of a prepayment plan than insurance in the usual sense, and things that do work like insurance in the usual sense will get into legal trouble (whether because of Obamacare or because of older laws; I’m not too clear on the details there). As in, instead of the insurance company paying you what they think it ought to cost for a certain thing, they pay whatever the doctor says it costs (although they will refuse things that are too expensive, and do negotiate as you mention, and of course patient still pays coinsurance and deductibles; so it’s not entirely like that). Thus the incentives problem you mention — lots of the “other people’s money” problem.

    I think a key thing here that gets overlooked a lot is, AIUI, health insurance companies do this because they have to — I’m not clear on the details here, but balance billing, the practice of saying, here’s what we’ll pay, bill the patient for the rest (or, from the doctor’s side, which is likely more important, the practice of saying, OK, guess that’s what the insurance is willing to pay, I’ll bill the patient for the rest) is often illegal. Medicare definitely bans it; they will not let you bill your patient for the balance. And I think other insurance companies may now be banned from doing it as well? I have a definite suspicion that getting rid of this, just letting doctors bill the balance, would make the healthcare market substantially more sane.

    Like why is this restriction even there? I have to imagine it’s there basically for convenience, so people don’t have to think about these sorts of bills, because it looks bad when insurance pays for something but also you get a bill. But as is so often the case, trying to make things look good makes the reality much worse…

    Anyway other people can probably make these points better than me or get right the things I got wrong but thought I should post this anyway. Like I said, not clear on how much this is actually true, just thought I should point out some natural other points in the hypothesis you suggest.

    …of course, if we’re talking about reasons healthcare is expensive, we can’t forget about the AMA and the ADA and their constant suppressing of supply, but…

    • Said Achmiz says:

      The story is Christopher Anvil’s “Positive Feedback” (which appears in the story collection Prescription for Chaos). It was published in 1965.

      • teageegeepea says:

        I had never heard of that before. I have however read “The Bladerunner”* by Alan E. Nourse, a scifi writer better known in his time as a medical doctor (since he also wrote about medicine). It depicts an alternate future in which government provision of healthcare resulted in the system being overwhelmed by demand and a system of eugenics laws being applied to anyone who receives care, which winds up causing another disaster during the “present day” of the novel due to all the people who avoid the official system in favor of the black market when a pandemic hits. I wish I could link to my review of it, but that Disqus page no longer exists.
        *The film got its name from William S. Burroughs’ attempted adaptation “Blade Runner (a movie)”, even though the plot is instead based on PKD’s “Do Androids Dream of Electric Sheep?”.

      • Sniffnoy says:

        Thanks!

      • mobile says:

        This was terrific.

    • spacecommie says:

      “Hospitals can’t or don’t price discriminate regarding insured vs uninsured — my understanding is that when they don’t it’s because they can’t; they have to quote one price, see.”

      It’s true that the sticker price stays the same, but there’s nothing preventing hospitals from quoting you the price minus a self-pay discount (or, for that matter, the cost out of pocket after your insurance company ponies up—they keep fee schedules on file). There’s pretty good odds you can go to the website for your local health system and get the real price of an appendectomy or whatever.

      In any case, there’s a bunch of price transparency regulations coming down the pike, so this will simply not be a feature of the market in a couple years, inshallah.

      • Scott Alexander says:

        There was a mention in one of my sources about how one hospital was willing to discount the Amish down to the price of Medicare, but not further, because if you charge anyone less than you charge Medicare, you can get sued for Medicare fraud. I don’t know how many things like that there are.

      • etheric42 says:

        My (unprinted, unadvertised, would-be-unknown if I hadn’t figured it out accidentally) options at the mega-practice I use (rarely) for general family-practice visits:

        A) Give them my insurance. I get a bill in a month for ~$210 because I’m on a High Deductible Health Plan that I need to pay in full, but goes towards my deductible it is incredibly unlikely I’ll meet unless I have a big unexpected health issue.

        B) Tell them I’m self-paying but will need to pay over time. I get billed for $250 and can string out payments as long as I want.

        C) Tell them I’m self-paying and can pay up front. I pay ~$100 immediately.

        This seems like a big multiplier over self-pay, although I honestly have no clue how much of that self-pay discount is a charity discount. This also makes it difficult to make a rational decision here. If I pay via insurance I’m paying significantly more if the year goes according to plan, but it’s effectively free if I end up having a bad health episode during that calendar year.

        Also, even though I have a choice which insurance company I go with because thankfully my employer doesn’t offer insurance, there is no realistic way to shop for which insurance company has better negotiated prices (and that seems an important part of the benefit when using HDHPs, which in my area are pretty optimal for most cost scenarios).

        • pochti says:

          I honestly have no clue how much of that self-pay discount is a charity discount.

          I would guess not much: if charity is motivating them, it doesn’t seem like they’d be charging 2.5x as much for people who can’t afford to pay $100 up front.

    • JPNunez says:

      Health Insurance does need to be all bad. The japanese system also uses Health Insurance. I dunno what the difference is; private health providers need to be non profits and prices seem to be regulated so there’s probably some stoppage to any price spiral.

      • teageegeepea says:

        By “health providers”, do you mean hospitals? Because I think they are mostly non-profits in the U.S. Blue Cross Blue Shield is also a non-profit, although it’s also a network of insurance companies which includes some for-profit ones.

        • JPNunez says:

          There must be something more insidious then. This reply only reinforces the idea that there’s nothing wrong with health insurance per se, but something exclusive to the american system.

    • etheric42 says:

      The balance billing restriction is pretty important. There are currently three “tiers” of balance billing I can think of. One prohibited, one permissible, and one currently being legislated out of existence in many places.

      The first is that in-network doctors can’t balance bill. As far as I understand it that is just a condition of being in-network and an aspect of their “pre-negotiated rates”. After all, if they are agreeing to the fee schedule the insurer sets up, why would there be any balance left?

      The second and third tiers are seeing out-of-network providers. If you know you are going out-of-network, why would you expect your insurance to cover everything? Well except your insurance says it covers a certain percentage of out-of-network bills (and in fine print clarify that’s only a percentage of “usual and customary” charges and it is really difficult to get the doctor to tell you what they are actually going to charge you and the insurer to clarify what what their usual and customary amount is, so it is kind of misleading to tell you they are covering 50% when it is really just whatever number the insurer comes up with). But theoretically if you have time to plan seeing the out-of-network provider then you have time to investigate the costs and it’s fair (Texas’s new law is intended to allow balance billing if the patient signs paperwork in advance and there is a waiting period between the paperwork and the service).

      The tier that is in the process of being banned due to national outrage is surprise balance bills. I blame hospitals and insurers for allowing it to exist, and I’m disappointed that various levels of government had to step in. To be fair, I’ve seen more insurance policies lately with surprise balance billing clauses to protect their customers, and I don’t know if this is due to additional regulation or is being done by the insurers independently to supplement the laws being passed lately.

      Anyway, I may be a little biased because I was hit with one of those surprise balance bills a couple of years ago. Our first kid was at home with a midwife and everything went without issue. The second kid was also planned that way, but we ended up having to be transferred to the hospital due to complications (everything worked out fine though!) With 9 months warning and not being my first rodeo, I thought I had done everything right. Midwife’s full fee paid by me outside of insurance (a rant for another time). Hospital nearby identified as in-network for insurance and enough money in the bank account to pay a the maximum amount for in-network service.

      Then the hospital bill came. Everything as expected. Paid in full. Then the ambulance bill came. Surprising, I didn’t think ambulances were in or out of network, but this one was out of network. A couple grand to drive a mile, but not the end of the world. Then the doctor’s bill Out of network. Out of network deductible+lower coverage+balance billing to the tune of 3-4x the maximum I could have paid in network. There wasn’t even a C-section, or complicated procedure, or pediatric intervention needed. My insurance already represented ~20% of my monetary annual compensation and at the end of the day I’m left with bills for another 30%+.

      I am glad the doctor was there, and I am glad he was experienced enough to not require any more intervention than breaking the waters and a mid-push twist. I am glad the hospital could afford to have an OB on call. But I needed to know what I did wrong so I could have prevented it in the future. The obvious one would have been to have planned a hospital birth from the beginning, which would have eliminated the ambulance, midwives, and out-of-network OB. It’s frustrating that the more resource-consuming solution would have been so cheap to me personally, but such is life sometimes. Then I found out even that wouldn’t have completely insulated me from the surprise balance bill.

      Just because the hospital was in-network doesn’t mean the doctors at it or on-call for it are. Just because you show up with your OB doesn’t mean your anesthesiologist will be in-network. Or if your OB is late arriving. Or if they need a specialist in a pinch. I asked the hospital what I could have done differently to make sure I only go to hospitals that not only are in network but have in-network providers. They told me there wasn’t anything. That I could have called that night before coming, asked who was on call, and then researched my plan. But they strenuously recommended not doing so in an emergency like we experienced (and there would have been nothing stopping the call schedule from rotating by the time we came in). I called the competing local hospital chain. Same answer. I called my insurance provider. Same answer. There was effectively nothing I could to to stop my “out of pocket maximum” from turning into a figment of the imagination.

      Thankfully we ended up okay. I put on my D&D hat and min-maxed in order to retroactively qualify for Medicaid and ended up having them pick up all of the bill (and even refund me the money I paid to the hospital). Sorry fellow taxpayers. We were on Medicaid for the first one and even though we didn’t use it we were trying to be more responsible for the rerun. We gave up on home birth after that and we just scheduled an OB for the third one, but then pandemic canceled that plan as staying out of the hospital sounds like a better idea so here we are back to home birth. At least the state now has surprise-balance-billing laws now, and strangely Medicaid was beating down our door this time insisting we take it even though I make significantly more now than when I had to fight the state for it. It’s funny how things work out.

      • albatross11 says:

        This kind of balance billing should be illegal, and as I understand it, it would be illegal if not for strenuous lobbying by the current beneficiaries.

        • etheric42 says:

          I still put most of the blame on the hospital and insurance company.

          A) Why are the doctors on call at the hospital not hospital employees? Or at least required to have the same relationships with the insurance companies the hospital does? Or (and this seems to make the most sense) be a contractor of the hospital such that the hospital bills the patient/insurance company according to its policies and then the doctor bills the hospital? If the hospital exists to care for the sick (as many non-profit and/or religiously-oriented hospitals are) why do they ever accept this kind of relationship that they know is so harmful to their patients/customers? No offense to sex-workers, but why are hospital doctors more akin to strippers who rent the pole? What other industry has that kind of billing model?

          B) I’m the insurance company’s customer. The insurance companies (particularly the big ones) have all sorts of leverage to extract agreements (it’s why the “negotiated rates” exist). Why would they ever allow a hospital to be in-network unless they had a contract with the hospital that barred them from facilitating these shenanigans?

          Instead I’m having to rely on the state government to ban it. Which means the first law they passed barely addressed it, the second law they passed on paper addressed it but left a loophole for the providers to use, and they are looking at having to pass a third law to close it next session (if they remember to do so after the current crisis).

          • Garrett says:

            > If the hospital exists to care for the sick

            I’ve had conversations like this elsewhere. My conclusion is that for historical reasons, that isn’t really what hospitals exist to do. They exist as a high-tech hotel where it’s safe to store patients between procedures or during treatment. So instead of a 24/7 concierge you have a 24/7 nursing staff, etc. In-general, hospitals employ almost no doctors. Instead, they contract with medical practices. Almost nothing in the healthcare industry works like you’d expect it would. It’s weird.

          • etheric42 says:

            @Garrett
            Any clue what causes them to be a high-tech hotel? What are the incentives driving it? Is there an economic advantage to them being this way? Is there an opportunity for a hospital that employs doctors to disrupt this, and has that already been done?

            I worked in a 24/7 answering service for doctors for years. The relationships I discovered were strange and difficult to work with. I loved working with the bigger, corporate-owned clients, but even they had their challenges between the doctor hierarchy and the administration hierarchy.

          • mfm32 says:

            If you are assigning blame, assign it to the doctors.

            Physician culture in the US has long been highly entrepreneurial. Physicians today are much likely to be employees than in the past (and some certainly prefer that), but in the US a large part of the allure of the profession is the independence and economic potential of being in private practice. Only very recently have the plurality of physicians not been equity holders in the practice that employs them. And even that overstates the number of physicians who are truly employees in the way you are describing, because some of the employee physicians are employees of a small group of owner-physicians. Some data is here if you’re interested. The AMA promotes private practice and always has. Protecting the economic interests of doctors as business owners was one of the organization’s founding goals.

            Doctors clearly benefit economically and non-economically from private practice. It is not at all clear that it serves patient or broader public health interests, except perhaps in some very convoluted way. I also see this interest cited as one of the major political obstacles to many forms of health care reform in the United States, including anything that looks like single payer.

            Do other countries have a similar model? I know the UK has private doctors, but they almost all also work for the NHS, right? Does anyone have experience of how that works in practice.

  4. Some Troll's Serious Discussion Alt says:

    . If occasionally scamming providers meant insurance companies had to pay more money total, then they would stop doing it.

    That definitely seems like a feat of coordination that would be beyond them.

    • Scott Alexander says:

      I don’t think it requires coordination – if a certain insurance company were known to be honorable, it could probably get the same deal the Amish do.

      • aristides says:

        Agreed, I switched health insurance from TriCare to Blue Cross Blue Shield, and it was incredible to see how many more provider options I had. Doctors have a tendency to be fairly selective with the health insurances they accept, but it seems that what they value is reimbursement rate over anything else. Pretty much the only Tri Care Providers I were able to find only did it for patriotic reasons.

      • Nicholas Weininger says:

        If social-trust-related scam and suit risks are major contributors to the cost gap, we should observe that other communities dominated by high-trust subcultures (Mormons, Orthodox Jews, others?) also have lower insurance premia. Do we in fact observe that?

  5. Jack V says:

    I don’t want to get too simplistic but this sounds like the NHS, or to a lesser extent other mostly nationalised systems. I don’t know what a good analysis would be: maybe compare different examples of something like pet insurance where more different options can exist, and comparing the costs.

    If hospitals just provide healthcare then most of the problems the Amish avoid, everyone avoids. Medical staff have a culture of providing treatment so don’t try to nickle and dime patients. Patients don’t have to pay, so don’t have the “get my money worth” attitude. There’s SOMETIMES a problem with “patients asking for too much stuff because it’s free” but not usually. Most of the overhead America labours under goes away: everything padded to cover other treatments; exaggerated temptation for malpractice suits and insurance against it; everyone employed by health insurance companies, and all the time everyone in a hospital spends dealing with them; all the effort to track bills and have different tiers and chase down payment. Plus, of course, the NHS can buy drugs effectively.

    PS. I should acknowledge that the Amish system like any other institutional system probably sucks for anyone who DOESN’T fit into Amish society because the people in charge may be less likely to believe them about what’s wrong. But I assume that’s only a small proportion of the overall costs.

    • Lambert says:

      Also the public seems to see the NHS as a vary big community health system, rather than a faceless organisation to be swindled.

    • exaggerated temptation for malpractice suits and insurance against it;

      Why would a nationalized system work any differently? If you want to make it harder for patients to sue doctors, you can just do that.

      everyone employed by health insurance companies, and all the time everyone in a hospital spends dealing with them; all the effort to track bills and have different tiers and chase down payment

      Nationalized systems have to deal with billing, it’s just not the customer being billed. All that bureaucracy is less visible to the user, but it’s still there.

      • zzzzort says:

        Why would a nationalized system work any differently?

        Social pressure is more effective than legal obstacles.

        All that bureaucracy is less visible to the user, but it’s still there.

        Some of that bureaucracy is still there, a lot of it has gone away. You don’t need to get insurance/payment info for every patient. You probably don’t need to track which patient got that bandaid. The pre-approval process is less onerous, and there are more situations where a doctor can approve things on their own.

        • Social pressure is more effective than legal obstacles.

          This social pressure exists in nationalized systems? First I’m hearing about it.

          You probably don’t need to track which patient got that bandaid.

          Of course you do, otherwise doctors will charge to serve non-existent patients.

          • thisheavenlyconjugation says:

            Of course you do, otherwise doctors will charge to serve non-existent patients.

            If that were a concern you would still only need to track how much of each doctor’s time each patient consumed, not how which resources. But I’m not at all convinced that it is a concern. How money does the average doctor committing fraud in this way get (in a system where we aren’t formally tracking anything)? What proportion of doctors would commit fraud? My guesses would be 20% and 1% at most, in which case if your tracking system costs more than 0.2% of doctors’ salaries it’s not worth it.

          • mfm32 says:

            Of course you do, otherwise doctors will charge to serve non-existent patients

            This is not how the NHS works. It’s not a national insurance policy. It’s more like a national healthcare provider. I believe the bulk of an NHS doctor’s income comes from the number of patients they serve as primary GP. They don’t get paid more for providing more services, in the main, which is closer to how the US system works.

            You still could commit fraud through non-existent patients, I suppose, but you’d have to invent whole identities and get them assigned to you as a GP. That’s a lot more difficult than the typical Medicare fraud of inventing procedures. The administrative burden should be lower as a result. It would be closer to (in many ways, easier than) policing Social Security OASI fraud, which I think is pretty rare.

      • 10240 says:

        The government has more incentive not to make malpractice lawsuits too easy, as it would eventually become the one who (indirectly) pay the cost. Or, rather, it’s taxpayers, but raising taxes makes the government unpopular. Private insurers raising insurance premiums just makes the insurance companies unpopular.

      • NostalgiaForInfinity says:

        The US has substantially more healthcare bureaucracy as a % of expenditures than other countries.

        https://www.americanprogress.org/issues/healthcare/reports/2019/04/08/468302/excess-administrative-costs-burden-u-s-health-care-system/

    • Charlie Lima says:

      The US has several nationalized health systems for various patient populations the size of small countries. The Indian Health Service and Tricare, for instance are operated very much like the British NHS. Yet their costs are not terribly out of line with other US healthcare systems. Likewise, their overhead costs are not spectacularly cheaper the other systems in the US.

      If it were as simple as changing the ownership and payment models, somebody would have done it already.

      • Scott Alexander says:

        Do you have any numbers for how much Tricare and IHS spend per person per year?

      • thisheavenlyconjugation says:

        You might expect them to have higher costs if they serve less healthy populations. But also, they have to compete with other healthcare systems on salary.

        • Charlie Lima says:

          1. Tricare is disproportionately young and fit, pretty much all of them are fit for active duty or are the dependents of folks who have to do physical training for work. Could be wrong, but I doubt the active or the reserve military forces and their families are particularly high on comorbidities.

          2. Well of course they have to bid higher. Just like how the healthcare system in the US has to be against Fortune 500 companies for talent. Your average physician is an excellent manager these days and has plenty of other options.

          3. If we enslaved all the docs that would still get us, maybe, a 20% reduction in healthcare costs. Whatever terms you hope to force onto physicians, we still are not looking at much in the way of savings. Allegedly going to single provider is supposed to cut overhead and allow for more rational care. But this is not what we see with American single payer systems. Tricare can literally order its patients to be healthier, under threat of prison if they fail. Yet its overhead and cost to provide services is not that far from a bargain basement ACA plan.

          There is no quick fix for US healthcare costs. If there were, somebody would have implemented it in Tricare, IHS, Kaiser, or any of the other nation-sized healthcare systems in the US.

          • disluckyperson says:

            Kaiser is the best model for reduced administrative costs that M4A folks are toting, because they are entirely self contained, ie their doctors and hospitals don’t have the administrative burden of dealing with a multiplicity of insurers. And indeed according to this they have the best “administrative ratios” in California. But yet, as you say, we don’t find that the prices of Kaiser are much cheaper than other insurers. Maybe a bit cheaper, but certainly nothing close to the nationalized health systems of other countries.

          • thisheavenlyconjugation says:

            1. I thought Tricare treated military retirees as well as active members, you are quite probably right that the overall population is healthier than average though.

            2. Competition within the same role is much more important than competition between roles, as can be seen from the fact that we tend to talk about salaries of doctors and teachers and lawyers, not “managerial roles” which have much greater variance.

            3.

            If we enslaved all the docs that would still get us, maybe, a 20% reduction in healthcare costs.

            Do you have a source for that?

          • Charlie Lima says:

            THC:
            1. They do, but not that many of them (most military members never retire which requires many years of service). Even when they do, a lot of care still ends being provided by the Veteran’s Administration.

            2. Perhaps, but we do talk about broad competition back when students are picking careers. Becoming a physician in the US is one of the hardest, most expensive, and longest gratification delayed paths in the country. If you are going to reduce their compensation from 4.5x median per hour to maybe 3x then a lot of those other career paths are going to come into play. Likewise, for a lot of the senior docs, reducing their hourly compensation will almost certainly result in reduced hours working, retiring early, or dropping the more annoying patients/cases.

            Again though, I do not see how going single payer is going to change salaries quickly. Everywhere else just grandfathered in docs at their current salaries and then slowed salary growth in real terms.

            3. Well Scott says here as well. But the magic comes from Health, United States, 2018 by the CDC. In that they state that “Physician Services” account for 23.4% of health expenditures (page 23). This is an upper bound as physician services include a lot other non-physicians’ salaries. A lower bound would be something like there are 860,000 physicians in the US (per the WHO) with a mean salary of $313,000. Total physician pay is $269 billion with around $3.5 trillion in total healthcare spending. This gives us a lower bound around 7.6% of total healthcare spending (i.e. a lower relative rate than the UK, Sweden, Germany, and basically everyone else).

            If single payer is going to dramatically lower costs, say by a third, it cannot do so just by bidding down physician salaries. It likely cannot do so even if you slapped a 25% across the board pay cut on everyone down to the janitor. You would need to fire people. Lots of them. And hence bringing American staffing levels down to European levels. That means fewer administrators, but also fewer nurses, nurses’ aids, and all the rest (oddly enough it would likely require more physicians).

            And that is why IHS and Tricare are such bad data points for saving buckets of money by going single payer. Getting to anywhere close to OECD average is going to require a lot more than pay cuts. Yet the places that should be able to most easily do this have not managed to make it work.

          • thisheavenlyconjugation says:

            @Charlie Lima
            Interesting, thanks. Do you know what the “Hospital” category covers in that report?

          • Charlie Lima says:

            THC: if memory serves that is basically all charges for hospital based care except the cost for the physician. Some costs are squirrely (e.g. the surgery scheduler may be rolled into the physician’s fee or into the hospital fee). I am not sure if they ever got around to properly reporting outpatient visits that occur on hospital grounds.

      • mfm32 says:

        Is TRICARE reimbursement capitated? I understand it to reimburse providers similar to the way a typical private insurance plan would, except at lower, Medicare-like rates. If so, it lacks some key features of the NHS model that proponents argue should lower costs.

        There is also the problem of anchoring. If a small nationalized health system (which I don’t understand TRICARE to be) exists in a market where the dominant model is private health insurance, the cost of doctors, other medical staff, and medical equipment and supplies will largely be set by the private health market. This is not a case where it’s obvious that a defector from the dominant model can reap the benefits that a near-monopoly like the NHS arguably enjoys.

        There is a related anchoring point on the patient side as well. Patients in the INS or TRICARE have a ready point of comparison to what healthcare in the US “should” look like. To the extent that cost controls in other countries work by lowering the (perceived) quality or quantity of care, then, it will be more difficult for a system like the INS or TRICARE to exercise similar controls. As an example, even though Medicare reimburses at relatively low rates compared to private plans, it is effectively prohibited from excluding drugs from its coverage. I think there is a strong argument that these sorts of controls are a material driver of cost savings in the US.

        • Charlie Lima says:

          Anchoring is going to continue though. For at least a generation, physicians are going to recall what a “fair salary” is and demand that. If you try to cut them under some nationalized scheme, you are likely going to pitting very unpopular politicians and bureaucrats against exceedingly popular doctors (“she saved my life”). And this is what England saw. The NHS did not lower salaries, it slowed their growth over decades while increasing their duties, again over decades.

          Likewise, nobody selling Medicare-for-all ever sells it as having truly less care. Oh they may talk a bit about private rooms or fewer brand new treatments. Nobody talks about having waiting times for elective surgeries where you get to live in pain for months in order to reduce the amount of capital tied up in surgical centers. And again the NHS did not reduce the level of care provided relative to the US overnight, they did it over decades by refusing to pay for new and expensive things until they became old and cheaper.

          Anchoring pretty much damns any US single payer system regardless of how implemented for at least a generation. Maybe when everyone who remembers the “good old days” is dead it would work. But again, a quick fix it ain’t. And even that is dicey.

          Lest we forget, we have monopoly models of government service provision in the US. Has this reduced the salaries of teachers? I have yet to find a locale in the US where the majority of private schools are cheaper than the public ones or where K-12 education is overall anywhere near as cheap as those in Europe. Or take subways. These are always government monopolies with a complete monopsony on any number of jobs, yet somehow US subways (be it NYC or DC) end up being far more expensive than those in other countries.

          For better or worse, Americans believe they are due certain things and even when you do have a monopsony for hiring, somehow we end up having illegal teachers’ strikes or something to bring wages and costs up to something more expensive than anywhere else in the world.

          Beyond unreasonably high American wages, one of the allegedly key benefits of single payer is the reduction in overhead. Supposedly we burn huge amounts of money by having insurance companies divvy up the market. Yet the overhead burden for IHS and Tricare is not all that much lower, certainly not near European levels, even though you get access to the former by dint of your ancestry and the latter because you join the military.

          And all this fits very nicely with the complete inability of any of the states to implement single payer. Vermont, one of the most left-wing states in the country, tried and could not make the money work. It went so poorly that they actually elected a Republican as governor shortly thereafter.

          My guess is that American healthcare is not expensive because of its org chart. My guess is that it is expensive because it employs Americans to treat Americans.

          • Edward Scizorhands says:

            Largely agree. Doctors have set their lifestyles to their current salaries. This doesn’t mean they deserve those salaries, but it means they will fight like hell against attempts to break their current lifestyles. And they are better at lobbying.

            We can plausibly hold their salaries flat on a nominal basis.

          • mfm32 says:

            Yes, I completely agree. Although it’s generally against my political inclinations, I actually have quite a bit of admiration for the NHS and think it is a workable system. But not in America. Our political and cultural system are not equipped for success in that model, for many of the reasons you outline. For example, you are exactly right that nearly no one in the American healthcare debate is willing to admit that less care will be provided. Even Scott in this essay expresses concern that alternative models might reduce the ability of some patients to carry out years long searches for treatment of challenging illnesses. And of course cost control will do that. And of course we as a society do not need to accept that trade off. But if we aren’t willing to accept the trade off, let’s accept that we will need to pay for it.

          • Charlie Lima says:

            ES: Physician salaries have decreased on an inflation adjusted basis for decades. Depending on the source physician incomes peaked in the 70s or in 1993. Regardless of source, inflation adjusted income for physicians fell for pretty much all of the 1990s and 2000s for physicians.

            This again, part of why I am skeptical that we can save all that much. Maybe we could have an even slower rise in nominal wages, but we have been doing most of the easy stuff already to no noticeable effect.

            mfm32:
            I think things were really historically contingent. The US is wealthy enough to afford expensive healthcare so it does. And the US has better mortality rates for basically any disease. A lot of the international comparison data suggests a giant Simpson’s paradox to me: Americans are more obese, more diabetic, get less exercise, etc. so they are less healthy to start with. American medicine does a lot more interventions, which restore a lot of time lost to poor decisions, but the net is still Americans dying sooner than everyone else.

  6. chaosmage says:

    > once the government made easy loans and subsidies available to everyone, prices rose

    Prices tulipped.

  7. Bugmaster says:

    I think that the Amish are making a reasonable trade-off: less medicine in exchange for lower expenses. They don’t have cosmetic surgery, little no no women’s reproductive health (and no abortions or contraceptives, obviously), virtually no preventative medicine (do the Amish even vaccinate ?), less end-of-life care (since their lives are shorter), virtually mental care, etc. I’m not sure how they deal with things like cancer, diabetes, or congenital disorders, but I have a feeling they just go untreated most of the time. Those who survive are selected for toughness.

    Here in the “English” society, we’ve made a different trade-off. We try to keep as many people alive and comfortable as possible, which imposes massive ongoing costs onto our healthcare system. Insulin costs really add up during a person’s lifetime, and so do Ritalin costs. Old people require a lot more care than young healthy ones. An emphasis on swift and accurate diagnosis means that we need to maintain massive devices powered by superconductive electromagnetic coils… and so on.

    You could argue all day about which trade-off is better, but since it comes down to fundamental moral values, it’s a subjective argument that can never be resolved; all I can say is that comparing Amish to the “English” is like comparing apples to oranges.

    • Scott Alexander says:

      I would agree with this except that the Amish are healthier by most measures, so it doesn’t seem like they’re trading off health for money.

      You could still make an argument that they’re much healthier because of their lifestyle, and then trade off a little health for money, and still end up ahead, but this would suggest a pretty high money per unit health exchange rate.

      • Bugmaster says:

        Are the Amish actually significantly healthier ? We know they go to the doctor less often, but that’s not the same thing. I think their lifestyle obviously does help a great deal — all that hard work and lack of alcohol or smoking surely prevents a lot of diseases that the English are prone to.

        On the other hand, I suspect that the selection bias is also quite strong. For example, an English child (or adult) who develops some chronic life-threatening condition can enjoy a long life as long as he receives daily treatment; an Amish in the same situation would probably just die, thus boosting the overall health statistics for the survivors.

        Come to think of it, how do the Amish handle occupational injuries ? What happens to an Amish lumberjack who cuts off his hand, or a farmer who’s kicked in the chest by a horse, etc. ?

        • Oldio says:

          Amish who become disabled/widowed/what have you are taken care of by their communities.
          As for medical care, that seems like one of those things that gets sent over to the english healthcare system.

        • Nick says:

          On the other hand, I suspect that the selection bias is also quite strong. For example, an English child (or adult) who develops some chronic life-threatening condition can enjoy a long life as long as he receives daily treatment; an Amish in the same situation would probably just die, thus boosting the overall health statistics for the survivors.

          Huh? Wouldn’t this appear in overall mortality? Imagine if Amish were letting children born with diabetes die. That would certainly show up in the life expectancy, wouldn’t you say?

          • Bugmaster says:

            Yes, exactly. The overall mortality would be higher, but the average health of the survivors would be higher as well. It’s kind of like evaporative cooling, but for people.

          • JPNunez says:

            And their life expectancy is lower than for the “English”.

            So maybe we are onto something here. They probably have a bunch of sickness where people either die or leave the community that helps their overall health and drives down their life expectancy.

        • eric23 says:

          Most chronic disease don’t kill you right away. You live with diminished quality of life until 50 or 60 and then die.

        • Scott Alexander says:

          Yes, see the link about showing lower blood pressure and less heart attack, stroke, etc.

        • John Richards says:

          Health is more than just being alive. Health is the ability to swing an axe, the ability to saw a plank, to hold a child in each arm, to squat to milk a cow. When we consider health as the ability for the human body to perform various necessities, I imagine that the Amish are much healthier, men and women alike, than us sedentary dwellers in modernity.

      • Imsoindiethatmyblogdontfit says:

        Isn’t it basically consensus that people in most western nations are using gigantic amounts of resources to buy very marginal gains in health care? (Or, according to Robin Hanson, zero or even negative gain in health but lots of signaling.) So the pretty high money per unit health seems entirely reasonable to me. I could see why people stop making this deal once they have to pay for it using their own money.

        • Bugmaster says:

          Well, in this case, I personally disagree with the consensus. There’s a massive difference between a modern hospital, which is equipped with MRI scanners, blood screening labs, vital signs monitors, etc. etc.; and your average village herbalist/midwife/barber-surgeon. The gains are by no means marginal.

          • Imsoindiethatmyblogdontfit says:

            True. But the difference between a modern hospital and a 1970s hospital is not that great. Once you have vaccines, antibiotics and the rest of the low-hanging stuff, you basically made your gains. In Considerations on Cost Disease, Scott calculates (with enormous caveats) that the almost tenfold increase in health care expenditure since 1970 has granted us about an extra year of life expectancy.

            “The Amish chose to stay at a 1970s level of healthcare” seem like a pretty good explanation of what is going on. I think that’s the choice most people would make if they had to pay for their healthcare with their own money.

          • North49 says:

            @bugmaster
            The only two options are: exactly the system we have now at present costs, or herbalists and barber-surgeons? This seems disingenuous as a line of argumentation.

            (@nick point taken)

            There are good reasons to think medical technology would be cheaper over time rather than more expensive over time in the absence of our current insurance system. 1) almost all other kinds of technology falls in price dramatically over time. 2) medical technology not covered by most insurance falls in cost over time.

            MRIs being expensive isn’t a justification for the medical system overall being expensive; rather it’s another symptom of the same cost disease.

          • Nick says:

            My impression is that the gains from all those fancy tools at modern hospitals are not evenly distributed. Most people do not stand to gain from an MRI machine, but the few people whose conditions require investigation with one do. Of course, in a veil of ignorance sense, we often don’t know who those people will be, so in that way it is even.

            It seems to me an argument like that can cut both ways, though. We might reason our way into buying an MRI machine because we/our family may someday need it, but we might also reason it’s not worth it for us/our family. We know life expectancy on the whole isn’t gaining that much thanks to these tools, so its being a bad tradeoff is not unthinkable.

            @North49

            Really giving your false dichotomy muscles a workout today, huh? Just don’t forget leg day.

            Less of this, please.

          • Bugmaster says:

            @Imsoindiethatmyblogdontfit:
            I guess it depends on what level of technology Amish healthcare actually supports. Do they have X-ray machines ? Dialysis machines ? Invasive respirators ? I have trouble envisioning how such devices would work without electricity — though obviously the Amish could still use them when they visit modern hospitals; but then, why wouldn’t they use MRI machines, as well ?

            I think that’s the choice most people would make if they had to pay for their healthcare with their own money.

            Like I said, it’s a genuine trade-off. For example, one time I personally developed persistent nausea, vomiting, and a stomach pain that kept getting worse. When it got to the point where I couldn’t handle the pain anymore, I had a friend drive me to the emergency room (since I was too incoherent to drive myself by then). After wasting what seemed like an eternity on insurance verification, the hospital ran a blood screen, put me on IV fluids and some serious painkillers, then ran an X-ray and a CT scan (plus probably some other stuff that I was too zoned out to notice). Eventually, they diagnosed me with “probably an ulcer, we guess”, and I recovered a few days later.

            There are two ways to interpret the hospital’s actions. On the one hand, you could say that they’re deliberately throwing every test in the book at the patient, just to run up their bill — since they know that insurance will be picking it up. On the other hand, you could also say that they definitively ruled out cancer, infection, physical trauma, and drug interactions, before moving on to less life-threatening conditions.

            The probability of any given person having stomach cancer is pretty low, so running the cancer test for every patient with potentially cancer-like symptoms is a total waste — from the point of view of maximizing health care per dollar. On the other hand, if you want to save as many people from cancer as possible, screening everyone is the way to go. Like I said, it’s a trade-off.

          • Imsoindiethatmyblogdontfit says:

            @Bugmaster:
            I agree with everything you wrote. My original point was that the tradeoff that is made in the western world right now doesn’t look that attractive. Spending $10 a month on healthcare to get insulin, antibiotics and vaccines looks like an amazing deal. Spending an extra $100 a month to also get x-rays and antidepressants looks like a decent deal. Spending an extra $1000 on top of this to get MRI scans, double lung transplantats for 80-year olds and Who By Very Slow Decay looks bad. We are making a bad tradeoff. All the easy gains have been taken, and we are throwing enormous amounts of money on healthcare to squeeze out an measly extra year or month of lifespan.

            This is a classic example of diminishing marginal utility. I made a quick graph: https://sketch.io/render/sk-a3978b8159c9516172d09d65ab037bb8.jpeg

            I think the tradeoff that is being done today is bad. Since healthcare is mostly bought with other peoples money, we overconsume it. Add the effect of healthcare as signaling (if you believe in Robin Hanson), and the fact that you’re obviously a cruel heartless monster if you think we should let six-year old kids die if they need new lungs and $1 million of chemo, and you get the system that we have today.

            For example, screening everyone with stomach pain for cancer is probably a bad tradeoff. I haven’t done the math on that exact case, but it probably costs a lot for each life it saves. If we compare it to buying malaria nets it’s probably horrendously inefficient (but so is everything). If the hospital offered you to skip the tests and instead just taking the money all those tests would have costed, knowing that this gave you an extra 1% risk of dying of cancer or something, I think many people would take that deal.

            I think the US should slice healthcare costs by a factor ten. Have the state pay for basic inexpensive medical needs. Let people get private insurgence for more expensive care, tax this like all other consumption and don’t do the employer-pays bullshit. Then people would be able to spend on healthcare as they do an all other goods, and consumption would go down to a sane level. But this is politically impossible since it requires “death panels” and letting people die from treatable diseases etc.

          • Bugmaster says:

            @Imsoindiethatmyblogdontfit:
            I wish I could say that I agree or disagree with you, but logically speaking I can’t, because the conclusion depends entirely on one’s fundamental values — and I don’t think there’s a reasonable way to reconcile them. As you implied: is it acceptable to allow a few children to die from cancer, if doing so results in significantly higher medical expenses for everyone ? Someone might answer, “no, we should strive to minimize childhood cancer at all costs”, whereas someone else might answer “yes, because any other strategy is suboptimal in terms of overall utility”. Each person would say that the other one is obviously wrong.

            As I said elsewhere, in general modern Western society places a much higher value on human life than medieval societies, or even modern societies such as e.g. Russia. This isn’t a matter of government or finance; I’m speaking strictly of the moral values held by the common people.

          • Imsoindiethatmyblogdontfit says:

            @Bugmaster
            Anyone who says “no, we should strive to minimize childhood cancer at all costs” is being hyperbolic, unless they are donating all their income to child cancer research. Or they are trying to create some kind of weird sacred value. What they actually mean is “We should strive to minimize childhood cancer at a really high costs”. And now we can have a discussion about what those costs should be.

            The problem is that we are currently spending others people money on minimizing childhood cancer, and that the amount of money is decided by political processes. And “we should spend less money on childhood cancer” is a politically unfeasible position. So we end up spending way too much money on childhood cancer and everyone would be better off if they just got the money in the hand instead. (Expect the children with cancer of course. But behind the veil of ignorance, almost everyone would prefer to be born in a world where we took the childhod cancer money and gave it to people as cash.) (Here “childhood cancer” is a stand-in for “healthcare with bad marginal returns”. I’m sure some childhood cancer treatments have good returns, and that there are other parts of healthcare with worse returns.)

            Or, some people might really really value healthy human children. In which case they should focus on malaria, not cancer. But say that they really really value healthy American children. Cancer still seems like a bad investment too me. Maybe they really really value cute middleclass white American children? I don’t know man, in the end peoples values can end up so weird that they are entirely foreign. What’s really happening here is not well modeled by people rationally prioritizing between their values: people react emotionally to children with cancer and overinvest. And it is ok, even good, that people act irrationally from time to time. That’s what makes us human. But they should care with their own money and resources. When they are using state violence and percentages of GDP, it gets kind of scary.

            I agree that we assign a high value to human life in western nations, and I think this is correct to a degree. But I think the valuation in medicine has spiraled out of control and needs to be brought back to the level of the rest of society.

            Like, would you accept $500 each month for the rest of your life for a 1 year decrease in life expectancy? (The real tradeoff should be somewhere in that magnitude.) (For comparison, the difference in average male life expectancy between New York and Kentucky is 4 years.) Would you accept $500 each month for the rest of your life on the condition that if you get sick, you’ll get treated in a 1970s hospital? I’m already rich and both of those still looks like great deals to me.

          • Beck says:

            There’s a massive difference between a modern hospital, which is equipped with MRI scanners, blood screening labs, vital signs monitors, etc. etc.; and your average village herbalist/midwife/barber-surgeon. The gains are by no means marginal.

            The Amish I knew up in Geauga County, Ohio would go to the Cleveland Clinic, which is a modern hospital, if they were seriously ill.

            They just went a lot less than non-Amish (who they call Yankees btw rather than English), I think.

          • Loriot says:

            True. But the difference between a modern hospital and a 1970s hospital is not that great.

            Someone with breast cancer would probably care about having a 40% lower mortality rate.

            Anyway, cost disease means that you couldn’t get 1970’s staffing at 1970’s costs under any circumstances anyway.

      • Charlie Lima says:

        We should expect the Amish to be much healthier.

        Attending religious services once weekly is associated with a 33% reduction in all cause mortality. (https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2521827)

        Being married (which virtually all adult Amish are) has a likewise pretty significant impact. Being single increases your risk of all-cause mortality. The odds ratio comes out as 1.6 (https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2521827).

        Even having children is associated with longer parental lifespans. https://jech.bmj.com/content/71/5/424 suggests 2 years of additional life expectancy at age 60 from having children.

        Above and beyond the well known risks of tobacco and alcohol, I would be shocked if any population of married, religious folks with kids (and heck, happen to be white) were NOT much more healthy than US average.

        These things are all confounded and difficult to parse, but my prior that obscure religious group that has persisted for centuries has good health outcomes is pretty high to start with long before getting into anything specific about their healthcare.

    • eyeballfrog says:

      The Amish do vaccinate, though at a lower rate (68%) than the general population.

  8. Said Achmiz says:

    Why did health care prices start rising? I’ve wondered about this a lot before – see here, here, and here. Looking into this issue, I noticed glimpses of a different possibility. The increase started around the same time that health insurance began to spread. In one sense, this is unsurprising – of course health insurance would become a thing around the time care became unaffordable. But I’ve never seen someone really try to tease out causality here. Might the two trends have been mutually self-reinforcing? The price of care rises due to some original shock. Someone invents health insurance, which seems like a good idea. But this creates a series of perverse incentives, which other actors figure out how to exploit (eg the Lucemyra example above). Insurance-based-health-care becomes less efficient, but hospitals can’t or don’t internalize this to the insured patients – they just raise the price for everyone, insurance or no. That makes even more people need health insurance, and the cycle repeats as prices grow higher and higher and insurance becomes more and more necessary.

    This, precisely this, is the point of Christopher Anvil’s story “Positive Feedback” (which appears in the story collection Prescription for Chaos). It’s a quick read; I recommend it.

    The story was written in 1965.

    • Chris Phoenix says:

      I read “Positive Feedback.” It’s unrealistic.

      In real life, a group of people who were making lots of money on Blue Wheel would have formed an Industry Association, which would have lobbied (for safety reasons!) to make it illegal for anyone to practice car maintenance without a license.

  9. expensivedupilumab says:

    A relevant anecdote: I’ve had moderate eczema my whole life. It’s not debilitating, but it’s annoying and cosmetically a bit ugly (not horrifying, just a bit noticeable). I eagerly watched the development of a promising new eczema drug, dupilumab. It was finally approved in 2017, and my insurance would cover it.

    They priced it at $37,000. Eczema is pretty annoying, but I wouldn’t pay anywhere near $37,000 to solve it if it was my own money.

    So when it came out, I didn’t apply for it. It’s not efficient! It’s moral hazard and principal-agent-bad and I’d be doing capitalism wrong if I did.

    Everyone I talked to about it said I should do it. You pay lots of taxes, they said. You give 10% to EA causes, they said.

    So one day, one particularly bad eczema day, I applied, and now I take dupilumab, and I’m really enjoying the vastly improved eczema.

    Sorry your insurance costs are so high everyone.

    • Bugmaster says:

      As I said in my comment above, this is a trade-off that our society chose to make. In Amish country, you’d just live with your eczema for your entire life. Arguably, you might be conditioned to mentally resist it to the point of near irrelevance — or, at least, to avoid having any trace of discomfort show.

      • expensivedupilumab says:

        I see what you mean, but it’s unfortunate that the non-Amish trade-off isn’t even Pareto-optimal. My insurance company could have said, “cool, you won the eczema lottery, do you want dupilumab (which will cost us $37,000) or $10,000 cash?” I’d have taken the $10,000, saved them $27,000, and everyone would be better off.

        Maybe we can tweak our system to align incentives better, without going full Amish.

        • Sortale says:

          I wonder if this kind of insurance can be allowed.

          “We can give you this course of life-saving chemotherapy treatment which would be in the 100k or we can give you 50k and you can die or find a cheaper alternative elsewhere.”

          I can see how it would disadvantage the poor by incentivizing taking the payout than the needed medical care.
          also what would people do about controllable (e.g. STD) or semi-controllable (e.g. obesity) causes of disease

          • keaswaran says:

            STDs obviously have the complicating factor of being infectious. People don’t fully internalize the externality of failing to infect others, or ideally helping eliminate the disease.

            I suppose I’m thinking of the STD from the point of view of “what can I do to not infect others?” rather than from the point of view of “what can I do to prevent others from infecting me?” But given how everything’s going right now, I would hope that most people have realized that the former is more relevant.

          • sidereal says:

            Can you disadvantage someone by giving them more choice? Homo economicus says no.

            But humans seem to have a really strong aversion to explicitly trading off sacred values for non-sacred ones. Even though we do it all the time implicitly. I wonder if this is irrational or if there is some deeper reason, like, we’ve taboo’d that line of inquiry to prevent some sort of race to the bottom.

          • 10240 says:

            I can see how it would disadvantage the poor by incentivizing taking the payout than the needed medical care.

            If they gain more from the money than from the medical care, it doesn’t disadvantage them.

            On the other hand, it would incentivize insurance fraud.

        • ShemTealeaf says:

          Wouldn’t this create an enormous incentive to report false illnesses? Right now, I have no incentive to falsely report that I have a chronic illness, but if someone were to offer me 25% of the treatment cost in cash, that would be pretty tempting.

        • ana53294 says:

          The problem is how do they keep you from taking the 10,000$ and coming back for the 37,000$ treatment. Especially if it’s a life-saving treatment and not eczema.

      • North49 says:

        this is a trade-off that our society chose to make

        Funny, I don’t recall getting invited to that meeting. In a country where we have to “pass the bill to find out what’s in the bill” I don’t think one can reasonably even argue that democracy is some kind of collective decision making process any more.

        • Bugmaster says:

          Well yes, but that’s a separate issue. I wasn’t talking about our government, necessarily; but rather about people’s attitudes toward illness and death. To use a deliberately extreme example, in our current society the death of an infant is seen as a major tragedy. In e.g. medieval Russia, it was seen as a sad yet unavoidable event; basically the “cost of doing business” in life.

          • North49 says:

            It seems like you’re imagining a single-axis chart with cost on one side and level of care on the other, and as a society “we” have picked a point on the line, and whatever the cost/care trade-off, that’s just what we have to live with now.

            I think this is wrong on two counts:

            1) I don’t think there’s any meaningful way in which our society ‘chose’ the current healthcare structure; it’s a causally-dense kludge of kludges, the outcome in totality of which could not be predicted at the outset of any particular kludge.

            2) I don’t think the cost/care tradeoff matrix is a single axis. I believe we can improve both metrics simultaneously… or at least dramatically improve the cost metric without an appreciable reduction in care outcomes.

    • Del Cotter says:

      This is something various national universal health care systems do: they determine a treatment is more expensive than it’s worth, and say to the patient, sorry, you’re welcome to go pay a private clinic for this, but you can’t have it on the taxpayer’s dime.

      And then Fox News hears about it and says “see, America, this is what the socialists do, ration your health care!”

      The detail that Fox News misses out is that, at least in my country, if you like your American-style out-of-pocket expenses, you can keep them, or if you like your American-style private insurance, you can keep it.

      It’s even a little cheaper as insurance than US plans, because the private sector is partly a free-rider on national universal health care infrastructure when it wants to be (they’ll often pay just to send their customer into a national hospital, and everyone’s happy, including the customer), and partly the private sector is always under competition pressure from the national system; the customer might just cancel their policy if it gets too expensive, because there’s always that guaranteed safety net.

      • Akhorahil says:

        This is important, and likely the main reason why the U.S. is ridiculously inefficient in comparison.

        If you do things right, you can even have a private insurance system like in Switzerland, or lots of private medical actors servicing the universal health care system like in Sweden, but acting under a thick regulatory system.

        • nupi says:

          Switzerland is a fairly bizarre hybrid.

          ‘Basic’ insurance (which pays for almost all treatments but not really luxury like high end dining etc) is a mandate with identical coverage offered by private insurers who are mandated to be nonprofits in that area and have to take on everyone at their listed price but allowed to sell for profit supplemental insurance on top of the mandate.

          And to make things more complicated:
          – government pays 55% of hospital bill directly UNLESS you leave the same day
          – providers can be for profit or governmental
          – government subsidizes premiums where needed

          In sum, standard of care is generally high (except when they screw up the diagnosis for years like with me but at least when they finally got it right, expensive ongoing treatment was approved immediately) but so is the expense – my per capita premium is in line with the obamacare quote in the article, for one. And that is obviously before the hospital subsidy…

      • Edward Scizorhands says:

        Fox News is a problem, but they aren’t the only problem. What do you think the liberals do when an insurance company decides someone’s “life-saving treatment” isn’t worth the cost? We already know the answer.

        • Del Cotter says:

          I hear what you’re saying, and I understand why you would think that was symmetrical, but it isn’t. Symmetrical would be pro-private saying “Universal takes its clients’ money but stiffs them in the end anyway” and anti-private retorting “oh yeah? well Private takes its clients’ money but stiffs them in the end anyway, too!”

          But pro-private, to its credit, usually takes the point about universal being cheaper (nice steelmanning on their part, kudos) but explains this is a trade-off for full service, and alleges that universal rations and private does not. Anti-private’s rebuttal alleges that private milks its clients for much more money but at least it doesn’t ration OH WAIT YES IT DOES. (sorry, not my caps)

          So it’s not a symmetric “I’m rubber and you’re glue”, it’s an asymmetric “your parents bought the expensive insurance and all you got was this lousy health care”. It’s a bit like More’s (Paul Scofield) snark to John Hurt in A Man for All Seasons, that Hurt betrayed him for a political appointment governing Wales? (respect to Wales but it was a funny line)

          This matches my experience of non-health insurance. I try to be my own underwriter, but not to a foolish degree: I buy insurance when I could not afford the loss. But it has never bought me peace of mind. I lie awake worrying that my insurers will find a way not to pay me in the event of a big disaster, as they have not paid me for small ones.

          • Del Cotter says:

            Speaking of Mo(o)re, the horrifying thing about Michael Moore’s Sicko was that it wasn’t about the uninsured, but people who actually had bought the insurance.

          • Edward Scizorhands says:

            You need to stop trying to read people’s minds over the Internet.

          • Ouroborobot says:

            That feels pretty strawman ish to me, and I lean single payer. I think most people know that their private insurance has limits. A better attempt to explain how pro-private people feel would give credit to the difference in choice, at least in theory. Public payer models of Bernie variety take your money by force and say “this is the care you are getting, and if you want more, too bad”. With a free private market in health insurance, yes, your insurer will ration care, but you get to choose what plan you buy. Obviously in practice most people are constrained by what their employer offers, but at least in theory the system retains more degrees of freedom for the individual.

          • Del Cotter says:

            I fear that private insurers are inherently set up to maximise what they take, and minimise what they pay. It’s not that there’s a practical treatment limit–there really is, because we live in the universe–it’s that I fear the pay out isn’t driven by that limit, but by the urge to maximise revenue minus cost.

            To the extent that’s inherent, they turn themselves to straw with no rhetorical effort on my part. If the opponent I’m debating actually is the Scarecrow, it’s not my fault I kicked the stuffing out of him; he came to me made of straw, I didn’t contrive that.

            That’s why I had a good word to say about private insurers in countries with a state sector: they have the classic check of risking their customers abandoning them for the competition. I don’t feel there’s enough of that in the US.

          • disluckyperson says:

            I think Edward’s point is that there’s a bit of a double standard among proponents of national health care. If the nationalized system says, “this treatment is not worth it”, their decision is assumed to be based on good motives, solid research, evidence based medicine, etc. If the for-profit system says “this treatment is not worth it”, their decision is assumed to based on greedy, conniving, nefarious motives. Maybe this double standard is justified, and we should apply more scrutiny to private sector motives than to government agency motives. But this scrutiny undoubtedly makes it more difficult for insurance companies to contain costs. Just look at the backlash to HMO cost containment practices in the 90s.

          • Edward Scizorhands says:

            I suspect that, were the US to try to establish something like NHS, while the intention would be to stick to the guidelines on what is and isn’t “worth it,” we would rapidly blow right past the guidelines, leaving us worse than we started.

            If we could accept such decisions, something like NHS would work. But if we could accept such decisions, our existing healthcare system wouldn’t be such a mess as it is right now.

          • Del Cotter says:

            I mentioned below advising not obsessing about the UK NHS. There are a hundred universal health care systems in a hundred countries, and most of them aren’t like the NHS.

          • Del Cotter says:

            @disluckyperson

            Good, bad… it’s not unfair to point out that for-profit companies have a particular motive: profit. This leads to motivated decision-making, and is a classic principal-agent conflict, where the principal is the policy holder, and the agent is the insurer. Or a principal is the policy holder, the agent is the company officers, and another principal is the company owners, a.k.a. shareholders, making it a principal-principal conflict if you like. You might not like that it’s described as a bad motivation, but it’s not hypocritical to point it out.

            To make it symmetric (that is, a single standard, you’re saying its a double standard) you need to identify a similarly problematic motivation to single-payer (or other) cost-benefit committees. Good luck ascribing motivated decision-making to NICE. And by good luck, I don’t mean try, I mean please don’t try, I find such attempts strained.

          • Del Cotter says:

            It reminds me of the straining to find a symmetry between oil companies, and climate scientists with their trillions in sweet research funding.

          • disluckyperson says:

            @Del Cotter

            I think this question of motives is at the heart a much broader ideological debate. If you would ask a libertarian, they would say OF COURSE there are all manner of nefarious or non-optimal motivations that politicians or bureaucrats could have. They don’t ascribe any special trust to people simply because those people don’t have a market profit incentive. The opposite in fact. In the case of the NHS, I could imagine that politicians are not eager to raise taxes as that affects their chances of re-election. Or in the case of drug purchasing, maybe there’s a strong antipathy towards “Big Pharma”. I think your analogy to climate science is good. There too, I wouldn’t say the motivation is “sweet funding”, but there has always been a very strong anti-capitalist trend in the environmental movement, even before climate change became the big thing it is now. And of course, it’s not surprising to see countless articles about how capitalism must be abolished to save the climate.

            But I think you’re right, to many people, profit opportunity is the ultimate reason for suspicion. And I suspect more people, even in the US, share your sensibilities than mine, hence the backlash to HMO cost containment efforts.

      • Charlie Lima says:

        That is how socialized systems say they work. Reality is, as always, messier. For instance HV vaccination was approved in 2007, but the NHS spent a full year dickering on price even though the asking price met the NICE’s metrics. So girls in the UK get an extra year of cervical cancer risk because the money had to come from somewhere. And lest we think this is unusual the NHS only started covering male HPV vaccination, in spite of it being a leading cause of anal cancer, in 2019. Again, long after the vaccine passed the NICE’s own metrics.

        And let’s be honest. Suppose a new drug comes out, is the NHS going to call up patients and say “Sorry your lung treatments are less cost effective than these new cancer pills, best of luck.” Yet this is exactly what happens if Parliament does not increase the budget when something new crops up.

        In any event we have the exact conditions you describe here in the US. The Indian Health Service is government run with a patient population larger than several European countries. Yet its costs are much more similar to US private hospitals than European ones.

        • Del Cotter says:

          Oh, I don’t doubt it can happen that treatments are not immediately rolled out even after they are judged by independent bodies not to be cost-ineffective. As David says, utopia is not an option. A system where every treatment that is stupid expensive is withheld and every treatment that is not is offered, is a Platonic impossibility, like a perfect circle with zero ellipticity.

          I wanted only to say that offering every $37k eczema treatment free at the point of use isn’t a systematically inevitable evolution, or an ineluctable cost disease. And, as I say, in most countries, the private alternative isn’t prohibited, so off you go, pay for it yourself if you feel different.

      • Scott Alexander says:

        “The detail that Fox News misses out is that, at least in my country, if you like your American-style out-of-pocket expenses, you can keep them, or if you like your American-style private insurance, you can keep it.”

        Is your country Ireland?

        • thisheavenlyconjugation says:

          I think it could be most developed countries with single payer systems except for Canada. AFAIK the idea of banning private healthcare is a weird North American (i.e. Canadian and various Democratic presidential candidates) thing.

          • Del Cotter says:

            Correct. When I heard about the Canadian ban-private-healthcare thing, I was surprised and shocked, but at the same time intrigued. I’ve often suspected that socialist experiments in democracies are fatally undercut by leaving a private sector in existence, as it’s like leaving a disease reservoir to break out again later.

            I didn’t know Canada was such an outlier though. I figured maybe a third to two thirds of countries might have gone that way. (as many of you know, there are a hundred healthcare systems in a hundred countries: if y’all don’t like the sound of the UK NHS, some of the others might be more attractive, so try not to imprint on the British alone)

          • Del Cotter says:

            As an example, we are beginning to suspect leaving any private schools (like Eton, school of the last two male Prime Ministers) in existence was a mistake, since guess where the political class including Diane Abbott send their kids?

            It was amusing/disgusting to realise when the drama The Night Manager came out, that all three male leads, Tom Hiddleston, Tom Hollander, and Hugh Laurie, went to Dragon School, which despite the Harry-Potter name is just a small prep school in Oxfordshire (also where Stephen Wolfram went to school, to tie in another recent thread).

            We begin to wonder if we just built ourselves a caste society instead of one with more equal opportunity. Which sounds a bit like the problem Americans have with their hated “public” school system [the US meaning of public school]

    • eric23 says:

      If it makes you feel better, I assume the insurance company didn’t actually pay $37000 but rather a discounted rate?

      Also, more people using the drug means the drug research investment is paid back sooner which encourage more drug development in the future.

      • expensivedupilumab says:

        Yeah, that was partly my thinking too – that dupilumab doesn’t actually cost $37,000 in materials to make and now those materials can’t be used for something else. Rather, it was a transfer of money that may help drug research. And as you say, it probably wasn’t the full price.

        • 10240 says:

          If drug research costs more than the value of the benefit provided by the drug, that’s just as bad as if manufacturing the drug costs more than the benefit.

          • eric23 says:

            The grandparent post had “moderate” eczema – for people with severe eczema I imagine the benefit would outweigh the cost.

    • John Richards says:

      Jokes on you, I don’t make enough money to have to pay for my insurance!

  10. Egregious says:

    Amish life expectancy (at birth?) around 70, and 5-10 kids per family. Seemed to me the amish should be about doubling every generation (accounting for people leaving). Turns out I was correct: the armish are growing exponentially!

    Relevant quote, from 2012:

    But according to a new census, the Amish are growing faster than ever. There are nearly 251,000 Amish people in America and Canada, according to Ohio State University researchers. That’s more than double the estimated population in 1989 of about 100,000. Researchers estimate the population will double again to half a million within about 21 years.

    • eric23 says:

      Every population with a constant fertility rate is growing (or shrinking) exponentially 🙂

      • Bugmaster says:

        Well yes, but I still find Egregious’s link somewhat surprising. Where are all those Amish living, physically ? They may be growing exponentially, but the land isn’t — and it’s not like they can build skyscrapers…

        • acymetric says:

          The Midwest (Ohio, Michigan, etc.) is very flat and undeveloped. Plenty of space there.

        • John Schilling says:

          A lot of the English are giving up on farming and moving to the city for the “good jobs”; that frees up land for Amish farmers. And the Amish are branching out into less land-intensive pursuits like woodworking and machining.

          The Mennonites, similar to the Amish but more numerous and a bit more cosmopolitan, have taken to looking for cheap land outside their traditional domains; there’s a fairly substantial community in Belize, for example. I expect the Amish will eventually follow have to follow suit.

  11. Murphy says:

    >Diabetes rates are 2% vs. 8%

    OK, diabetes is an unusual case. Because of the fact that developing diabetes was basically a death sentence until the discovery/invention of insulin it makes for an interesting case study.

    Until 1922, childhood diabetes meant death.

    As such genetic alleles that caused childhood diabetes were removed from the population by death until 1922.

    So the alleles for childhood diabetes didn’t tend to survive long in families so the background rate was roughly equivalent to the rate at which new mutations sprung up.

    In one of medicine’s more dramatic moments Dr. Frederick Grant Banting, Charles Best, Dr. James Macleod and Collip. went from bed to bed, injecting an entire ward with the new purified extract. Before they had reached the last dying child, the first few were awakening from their coma, to the joyous exclamations of their families.

    Insulin changed that beause it allowed people to survive and have families and live normal lives.

    So the background rate slowly increased.

    Childhood type 1 diabetes tends to be genetic rather than lifestyle because kids tend to be more active and it typically takes time for unhealthy kids to fuck up their pancreas.

    I can’t find figures that seperate childhood diabetes for the amish but it might be used as a way to judge effiacy of their healthcare system since 1922. If they successfully save the lives of diabetic children in their communities then we would expect them to have a similar rate of childhood diabetes to other populations. If a lot of those kids quietly die because no doctor noticed and said “get this kid on insulin” then we would expect a much lower rate of childhood diabetes because of selective pressure on the alleles involved.

    Note that this doesn’t apply very much to diabetes that sets in in later life.

    • George3d6 says:

      I’m not sure why you are assuming autoimmune diabetes is hereditary… it’s, autoimmune.

      Yes, genetics plays a role, but so does the environment of the fetus/mother and the environment after birth, as well as just random chance.

      It’s hereditary as in the best marker we can find for predicting it is heredity, but heredity can’t make accurate predictions about it, only increase the probability. It was also never really cleared out of the gene pool in any way because there is no “diabetes genes”, there’s certain genes that under certain conditions increase the risk of autoimmune diabetes.

      Also, looking at metabolic diabetes alone is not the whole picture, it’s worth looking at metabolic syndrome, or at least at the probability of NAFLD plus diabetes since they are usually exclusionary but stem from the same set of issue.

      • mike150160 says:

        Indeed, I believe the evidence from the twins studies was that Type 1 had almost no genetic component whereas type 2 had a high heritable component.

        But somehow people just assume that 1 is genetic and 2 is not.

  12. LosLorenzo says:

    With regard to the rising price > increased insurance demand > rising price dynamic, it seems at least plausible. On the face of it, you’d expect this dynamic to apply to other types of insurance as well, wouldn’t you? So to support this view you’d have to either show that this does happen with other classes of insurance (the extent to which it affects other sectors should then frame expectations of the effect size on health spending), or explain why health insurance is different.

    • Del Cotter says:

      Health insurance is different: it’s your life. Car insurance, you can pay out of pocket and if it’s too much you can get a more insurable car. House insurance, they can’t offer to rebuild your house better, or offer you expensive fire-prevention services (at least they don’t yet), contents insurance they can only offer to replace your iphone, not give you a gold iphone. If they offered you the gold iphone you’d buy the policy that didn’t–you don’t need the gold iphone.

      But you can’t buy a better life with the money you saved not saving your life.

      • Del Cotter says:

        Note the Christopher Anvil story ended with car owners telling the insurers and their car diagnostics machines to take a hike. Taking inspiration from medical insurance turned out to be a mistake for the equipment manufacturers.

        • John Schilling says:

          Anvil failed to anticipate the world where an automobile’s finely-optimized engine could not operate if the embedded computer with its obfuscated code didn’t get the right software patches from an Officially Approved Source, and where anyone who tried to build anything else would find that the less-optimized and -computerized engines would be barred from selling it due to fuel economy and emissions regulations.

      • LosLorenzo says:

        I’ll certainly grant that health insurance is not identical to eg. home or car insurance. But just as a person without car insurance could buy a cheaper car to reduce the risk of financial ruin in case of an accident, a person without health insurance can decide that they will not buy $500,000 of cancer drugs to extend life expectancy by 6 months in case of stage 4 lung cancer. So that part of the inflationary dynamic could still be in place. To be clear, I’m not sold on this, but we still haven’t conclusively shown that the mechanism could not contribute to spiraling home costs. I’m still not seeing ‘that one thing’ that would make me not expect to see this inflationary effect on other big-ticket, commonly insured ‘things’.

        But thinking about it some more, I suspect the original claim is flawed. Scott wondered if rising healthcare costs lead to increased demand for insurance, which leads to increased price of healthcare. The problem is that this equivocates between price of healthcare and price of insurance.

        As the cost of like-for-like medical procedures/drugs rise, so will the demand for insurance. And as demand rises, so will the price of insurance. So far, so good. But the increased demand for insurance does not directly increase the underlying cost of procedures/drugs, only the cost of the insurance. You could still claim that increased insurer profits would be to some degree captured by healthcare providers, which they no doubt would be in equilibrium. But insofar as providers don’t capture the whole increase in profit, this cycle diminishes rapidly and likely can’t explain more than a small part of the steadily and rapidly rising healthcare costs.

        • LosLorenzo says:

          Also it’s worth pointing out that as the price of like-for-like medical care rises, demand for it will fall. Econ 101. So it is true that as healthcare prices rise, more procedures and drugs would necessitate insurance for Joe Average, should he wants the procedure. But the increased price should also make Joe demand the procedure less (ie. but a plan that covers less). All else equal, this alone should convince us there’s no insurance-driven inflationary price pump. But of course, all else is not equal…

          • Garrett says:

            How practical is it to set up a health insurance plan which explicitly excludes a lot of high-dollar treatments of marginal value?

          • LosLorenzo says:

            @garrett
            As far as the law permits, this exists today, no? Seems trivial, to the extent that any change for a giant insurance bureaucracy/industrial complex can be trivial. There are some strong arguments for setting lower bounds on coverage. At what level that should be is trickier.

  13. Akhorahil says:

    It just struck me that the Amish is one of the very few examples we have of what Neal Stephenson calls a “phyle” (like the Victorians in The Diamond Age).

    And the trust thing is important – Hasidic Jews dominated the diamond trade in Antwerp for a long time, because it was all but impossible to undercut them when they could assume trust between each other and save a lot of money that way, where others would have to put in measures.

  14. Gumpalonia says:

    I’m fascinated by how many of today’s biggest economic problems just mysteriously failed to exist in the past. Our grandparents easily paid for college with summer jobs, raised three or four kids on a single income, and bought houses in their 20s or 30s and never worried about rent or eviction again.

    Would be really interesting to read what you think of the less quantifiable explanations like a reduction in general trust. Intuitively it makes sense that if people trust each other less, every interaction requires more checks and balances than before, and thus also becomes more costly. Amish manage this partly because of high ingroup trust, which is apparantly seen by “outsiders” as a sign of trust between groups as well. But explaining anything as a cause or a result of trust is difficult and the whole model looks pretty politicized nowdays.

    • Simon_Jester says:

      The interesting question is, what prices were we paying for that increased level of social trust?

  15. Meneldur says:

    I generally lurk, but I though mentioning something about life insurance might be pertinent here…

    It’s not my actual job, but I somehow became a liaison between several educational institutions and insurance companies, and the thing that hit the hardest was how horrible life insurance is compared to the system that’s in place for those educational institutions. They’re ultra-orthodox Jewish institutions, mostly, so a liaison who uses the internet is pretty useful, but I digress….
    In many ultra-orthodox Jewish communities, especially Hasidic ones that are a large group, the religious leader has someone organize health insurance. Much like the Amish, we’re talking large families – 8 kids is the average, I’d say. The way it works is actually pretty simple. Every family in the sect signs on to the insurance scheme. The way it works is, you usually pay nothing. If a parent of one family passes away, every family pays a set sum which is then transferred to that family. For example, in the Hasidic community X there are 30,000 families. Family A has lost their father. Every family transfers 100 dollars to a bank account, which is then transferred to Family A. Boom, 3 Million in life insurance (which is critical when you’re a single parent with 8 kids to raise marry off).
    The beautiful part about this is that no one needs to worry about the constant costs. Payments are only made out to families who still have unmarried kids – if the father of family B passed away at 65 but all their kids are married, you don’t need to pay, because the assumption is that the mother can get by on her own savings and her kids’ help. The payments are thus infrequent – only when a parent dies young enough so that their surviving family needs the money. Maybe in the current epidemic this might seem more problematic, but again, if most of the dead are elderly, it’s not an issue – you’d still only be paying out infrequently.

  16. AlphaGamma says:

    I’m slightly reminded of the Dutch system for people with a religious objection to the state-mandated health insurance (who tend to be orthodox Calvinists- from what I can find, Dutch Muslims tend to buy insurance even though Islamic law forbids it). As I said in a recent OT comment, most Dutch residents are required to buy health insurance, which is heavily regulated and subsidised. Premiums are around €100 per month, and people with pre-existing conditions don’t pay more.

    If you state that insurance is against your religion, you are required to put an amount comparable to what your insurance premium would cost in a healthcare savings account each month. You can’t take money out of this account for anything except paying for healthcare. If it runs out, you are expected, like the Amish, to rely on the support of your community, though I think there may be some provision for emergency care. If your views change and you decide to buy insurance, anything left in the healthcare account is confiscated (and goes in to the government healthcare budget) to avoid possible free-rider problems. On the other hand, anything left in the account when you die goes to your heirs.

    • MVDZ says:

      Interesting, I didn’t know that. Many other countries may allow weird minorities and sects like the Netherlands, but I can think of few that actively try to accommodate them as much. (I’m also Dutch, by the way.)

      • AlphaGamma says:

        (I’m not actually Dutch, I just live here…)

        And in terms of accommodating minority religions, lots of places seem to have some form of accommodation for various minority religions where the law would, if applied equally, require them to do something that is against their religion. Possibly the prototypical example is the countries that allow alternatives or exceptions to compulsory military service for members of pacifist sects. In the UK, where I grew up, turbaned Sikhs are exempt from motorcycle helmet laws- while the British system to accommodate members of religions opposed to taking oaths by allowing them to affirm instead has expanded to almost become the default in an increasingly atheist population. The US has a whole range of things that only members of certain (often Native) religions can do, such as using peyote or eagle feathers in ceremonies.

    • Gnecht says:

      I would love to read up on the Dutch system for state-mandated health insurance exemptions. What are good search keywords for this? Names of government agencies maybe?

  17. oldman says:

    I don’t see how administrative cost could explain it. Surely administrative cost is in the prices hospitals charge, and Scott said in the collective bargaining costs that the prices charged by hospitals to the Amish are similar to what insurance companies get.

  18. meh says:

    fascinated by how many of today’s biggest economic problems just mysteriously failed to exist in the past

    There is a tendency for people to look at a lifestyle that only existed for maybe a generation and a half, and assume it has been some ancient constant way of life for thousands of years. Things are always different from what they were.

  19. Telemachus says:

    Can I ask whether the differences between healthcare outcomes/rates of disease have been controlled for the genetic homogeneity (relatively) of the Amish population compared to the comparison (e.g. the US population as a whole)? Any Founder Effect due to this may account for some discrepancies in health on a population level, and perhaps make it more difficult to draw any firm conclusions about the possible benefits of the Amish healthcare system vs. that used by the general populace.

    Don’t have any information to suggest either way, but just a point to think about, and I’m curious to see if there are any thoughts/answers.

  20. arnold kling says:

    In my book Crisis of Abundance, I asked why health care was so much more expensive in 2006 (when I wrote the book) than in 1970. The answer was what I called “premium medicine,” with lots of capital equipment and medical specialists. I suggested that premium medicine makes a big difference in health care spending but only a small difference in health care outcomes. Perhaps the Amish are not as eager to consume premium medicine. Maybe they don’t get routine colonoscopies starting at age 65 the way we are advised to do. Maybe they don’t go for an MRI every time they get a backache. etc.

    • joshk says:

      I think that’s a big part of it. I’ve seen people spend so much time and money with specialists and very marginal improvements. How many people run to a neurologist to get Aderall?

  21. wiles3142 says:

    I believe the late Prof. Uwe Reinhardt argued along these lines: There are approximately 3000 hospitals in the USA. They are geographically distributed, except in large conurbations, and each has an effective monopoly in its zone. They can charge pretty much whatever they want as consumers (a) have little choice, (b) think someone else is paying, (c) are often in urgent need of care, and (d) are offered no way to compare the costs of different providers (this is true across health care). Further, hospitals compute the price per item (the hospital master) not by looking bottom up at costs but top down starting with the cost of running the service plus the desired profit (sorry, surplus). They then distribute this sum across billable items in ways which are unique to each hospital, making comparisons across hospitals impossible.
    Finally, and I don’t know if this is still the case, Reinhardt pointed out that insurance companies are regulated at the State level but hospital chains can operate across States. If health insurance companies could negotiate nationally they might be able to enforce (a) standard pricing methods and (b) effective price negotiation. He explained that insurance regulators oppose Federal regulation because they want to keep their jobs. So we all suffer to keep an handful of bureaucrats employed. Of course, Medicare does negotiate nationally and surprise surprise obtains lower prices.

    • keaswaran says:

      How many hospitals have effective monopolies? And how many people live in the areas served by these hospitals? My understanding is that in recent years rural hospitals have been going bankrupt at a pretty high rate, which suggests that they’re at least not effectively extracting monopoly prices (unless it’s just continued depopulation of rural areas that is doing them in).

    • BBA says:

      Insurance is still regulated at the state level but there are multi-state health insurance “chains” now too. If United Health isn’t negotiating with HCA on a nationwide basis, what was the point of all those mergers then?

  22. Del Cotter says:

    I’m fascinated by how many of today’s biggest economic problems just mysteriously failed to exist in the past. Our grandparents easily paid for college with summer jobs, raised three or four kids on a single income, and bought houses in their 20s or 30s and never worried about rent or eviction again

    Maybe this was only the happy experience of one or two generations, rather than all the past. I feel many of my 19th century working class ancestors had no medical care at all, college was a fantasy for members of their class, they raised four children in poverty, and were beholden to a landlord all their lives. I credit socialism for the post-war period when anything seemed possible, what the French call Les Trente Gloriouses, and Thatcher and Reagan for destroying that.

    edit: ninja’d by meh

    • Alkatyn says:

      Yeah, and the semi mythical period of American middle class affluence was only really possible because of the aftermath of the second world war meaning America had minimal competition in international trade, making the wages for workers artificially high.Along with government programs like the GI bill. And even then it didn’t really apply to minorities

  23. Frederic Mari says:

    This is the extract I think is most interesting as it’s the reason for the whole blog post.

    “I’m fascinated by how many of today’s biggest economic problems just mysteriously failed to exist in the past. Our grandparents easily paid for college with summer jobs, raised three or four kids on a single income, and bought houses in their 20s or 30s and never worried about rent or eviction again. And yes, they got medical care without health insurance, and avoided the kind of medical bankruptcies we see too frequently today. How did this work so well? Are there ways to make it work today? The Amish are an extreme example of people who try to make traditional systems work in the modern world, which makes them a natural laboratory for this kind of question”.

    First, I think you don’t fully account for how people just used to die in the past. So your grandmother nurse telling you all was fine also reflects the fact that, in lots of cases, people did die often. Obviously, no longer costing anything. Ditto with living with infirmities and chronic conditions. If the system couldn’t treat you and it wasn’t killing you, you just lived with it. You mention that but I’m pretty sure the savings are quite amazing compared to our system’s ability to keep a heck of a lot people alive and minimize chronic conditions but at great cost. After all, third world countries health care isn’t crumbling under excessive costs either… but I wouldn’t recommend it.

    Two, when people bring up the Aemish or the Mormons as examples of successful communities (something conservatives love to do), I’ve always wondered – How do you account for self selection? i.e. if you’re chronically sick and your 30 neighbours are getting seriously fed up with putting up with your deadweight, do you leave the community and join the English? Well, that would make the Aemish stats look that much better.

    Ditto the teenagers who love their snow bikes and decide “screw this sect of 17th century nutters, I’ll join the English”. You got one risk taker out of your community. Good for your stats.

    Same thing with Mormons and the unemployed. You get your elder to come and help a few times. Then, he gets fed up and you get the message it’s time to move on.

    Etc.

    Most problems we have didn’t exist in the past b/c

    1- our technology didn’t allow us to attempt to address those. I got the feeling (no hard data) that we’ve progressed our ability to attempt solving hard problems at the margin faster than we’ve been able to reduce the costs of those technologies allowing us to make those attempts.

    2- expectations have evolved possibly faster than our ability to use tech to satisfy them. Broader point but similar to 1.

    3- male average salaries haven’t kept up. Both women joining the workforce, legal and illegal immigration and international trade have depressed salaries for a lot of men. We moved to a winner take all economy. It’s great for women, immigrants, innovators and we slashed poverty across the globe. As a result, median earnings are not longer enough to afford almost best in class lifestyle.

    NB: We have to circumvent the exact historical time you’re talking about. I don’t think being a factory worker in the 1910s was any kind of healthy living…

    • keaswaran says:

      > I got the feeling (no hard data) that we’ve progressed our ability to attempt solving hard problems at the margin faster than we’ve been able to reduce the costs of those technologies allowing us to make those attempts.

      See the use of mass isolation to prevent pandemic disease. Is it worth it? I suspect it is. But we’re paying a lot more in temporary economic activity for the sake of killing the virus, while paying a lot less in the permanent loss of friends and family to the disease.

      • Frederic Mari says:

        That’s a good example, yes.

        The Spanish flu killed 2-3M people, people cried, some never quite got over the loss of a sibling or young husband/wife but the economy bounced back quite quickly apparently.

        The UK was almost willing to try that and then, seeing the risk of dead people piling up, walked back and went into quarantine like almost everybody else.

    • Matthias says:

      You say it’s a winner take all economy, but also that previously poorer people are doing better. What gives?

      • Frederic Mari says:

        Sorry, I should have been clearer. In the West, with manufacturing either hollowed out or automated, it’s a winner take all economy.

        In the RoW, notably China but not only, where we have sent a lot of the manufacturing, the progress from subsistence agriculture and the slow weeding out of dreadfully bad economic policies (i.e. communism) has led to a slashing of global poverty.

        In short, the blue collar white western male paid the price of reducing global poverty.

        That’s leaving a lot of nuances out but it’s not wrong.

    • sharper13 says:

      Based on your post, I get the impression you don’t have much experience with the way groups like the Mormons actually work in practice. It’s way more organized, with published standards and rules and policies and procedures, than you appear to imagine.

      Let’s take your example in practice:

      Same thing with Mormons and the unemployed. You get your elder to come and help a few times. Then, he gets fed up and you get the message it’s time to move on.

      The reality is more like, you lose your job, you go see your local Bishop for help (who already knows who you are and your general situation). He:
      1. Asks you about your finances and what you’re already working on with your extended family.
      2. Gives you what you need to pay your rent/mortgage (or whatever is needed to keep your family off the street).
      3. Arranges for you to go “shopping” on a regular basis at the Bishop’s Storehouse (which is similar to a food bank, except closer to a grocery store stocked with essentials which mostly has stuff the larger Church has from various farms, orchards, and similar enterprises).
      4. Sets you up with an employment specialist (assuming you’re physically able, sick/injured would be different) who works with you on both short-term employment and on improving skills/resume/money management/etc… with the goal of getting you self-sufficient again as soon as possible. Depending on the area, they may already know other church members with low-skill jobs available for you as well.
      5. While you’re trying to get self-sufficient, finds something (cleaning up the church parking lot, working at a thrift store, at one of those food producing places mentioned above, etc…) you’re able to do in order to feel like you’re at least partially “earning” what you’re getting by not contributing perhaps economically enough to it, but at least doing what you realistically can.

      Nobody “gets fed up with you” unless you’re just unwilling to do anything for yourself that you’re clearly able to do. Even then, it likely takes years to get to that point. Up until then, it’s going to be “We can keep covering your rent, but you need to be working with the employment specialist.”

      The level of personalization based on people’s needs and abilities is the opposite of government one-size-fits-all-but-nobody-cares bureaucracy in terms of helping.

      • Frederic Mari says:

        Thanks for your informative reply, sharper13 and I hope you’ll see my comments despite the belated reply.

        I had/I still assume that the local bishop has a fair amount of leeway. In a way, your own description of the extreme personalisation of the service makes me doubt the system could work well if there are a lot of stringent, mandatory “published standards and rules and policies and procedures”. If these were truly constraining, then it’d be exactly the secular bureaucracy – with, presumably, the same results.

        Nobody “gets fed up with you” unless you’re just unwilling to do anything for yourself that you’re clearly able to do.

        That’s kind of what I mean. There are people like that. Antisocial assholes, mentally unstable or unwell people, whatever. The rest of society has no choice but to cope with them one way or another. The Mormons have an out. They can show these people the door.

        Finally, the services you describe are provided more or less by secular society. I’ve been unemployed (in France, not the USA) and the person at the unemployment bureau who took my case for the job hunt part of the procedure (i.e. the part with high social contact/touchy feely bits) was rather nice. She was sorry for my problems – or at least was able to express sympathy for them and while she wasn’t a close acquaintance like a local priest or pastor might be, she did a pretty good job of conveying enthusiasm for my job search. So it seems to me that one-size-fits-all-but-nobody-cares bureaucracy
        is a bit glib. Unemployment agencies have workshops and contacts and programs etc. mimicking all the help you described. But maybe you’re right it’s too impersonal still for people who have problems helping themselves and who would benefit from a more constant/closer mentoring/monitoring.

        I’m interested in knowing the standard size of a flock for a bishop in Mormon communities. And how do they deal with concentrated bad luck. Like, a local business closes down. It triggers a very localised 20% unemployment spike. How do they deal with that?

        Also – could we have a secular version of a bishop? A mentor/elder/wise man for a limited sized flock. The equivalent of an elderman for a medieval village. The issues with that are numerous – first, it’s a terrible invasion of privacy. second, it doesn’t work if people want to remain mobile.

        I believe the Chinese had or have a similar system but, obviously, their aims were to increase and deepen their dictatorship as well as helping people. I’d be personally uninterested in the trade off but some might.

        So – yeah – I’m still left with the feeling that the Mormon system works well for Mormons due to self selection. Only specific people become or stay Mormons. And within that population, sure, the system works.

        But it’s not obvious we can reproduce it for the wider society, not because we don’t have the resources or enough caring people but because the population is meaningfully different and would react differently to the trade off between privacy and mentoring/monitoring.

        What’s your take?

  24. Sandpaper26 says:

    I don’t work in the medical field, but I think you got the point about the cost of lawyers, clerks, and other administrative staff dead wrong. Three points:
    1. The existence of the hospital’s staff of lawyers et al necessitates a symmetry of lawyers et al hired by the insurance companies who try to do the swindling. Whether they succeed or fail, their cost goes to the consumer. When they succeed, it saves the insurance company money — maybe (see point 2). The consumer has already paid their deductible, so it only helps the consumer in the sense that the premium costs might be slightly lower (though we have no reason to assume this savings is passed on). But the administrative costs are constant, borne by the consumer, and only exist because the insurance companies try to pay less.
    2. Hospitals and clinics have costs of their own that require recoupment from some entity. When insurance companies refuse to pay for one thing, the hospital still has to finance it somehow. So why wouldn’t they raise the price of the things insurance companies will pay for? Why wouldn’t the price be statically and artificially inflated across all services to account for when the insurance company won’t pay? Won’t this cost everyone more in the long run?
    3. The claim that an insurance company could become “honorable” enough to justify not being charged administrative costs is laughable. The hospital is keeping this staff on payroll. The only reason they can waive the cost for Amish people is because it’s such a small percentage of total patients. If an insurer representing ~20% of patients decided it would always pay everything, do you think the hospital is going to lay off 20% of its legal team? Or is it going to keep charging the admin fees? At what percentage of patients might a hospital decide to do one over the other?

    • spacecommie says:

      “If an insurer representing ~20% of patients decided it would always pay everything, do you think the hospital is going to lay off 20% of its legal team?”

      Trivially, if they have a hundred people working 10,000 denials a week and they go down to 8,000 denials, you should be able to lay off twenty people or devote more resources to following up with other insurers, which also improves the bottom line. So yeah, maybe. If anything it seems more mysterious that Amish get administrative costs waived now, when they have an extremely small impact on overhead.

      The other thing I think this misses is that insurers are not “charged” in the normal sense: they get a claim for $X and they pay $X-Y of it. Y is usually set by contract between the hospital and the insurer, or declared by fiat if the insurer is the 800 pound gorilla in the room (e.g. Medicare, sometimes the bigger private companies). The relationship between the insurer and the hospital matters.

      • Sandpaper26 says:

        The fact that Amish get administrative costs waived isn’t at all mysterious to me — they’ve demonstrated that they’ll go as far as Mexico for care, so it’s in the hospitals’ best interests to incentivize them to stay local, and waiving admin costs is an easy way to do that.

      • Thomas Jorgensen says:

        The real problem is that no US insurance company can credibly commit to being honorable.

        Assume the hospital has absolute trust in the current CEO of “Honest Jane Insurance”.

        Further, assume Jane is, in fact, a living saint who somehow managed to ascend to the top of a corporate pile and is worthy of that trust. Jane gets the same discount they give the Amish and does not abuse it. Fine? No. Because then Jane suffers an unfortunate skydiving incident, and is replaced by T. Ypical MBA.

        Defecting on Janes agreement and taking the provider for everything they possibly can will produce at least one quarter of extremely high profits. That will get T. Ypical MBA a new yatch, so it is more or less certain to happen.

        The provider knows this, so will not give Jane the discount in the first place, unless she takes the entire firm private and gives them assurances regarding succession.

        The only way for a corporate insurance firm to realize this particular savings is by vertical integration.
        An insurance firm that just directly runs health care and dispenses with billing all together for inhouse procedures could credibly commit to honor, because commitment would be written in concrete poured into hospital walls, tiles laid in surgical suites and so on, which new management could not profitably abandon on a whim.

    • TheObeseDog says:

      Savings is definitely passed on. Insurance is very competitively priced. Assuming savings is not passed would require VERY strong evidence. I say this from direct experience.

  25. Well... says:

    Maybe a real Amish expert will correct me, but I’ve read quite a bit on the Amish and there are some trivial facts about them I think you (Scott) got slightly wrong…

    Pretty sure the Amish are mostly descended from the Swiss, not from Germans. Also, if you were going to generalize when most of their technology is “from” I think it’d be around the late-19th/early-20th century, which is when they earnestly began their extreme selectivity of new technologies.

    The Amish exceptions to things like health insurance and social security are based not on a miracle but on their religious forswearing of gambling (because insurance is basically betting against yourself); I believe they won the exemption based on a 1st Amendment ruling.

    • acymetric says:

      The Amish exceptions to things like health insurance and social security are based not on a miracle but on their religious forswearing of gambling (because insurance is basically betting against yourself); I believe they won the exemption based on a 1st Amendment ruling.

      Well that’s what insurance used to be and is supposed to be. Mostly now it is just an expensive discount plan.

      • Well... says:

        It’s now both. Yes, your health insurance premiums make your specialist visit copay $40 instead of $200, but they will also make your catastrophic hospitalization bill $5,000 instead of half a million.

        • acymetric says:

          Well yes, that’s why I said “mostly” and not “entirely”. I stand by “mostly” though.

      • Wency says:

        “expensive discount plan”

        When using “insurance” to pay for low-cost, routine expenses, I think the term is “prepaid consumption”.

        I would think a very large percentage of insurance expenditures, in dollar terms, are still for true catastrophic causes, even if the vast majority of insurance *claims*, in absolute numbers, are for prepaid consumption.

        • mitv150 says:

          The large majority of U.S. health care spending is indeed “prepaid consumption” in the form of chronic disease management. This article suggests that, in 2005, chronic disease management consumed 78% of health care spending in the U.S. and was rising. These expenses are, however, more likely to be high-cost routine expenses and not low-cost routine expenses. In a sense, these are indeed “catastrophic” in that a diabetic cannot live without appropriate medication.

          A common objection to a health care system that incentivizes shopping around for best prices is: “you don’t have this option when you have a heart attack.” This is true, but those acute “don’t have a choice” requirements are a much smaller portion of the pie than people expect. It is very expensive to be treated for a heart attack. It is also very expensive to take blood pressure medication continuously for 35 years.

          The point to recognize here is that, for chronic disease management, a consumer could reasonably shop around for lower prices if the incentive structure of the system lined up to permit such price discrimination. In the current system, for many people (clearly not all), there is zero incentive or reason to shop for a lower cost option.

          • Garrett says:

            > It is also very expensive to take blood pressure medication continuously for 35 years.

            One of the best first-line medications, losartan, an Angiotensin Receptor Blocker, is available for about $5/month from GoodRx, and 9$/month via Walmart’s cheap med list. That’s not exactly a high cost.

    • Beck says:

      There are Swiss Amish in the US (maybe in Indiana?), but they’re a separate crowd.
      The better know groups in Pennsylvania and Ohio are mostly German, I think.

      Edit: Wikipedia link

    • The religious ideology originated in Switzerland, the people originated in Germany.

      • Wency says:

        In addition to this point, the distinction seems pretty artificial to me for much of the relevant period. Would a 17th-century inhabitant of the Palatinate consider the German Swiss to be foreigners? Probably less foreign, at least, than the Prussians.

  26. jrdougan says:

    As I understand it the private mutual-aid societies (aka friendly society, benefit society, etc.) were an significant factor and what they did was absolutely a form of risk pooling. So I don’t think you can discount risk pooling organizations as as part of a solution.

    When and where did your grandmother work as a nurse? She may not have understood that many of the “religious organizations” were risk pooling societies (eg. Ladies Friends of Faith Benevolent Association), and she may have nursed late enough to be in the period where they were losing influence. It is also my understanding the immigrants and other low-income people were big participants in the mutual-aid societies, so I’m not sure about the claim that poor people were free, except possibly in the cases where the hospitals were run by charitable orders.

  27. Filareta says:

    Can’t Singaporean health care system be described as modernized and adapted to industrial and urban society Amish one? You have personal account for some savings forced by government, which is your main source for funding your health care. You can use it later in life for retirenment and savings can be inherited (if I remember correctly). So everyone has clear incentive to not overspend. There’s a state insurance, but only for rare diseases and catastrophic admissions, so it’s a proper insurance, not prepaid scheme. Finally, treatment of all those with chronic pain, mental health, end of life care and simply treatment of impoverished are all subsidized by taxpayers. There are various classes of quality of care – if you want to cut your bills, simply choose slightly worse treatment, with more crowded rooms and so on. Also, being a city state, Singapore has great advantage – if you want to chose hospital with best quality or lowest price, you don’t fly from Pennsylvania to Tijuana. You simply go with car or subway, in worst case to the other side of the island.

    Results? Singaporeans are very healthy, their health care is one of the best in the world according to various ratings, and it’s also cheap – both on the personal and social level (total spending only 3% of GDP, if I remember correctly).

    • keaswaran says:

      I believe Singapore also operates at an interesting margin between the high-cost developed world and the low-cost developing world. When I visited last year, for instance, it was clear that you could get meals at either price ($15 for a dish at restaurants or $1 for a dish at hawker centers). Are hospitals able to ride this margin by hiring a low of low cost Malay labor for janitorial duties and lower-skill nursing activities?

      • Matthias says:

        There’s no upper limit for prices for restaurant food.

        But hawkers are more like 2 to 3 USD, not 1 dollar.

        Singapore has relatively free skilled migration. Nurses from eg the Philippines are welcome. Not sure how much of an impact on costs this has.

        Almost all medical treatment comes with some co-payment.

    • eric23 says:

      I don’t know if those “results” are results of their health care funding policy, or results of other social factors that exist in Singapore.

  28. PhaedrusV says:

    It’s interesting to see this deep dive into the cost increases predicted by economics theory. Basic supply and demand theory predicts that increased demand for medical care (i.e. more people with more coverage due to use of insurance) increases costs, but the specific mechanism is left blank.

    Since Scott also brought up college costs it’s neat to compare how the two institutions responded as economic theory would predict: to arrive in the same place of highly inflated costs, but by different paths.

    It’s interesting to see one of the few areas where modern economics actually make solid real-world predictions, yet appropriately refrains from predicting the specific mechanism. Being able to predict the endpoint without solving each step, (see also: 3 body problem) is somewhat rare.

  29. Nick says:

    Scott, a helpful contrast here might be healthcare cost-sharing ministries, which have advantages similar to the Amish and can likewise offer prices much lower than insurance:
    1) they’re nonprofits
    2) they bargain with providers
    3) members are required to avoid certain risks, like smoking
    4) members are asked to do some comparison shopping.

    The question of honoring bills on the other hand is a fraught one. Sometimes members will get procedures that are obviously not covered under a ministry’s up-front, when-you-sign-up guidelines. In that case they will refuse to pay for the procedure, which the hospital no doubt pattern matches to “insurance company trying to scam us.” Refusing to pay for certain things keeps costs low elsewhere, though.

    Cards on the table, I work for a healthcare cost-sharing ministry, so I have experience with how this works under our guidelines. I can’t speak for other ministries, but until recent price changes we covered individuals for <$50/mo and families for <$150/mo if you were willing to accept a high deductible and some restrictions on what we'd pay for, and the no deductible plan was about three times as much.

    AMA.

    • Randy M says:

      In that case they will refuse to pay for the procedure, which the hospital no doubt pattern matches to “insurance company trying to scam us.”

      How often is this clearly understood by the hospital before the procedure (and surely it doesn’t include many emergency procedures like broken bones or ruptured appendices)?
      It seems hard to call it fraud if everyone knows the policy upfront. It could be that the hospital is used to hard line negotiations that fold later and figure it’s better to send a bill than turn away a customer.

      • Nick says:

        The hospital isn’t going to know in advance what our guidelines are, though there are of course categories most health insurance won’t cover. And there is definitely an aspect of “send bill first, ask questions later”; they won’t literally accuse us of fraud, but they may think we’re playing games with them, and will negotiate accordingly.

        The commonest thing we refuse to pay for is probably pointless tests, but that’s more a problem we have with alternative medicine people. Another big one is pointless ambulance rides, which is one I could see surprising providers.

        • Randy M says:

          The first thing done in a medical waiting room is to fill out a medical history survey and insurance information. Is it not common practice for hospitals to run this information to find out things like what policies are covered, whether they are in network, etc.? It seems like that could be done via automation.

          I hate to be the “there ought to be a law” guy, but hospitals should be required to tell you before admittance (in non-life threatening emergency situations) how their treatment fits into your policy, and insurers should be required to have a way for hospitals to instantly know this.

          • Nick says:

            On the contrary, automating something like that would be very hard. We can’t even automate our own eligibility process, and we have CPT codes + medical history for every procedure we’re asked to cover. Humans have a knack for designing complex things that are not so hard to wrap your own mind around but really stupidly incredibly hard to describe correctly in a formal system like a computer program.

            It would be really nice if hospitals could do this, but it’s impractical. Better that the responsibility falls on the patient to know the policies of their insurance (or cost-sharing) program.

            (I think I could get behind a similar law, requiring insurance/cost-share/whatever places to provide an “eligibility calculator” that spits out a yes-no answer based on the information provided on the waiting room form you mentioned. Difficult or not, we would build it if we legally had to, and it would incentivize us to simplify our eligibility rules, which is a good thing. We could produce savings elsewhere by eliminating positions that negotiate with providers. But placing the responsibility to build such a thing on hospitals would be a terrible idea.)

          • Randy M says:

            What I’m envisioning is that a hospital receptionist will take your number and provider, go to the provider’s website, enter your number and the procedure code, and get a form to print out that says “You are here for [X]. This hospital [is/is not] IN YOUR NETWORK, and this service [Does/Does not] qualify for coverage, but you will be required to pay [$Y] copay per your policy terms. Additional services will require additional authorization. If your condition worsens to a state where you cannot consent to further treatment, the hospital is authorized to make medical decisions in accordance with best practice. To dispute, call 555-1234. Sign here _ _ _”

            Finding all this out obviously happens later in the billing department. It should be done up front. Preferably with the actual price or a price range included.
            I understand it is in various parties interest to haggle a lot of this behind closed doors with official negotiators trying to decrease and increase various numbers, but I don’t think it is in the consumers collective long term interest.

          • Nick says:

            Eligibility doesn’t always come down to procedure code; medical history is important, too, like for preexisting conditions. Does the receptionist have to check that, too? For some insurance companies a little checkbox saying “this is a preexisting condition” might suffice. But our rules for covering preexisting conditions are pretty complex. Medical history also matters because our cheaper plans cap coverage for illnesses at $125,000, unless you have our catastrophic plan (which is unlimited). Good luck getting all that from the waiting room form.

            If there were a law, we might be forced to simplify. I’d really like to see that, but I dunno how well it would work; many regulations just distort the market in ways that are hard to see ahead of time. I’m sympathetic to the idea, because I want to live not by kludge, but I’m not confident in it.

          • Randy M says:

            That’s a good point. I’d settle for “This hospital [is/is not] in your network based on information provided. You are here for procedure X, which is classified as [emergency/essential/elective]. Please call your insurance at __ to ensure the procedure is covered, or you may be liable for the full amount of $X for this visit.”

            Then again, I think this is probably already done, at least verbally, for many visits. But it was pretty annoying when we got billed out of network for an annesthiologist despite every other part of the operation being in-network. I’m not quite sure what we would have done differently in the moment, but when planning it would have been good to know this would be an issue.

          • JimJ says:

            I’ve never been asked (or even allowed, really) to type that information in; I assume that part of the administrative burden is getting someone to type it up later. I have called in advance, for things like imaging, and the insurance company insisted they couldn’t know the cost because it depended on the hospital (and telling them the hospital wasn’t enough, because the people allowed to answer customers weren’t able to see that information), but the hospital insisted they couldn’t know because it depended on the policy, and they had no way to look it up in advance even with that information; at least for the department I knew how to contact, they could only verify that I was “in-network”, then submit a bill afterwards and see what happened. I got this story from multiple hospital chains, and have no reason to believe that my insurance that year was any less transparent than other years.

            My Doctor’s office is owned by the same corporation that owns a local hospital and provides my current insurance. Even within those corporate walls, my Doctor (and his staff) have been very frustrated by trying to figure out why certain medications are being rejected by insurance. They are NOT told when writing the script (through an automated system), nor when sending (through an automated system) to the pharmacy. The pharmacy is only told when they start to fill-and-bill, and no one (the pharmacy, the doctor’s office, me) gets told what the problem is. We do get (different) numbers for appeals, and between the three, we eventually figured it out, though I was on samples for well over a month.

            A few months later, that same medicine is again being rejected for need of (another?) pre-auth, again with no indication of why the previous pre-auth is no longer sufficient. Technical incompetence can explain a lot, but this level… well, I suspect it is not a coincidence that they save a month’s worth of costs when either I give up and just pay or the Doctor hands out samples.

            I’m sure they are following their procedures, and these procedures even minimize costs at one particular cost center, but they certainly don’t minimize costs for the system. But incentives are at the cost-center level.

        • Garrett says:

          > Another big one is pointless ambulance rides, which is one I could see surprising providers.

          As someone who volunteers in EMS, I’m deeply interested in how you categorize this. There are patients that I’m reasonably certain don’t need an ambulance, but which I’m preventing by law from dumping on the curb. I’ve also dealt with hospitals which would frequently pay us the BLS rate for what is effectively a wheelchair van trip simply to be able to get the patient out the door.

          • Nick says:

            I don’t think it’ll be too helpful; our rule is too strict to cover all worthwhile ambulance rides. We only cover rides where
            a) your condition is life-threatening, and
            b) the hospital you’re at can’t adequately treat you, but
            c) the hospital you’re being transferred to can.

            For example, if you’re at a local emergency room but you need a major trauma center, they can send you to one and we’ll cover it. Likewise a lifeflight. But we’ll never e.g. pay for an ambulance from your home to the hospital.

          • teageegeepea says:

            we’ll never e.g. pay for an ambulance from your home to the hospital

            Huh, I tended to assume that’s the typical ambulance route, rather than from one hospital to another. I also remember the advice that you don’t want to move an injured person yourself, but instead call someone who can move them safely.

          • Garrett says:

            @Nick:

            The intent as admirable, but I’m uncertain how this would work in practice for a lot of the cases I deal with.

            For example, we do a lot of transfers from a small community hospital to a larger, comprehensive facility. The patients need services/consults which aren’t available at the originating facility. They also fall into this weird area where the patient condition isn’t serious enough to qualify as life-threatening, but not certainly safe enough to discharge for follow-up. The solution I’m familiar with is to use an ambulance because it provides for a “continuity of care” and would have a medical necessity form signed. So in this case, how would this be handled? The things I can think of would be to have insurance pay (which your criteria would probably exclude). Have the patient pay directly (if they don’t want to, they can sign AMA paperwork and transport themselves). Or, have the hospital eat the cost and hide it in general markup elsewhere (this happens a lot for indigent patients).

            Do you cover prehospital treat & refuse? We get a fair number of unresponsive hypoglycemia patients who we are able to treat on-scene and safely release. The standard model for insurance pays only for transport (and thus an additional ER visit) even though the hospital visit wouldn’t be required.

            And, are you not covering legitimate emergency transports from home? I understand that most calls really are bullshit. But what about severe injuries or medical complications which require emergency interventions? Is the expectation that people pay out-of-pocket? If so, does this result in people taking on a lot more risk to avoid costs?

            @teageegeepea:

            Most people don’t understand that most hospitals can’t provide every service in a timely fashion. So people get shuffled around a *lot*. Also, nursing home to dialysis clinic and back. There are whole private ambulance companies which operate (before going bankrupt) to do just that kind of interfacility transport (IFT). Even for the 911 services I volunteer with, about half of our volume is just hospital upgrades.

            > you don’t want to move an injured person yourself

            Err. It’s complicated. Imagine yourself lying on the floor, healthy, of the 2nd floor of a house. Would you trust that 2 of your friends would be able to pick you up and get you outside your house onto the sidewalk safely? On the ambulance we have lots of simple-but-effective equipment to do just that. And then you have to worry about the risks of exacerbating an injury, but that’s mostly broken bones and penetrating objects, though that’s been overblown. Emergency Medicine, as mandated by orthopedic surgeons. *Grumble*

          • Nick says:

            @Garrett

            They also fall into this weird area where the patient condition isn’t serious enough to qualify as life-threatening, but not certainly safe enough to discharge for follow-up.

            It usually wouldn’t be covered (I say usually only because a doctor might say it’s life-threatening anyway); patients would call family to transport them instead or something. The same is done in many home to hospital cases where there isn’t a clear need for an ambulance.

            And, are you not covering legitimate emergency transports from home? I understand that most calls really are bullshit. But what about severe injuries or medical complications which require emergency interventions? Is the expectation that people pay out-of-pocket? If so, does this result in people taking on a lot more risk to avoid costs?

            In legitimate cases we recommend private services, or, where available, free or cheaper services like fire departments transporting. But it still won’t be covered, so people do end up paying out of pocket. It doesn’t seem to result in much more risk; members are more willing to eat the cost of a $500 or $1500 ambulance ride than risk complications.

            Do you cover prehospital treat & refuse? We get a fair number of unresponsive hypoglycemia patients who we are able to treat on-scene and safely release. The standard model for insurance pays only for transport (and thus an additional ER visit) even though the hospital visit wouldn’t be required.

            Not sure, TBH; I think it would depend on whether the onsite treatment is separately itemized. I looked for a few cases in our system, and they weren’t, so it wasn’t covered.

  30. Matt M says:

    My understanding is that health “insurance” in the US was less an innovative response to a problem of high market prices for health care, and more of an innovative response to new deal price controls and extremely high nominal tax rates.

    In the 1930s-40s, it was in most cases literally illegal (or extremely inefficient) to offer your employees a market-clearing wage. So employers had to get creative to attract and retain quality employees. Thus, they came up with schemes wherein the employer would provide goods-in-kind rather than money wages, which were non-taxable and exempt from controls. One such good was basically a guarantee to provide free medical care. Note that this isn’t really “insurance” in the traditional sense (the whole ‘your car insurance doesn’t pay for an oil change’ analogy is cliche at this point, but remains basically accurate).

    As soon as that became relatively common, we suffer from the “spending other people’s money” incentive problem you outline and everything goes downhill from there…

    • North49 says:

      This is more or less my understanding of the chain of events as well. One perhaps relevant detail you left out of the story is that unlike other forms of in-kind/nonmonetary remuneration, employer sponsored healthcare got codified as permanently nontaxable when they were closing other loopholes, whereas out of pocket medical expenditures were not similarly tax-advantaged. Which is why insurance remained so pervasive beyond the wage controls.

      This story seems uncontroversial as the “original sin of US healthcare” among my various economics podcasts. Though my listening preferences skew neoclassical/free market.

      Given how often I encounter it, I thought it was more or less common knowledge, and was surprised Scott didn’t even mention it.

      • sclmlw says:

        I’ve heard the same story, but expanded in the following way:

        Employers were responding to salary/compensation ceilings instituted during WWII by offering free health care to their employees. The members of Congress who instituted the wage price controls saw this happening and cried foul. They started working toward ending the practice legislatively; they wanted to count the in-kind benefits as monetary compensation that would count toward the wage price ceiling.

        Meanwhile, FDR was facing a strong public movement toward socializing several major industries – especially health care. Despite his reputation on the Right as being practically a Socialist, he didn’t support this move. He couldn’t shut down the movement by just saying ‘no’, especially as it was an international political call. He needed an alternative to propose to voters. Employer subsidies was already happening, and was a viable alternative that would shut down the socialists. (Indeed, it has continued to be effective at doing just that despite a lot of subsequent dissatisfaction with the current system.)

        Around this time certain employers were pushing back on the attempts by Congress to count health care subsidies as part of price-controlled compensation limits; they did this by proposing contrary legislation to actually subsidize the practice. This was credible because FDR and anti-socialists got on board with the idea. Which is how we ended up with the current employer-based system. It’s a very ‘bootlegger and baptist’ story, not a system that was rationally designed to solve problems specific to delivering health care in an affordable/effective/responsible way.

        I don’t see any way to directly eliminate employer subsidies, since it would never pass politically. Perhaps the only way around this would be to expand the subsidies employers get to individuals, who could then form their own co-ops and shop around as they see fit. If that happens I predict that the subsidy itself becomes part of political platforms. “Vote for me and your employer won’t have to pay as much for your health insurance!” isn’t a viable campaign slogan. But on an individual level it is. At a certain point the subsidy increases and you end up with the equivalent of a voucher-based system of health care, with the possibility of paying extra for additional care.

        • North49 says:

          I would love to go to a ‘bootlegger and baptist’ themed party some time; I’m sure the antics would be legendary and self-reinforcing.

          There has been some positive movement recently in giving individuals access to pre-tax Health Savings Accounts, which have the potential to disrupt some of the worst parts of the insurance/healthcare industrial complex. They’re currently extremely nerfed, but they do give me a tiny bit of hope!

          • sclmlw says:

            Bootlegger and Amish parties are off the plug!

            I think the biggest hurdle is from libertarians and conservatives who don’t want to expand government influence/expenditure into health care more. The problem is that government is already surreptitiously spending most of the money in health care and creating distortions and perverse incentives all over the place. And you’ll never get that money back out again. You have to replace it in a way that doesn’t affect current recipients. Thus, giving away subsidies to individuals to create an individual market/voucher system is the only way out that still preserves a market-driven approach.

  31. AlexanderTheGrand says:

    Common understanding is that a large amount of the low life expectancy in the past was skewed towards infant and early childhood deaths. It seems almost unbelievable to me that the Amish, with their enormous families and non-modern healthcare practices, would do much better on this front than the rest of the country. And I assume they couldn’t possibly have those numbers if this wasn’t the case.

    Long colonial life-expectancy therefore seems like it could be largely a reporting error. If Amish families were largely outside of the hospital/birth-certificate system in the past, they may not report the life OR death of a 2-week-old infant. Since this can skew the average so strongly, that could be the bulk of the difference between 50-year expectancy and 70.

    • John Schilling says:

      What do you mean by “non-modern health care practices”? The Amish aren’t using herbs and leeches; they mostly vaccinate their children, and when they become seriously ill send them to English hospitals. For most things they are going to skew closer to using a small-town general practitioner than flying off to the bestest most elite specialist at the Mayo Clinic, but in terms of childhood mortality they’re using pretty much everything we developed over the century or two that childhood mortality went from horrific to almost negligible.

      And that’s pretty much always been the case, AFIK. The Amish have never been averse to using something reasonably close to state-of-the-art medical technology to keep their children (and in most cases their adults) from dying, whether that was the early 18th century state of the art 200 years ago or the current state of the art today. Maybe a generation or so behind, more because they are rural and reluctant to travel rather than from any religious opposition, but that’s not going to make a huge difference in infant or childhood mortality.

      There’s no need to posit secret graves filled with unrecorded Amish baby-corpses, and I don’t think they are hiding their dead babies from researchers who are honestly and respectfully asking about how their health care system works.

      • Garrett says:

        Childbirth complications happen. Many of them can be handled by midwives, etc. And there are a small number of healthcare providers I’ve dealt with who are trusted by the Amish. They aren’t called unless something is wrong, but they will go in and help with some of the severe cases.

        And a lot of died-in-childbirth can be addressed with some basic surgical knowledge and antibiotics.

  32. Roy G Biv says:

    I think any discussion of health care costs has to address the distorting effect of Medicare/Medicaid on health care prices.

    This is how it was once explained to me by a health care administrator. Imagine we created food insurance to ensure that everyone could afford food. Each insurance company would negotiate with grocery stores for what it could cover and at what price. Can of corn is 80 cents? United might say, “We’ll cover 65 cents with a 15 cent co-pay.” Blue Cross says it will cover 68 cents with a 12 cent co-pay. And the uninsured pays 80 cents.

    Medicare/Medicaid enters the picture and says, “We will pay 63 cents for that can of corn with no co-pay.” Now, the store has to raise prices on the uninsured in order to make up the loss it takes on the M/M cans. Before long, cans of corn are $2.79 sticker price. The United/BCBS shield customers are still paying 80 cents between insurance/co-pay and the M/M customers are still paying 63 cents w/no co-pay. For those customers, the hospital writes off the difference between the sticker price and the price paid. But the uninsured customer is paying $2.79 and before long, can’t afford food.

    My guess is that because the Amish are reliable customers in terms of payment, the hospitals are charging them 80 cents for the can of corn rather than $2.79.

    There is no optimal solution to this problem. I think the fairest solution, which spreads out the costs across society, is to mandate that hospitals have one price for a can of corn for everyone, regardless of private insurance, government insurance or no insurance. Each hospital can set its own price (with regulations protecting against price gouging) and insurance companies can determine how much of that price it will meet. And M/M has to cover sticker price. This will result in higher taxes, but will also keep healthcare prices honest.

    • methylethyl says:

      When our son was born, the hospital billed us $1000 for the nursery. Which our son never entered. At the time, Medicaid was on the hook for the bill. I tried to call up and get it changed, because the naked fraud there bothered me. But I couldn’t get any traction on it because I wasn’t the payer. And I had no idea how to get in touch with the actual payer to report said fraud.

      So… it’s not that M/M pays 63 cents for the can of corn. It’s that M/M patients get a can of corn, but M/M is paying for a can of corn, and also a can of peas, and seventeen pounds of diced frozen carrots that don’t actually exist.

      This fraud has been around for a long time. When I was born, my mother was charged for a private hospital room (she’d been in a shared room). She dislikes fraud as much as I, and fought it for months. She ended up asking for he room number of her so-called private room, and then went to the hosptial, found the room number, knocked on the door, and found it was… an office. She then threatened to go to the newspaper with her billing issue, and they finally relented. This billing odyssey lasted over six months.

      They count on everyone being too exhausted to fight.

    • 10240 says:

      Another thing that would help is to require disclosing and agreeing upon the price ahead of time, to allow comparison shopping. At least outside emergency situations; those should be required to have a published price list.

      Do I understand correctly that in America, the hospital just makes up a number after they treated you, with you having an obligation to pay it, and basically no right to dispute it? That sounds like they have no incentive not to charge as much as they feel like.

      • Roy G Biv says:

        It’s not that they make up a number after they treat you. It’s that the published price is outrageous and the only people who pay the published price are the un-insured.

        To use a very hypothetical example: Hospital has a published price of $787 for a simple blood test to check cholesterol. Patient A has Medicare/Medicaid, which will pay $57 for that test. Patient B has private insurance, which will pay $87 for that test with a $25 co-pay that the patient pays directly to the hospital (so $113 total). Patient C has no insurance and is charged $787. If Patient C is tenacious and bureaucratically savvy, they can probably negotiate/harass the hospital into cutting that price down. But most uninsured are not tenacious and savvy in that way.

        My argument is that the best way to handle it is that the blood test costs $113 for every patient. But that will require a pretty massive tax increase because M/M’s prices are relatively low.

        • SamChevre says:

          The orders of magnitude of difference between insured price and uninsured list price here are, if anything, too low.

          • Matt M says:

            Yeah… the viability of the entire system seems contingent on hospitals to trick a small quantity of poor people into paying 10x what the normal customer would pay for simple/routine procedures.

            It’s almost like a complete inverse of the “whale theory” of how free to play games work, wherein a few rich people spending a lot of money subsidizes the game for the vast majority of users who never pay anything.

          • Edward Scizorhands says:

            It should be obvious that the poor people aren’t paying $800 for a blood test, either. They don’t have the money. If you bill them and try to collect, it goes to collections for $8.

            The insane list price gives the provider room to negotiate however it wants.

            I’m not sure which is worse, this or colleges with insane list prices. They are each horrible in their own way.

          • Garrett says:

            Medical bills rightfully get the combined distrust of airlines and car repair shops.

            Airlines, because you’re always pretty certain that someone else on the plane managed to get a better deal than you did, but you don’t know how and never think it’s fair. And car repair shops because you aren’t certain whether the treatment being proposed is needed for your benefit or their pocket book’s. And because even if a procedure doesn’t fix the problem you’re still on the hook to pay for it.

        • 10240 says:

          Why do hospitals not have to compete, when it comes to uninsured patients? What happens if you ask multiple hospitals, and tell them you are going to do your blood test at the one that offers you the lowest price? Hundreds of dollars are something it’s worth to even travel to a different city to, especially if you are poor.

          In a big city, why don’t providers that cater primarily to uninsured people at reasonable prices spring up? Or ones that cater to anyone except Medicare?

          • SamChevre says:

            Because most uninsured patients can’t pay anyway.

          • Loriot says:

            Or ones that cater to anyone except Medicare?

            My understanding is that many practices do refuse to accept medicare patients.

          • Garrett says:

            > Or ones that cater to anyone except Medicare?

            If you commit to not accepting Medicare, Medicaid and Tri-care, you can avoid the mandates of EMTLA. This means that you don’t have the mandate from Federal law to accept patients who can’t pay.

            The problem is that you forgo Medicare reimbursement which is a major part of hospital revenue. I can’t easily find specific numbers for percentages. Remember that hospitals bill based on admissions whereas a lot of the medical care younger folks get is likely to be done on an outpatient basis. In other words, the revenue for hospitals is in Medicare.

            Also, in my area, as a consequence of State law, ambulance services cannot refuse to accept Medicaid. Woo! $180! *Grumble*

          • 10240 says:

            @Garrett The thread starter claimed that Medicare pays basically below cost, which is why hospitals have to overcharge the uninsured. I have no idea if this is true, but if it is, then not accepting Medicare shouldn’t hurt the hospital.

            That ignores economies of scale, of course; perhaps Medicare pays more than the marginal cost, but less than the mean cost. However, in a big city with several hospitals (or a bigger region if people are willing to go to a different city for lower prices), a hospital should have enough scale to keep prices at a reasonable level even if it receives only a fraction of the patients in the area.

          • Garrett says:

            @10240:

            Medicaid pays below cost. Laughably so at times.

            The Medicare fee schedule is based on the full cost of the treatment at whatever the government has decided it “should” cost to provide that service, explicitly including things like cost of facility operations, etc. The real question is whether the costs determined by the government match the real costs.

            I’ve also read some people writing about hospitals, for some reason, being pathologically interested in growing revenue rather than growing net profit which might entail cutting costs in a reasonable manner.

      • JimJ says:

        Essentially, that is correct. In theory, the number is made up in advance, and they just won’t tell you what it is. In practice, going in for what seems like one thing will involve hundreds of separate services, and precisely which of them get itemized (let alone charged for or paid for) is somewhat random. If the same patient with the same insurance returns for the same services (e.g., a second infusion of the same cancer treatment) at the same provider, then … the level of itemization still includes some randomness, though it is reduced.

  33. nkurz says:

    The Milwaukee Journal Sentinel ran a great article last year describing the life of a rural doctor who cares for the Amish community in a small Wisconsin town: https://www.jsonline.com/in-depth/news/special-reports/2019/10/10/wisconsin-country-doctor-treats-amish-health-studies-genetic-diseases-medicine-la-farge-james-deline/3901472002/

    It also addresses a lot of the questions raised in this post about how the system works: hard-working low-paid medical practitioners, generally lower healthcare expectations, and resilient community spirit. It also goes into how chronic diseases are dealt with, given a prevalence of rare genetic diseases in the community. It’s one of the best news articles I read last year — highly recommended.

  34. methylethyl says:

    I have heard that one of the primary functions of old “secret brotherhood” organizations (masons, shriners, elks, lions, moose, oddfellows, Knights of Columbus, etc.) was basically insurance: take care of a member’s family if the member dies, cover extraordinary medical expenses, etc. The small memberships, secret oaths, and attendance requirements helped to cut down on fraud and free riders, to keep expenses low. Possibly one of the key reasons for the decline of these organizations was the rise of employer-provided insurance.

    • Nick says:

      The Knights of Columbus still do life insurance! I have a friend who works for them selling life insurance.

  35. Anonymous Bosch says:

    Lots of Americans get an insurance plan from their employer, and then consume health services in a price-insensitive way, knowing very well that their insurance will pay for it. Sometimes they will briefly be limited by deductibles or out-of-pocket charges, but after these are used up, they’ll go crazy.

    As someone who also has a very good insurance plan but is aware of it, you really should look into recent trends regarding deductibles.

    In particular, KFF has a metric which tracks the proportions of medical spending and insurance deductibles. On average, people now spend close to half of the year out of pocket before meeting their deductible, a big jump from 15 years ago where the average deductible was met by February.

    I consider this to be an inadequate metric because deductibles are legally capped while spending is not and so the high spenders are no doubt pushing up the average much higher. I can’t find the same data using the median, but I am pretty confident it would be worse (meaning the median worker who doesn’t have stage 4 outliersarcoma has to go longer out-of-pocket before meeting their deductible). But I can’t find a version of this same data using the median.

    My point though is the whole “people over-consume health care because insurers handle it all” framework seems predicated on 90s and 00s numbers that don’t hold up as well. Talking about how people are “briefly” limited by deductibles does not jibe with how I hear my working-class family and friends talk about insurance.

    • Matt M says:

      I’ve been predicting/hoping that the increasing adoption of high-deductible plans will ultimately, in a way that nobody ever really planned/wanted, lead us to something much closer to a free market system where prices actually matter, people comparison shop, etc.

      • Anonymous Bosch says:

        Maybe very marginally but for the most part people in high-deductible plans can’t do much shopping. They’re stuck in-network if they want the expense to apply towards their deductible and there won’t be much diversity in prices since providers will converge on the reimbursement rate. And meta-shopping for good insurance plans is still mostly not a thing outside Obamacare since they’re tied to your job.

        • JimJ says:

          And there is no price transparency, and if there is a published price, it tends to be so outrageous that the only meaningful price differentiation is “how often will I not actually be charged, or at least not today, and maybe not turned over to collections”

  36. smilerz says:

    I don’t want to ignore the fact that the Amish do live a little bit shorter lives.

    Above, I said that the most important factor is that the Amish comparison shop. Everyone needs to use other people’s money to afford expensive procedures.

    I think you are underestimating how much the end of life treatments are and how much they contribute to the overall cost of healthcare. Crisis of Abundance by Arnold Kling went into it a number of years ago and found a pretty substantial factor.

    You also have to wonder what the QALY of those incremental years really are.

    “”

    • Canyon Fern says:

      And, for those who missed it, the Arnold Kling that @smilerz mentions showed up in this thread with a mention of his book.

    • JPNunez says:

      But the QALYs for old people who had an active life are way higher than for sedentary old people. No smoking, no drinking will get you a lot less problems in old age. Also big families with tons of sons and grandsons present to help with care.

      Which makes their lower life expectancy worse. I wonder what number we get if we take out a bunch of deaths by something the english prevent more easily, but would need disaggregated data.

  37. Ouroborobot says:

    I’d love to see some data on the prevalence of payers “scamming” providers. It sounds like Scott has had some bad experiences. Full disclosure: I work for a nonprofit medicare advantage payer, and have previously worked for an LTC insurer, though my speciality is data engineering and BI, not insurance per se. My experience in ten years has been that everyone I have worked with is honest, and has done nothing more than hold members and providers to the letter of the policies, and frequently generously bend the rules in favor of members and providers. It is quite common, however, for us to encounter fraudulent claims submitted by providers. Granted, I have never worked for a large for-profit insurer, and perhaps my experience would be different there.

    • J Mann says:

      I’d love to see some data on the prevalence of payers “scamming” providers

      I’d also like to see some more info on what Scott has in mind. In my experience, a lot of the administrative cost runs along the lines of:

      1) Insurer: We’re not paying because [our information shows that the husband’s insurance has first duty to pay/we don’t see that the symptoms justified a bill for a high level office visit/you haven’t shown that the member tried this lower cost alternative to surgery first/etc]

      2) Hospital: You should pay because [here’s a letter showing that husband’s insurance has already paid everything they’re going to/here’s some medical records showing that the high level office visit charge or surgery is appropriate under your policies]

      3) Insurer: Ok, here’s some money or No, you still haven’t met our policies because . . .

      The problem with the idea of insurers reducing costs by just paying everything is that some provider will start rounding more and more of their office visits up to the higher levels, ordering MRIs on the machine they just bought when they don’t necessarily need to, etc.

      One way the Amish might be addressing this prisoner’s dilemma would be if they push the providers hard on “did you bill the auto insurance carrier” “does this really need to be a 45 minute visit” “do I really need an MRI,” etc.

      • JimJ says:

        What I have seen (as a consumer, and in one job writing a payment system for an insurance company — the project that left me feeling the most queasy of anything I’ve worked on) sounds sort of like that, except that often there is no reason given in step one, or the reason is unrelated (as if someone made a single-key typo in the denial, and this brought up a bizarre “explanation” that didn’t get checked), or the reason is just untrue (as admitted if — but only if — the provider or patient has the energy to fight).

        And then in step three, “you still haven’t met our policies” is pretty common, and often for reasons that were not mentioned in step one, and are unrelated to anything that happened in step two, and often has the same problems I mentioned about step one.

        I suspect the way they get out of the prisoner’s dilemma is that they are reasonably large portions of the local population in areas with relatively few doctors, and they just start to avoid the doctors who routinely round up. (And, of course, the doctors won’t have as much anger/revenge motivation to pay someone to spend time rounding up if the bill isn’t going to be disputed afterwards.)

    • TheObeseDog says:

      Same here. I also work in insurance and I’ve only ever seen denials for things that are clearly not covered. Moreover, insurance companies cannot just deny willy-nilly. They can and will get sued ALL the time. Most of these lawsuits are without merit (my perception, not a lawyer). Ive never been in a meeting where I’ve ever seen any pressure to deny legitimate claims.

      Maybe insurance contracts are just to hard for laypeople to understand so it feels like fraud?

      • The Nybbler says:

        My experience is that insurance companies will often issue an initial denial for something that clearly IS covered, in the hopes that the customer will just give in. This isn’t technically insurance fraud, but only because the insurers helped write the laws.

        This happens in other fields of insurance too; insurance companies can and do freely lowball claims, but if the insured pads one, they’re subject to criminal penalties.

        • Edward Scizorhands says:

          My experience is that insurance companies will often issue an initial denial for something that clearly IS covered, in the hopes that the customer will just give in.

          I know someone who will absolutely swear to this behavior from the inside. They worked for an insurance company, and they were told to deny the claim even though it was covered. “Don’t worry, they’ll just refile later.”

  38. n-alexander says:

    This is a fascinating topic that honestly cannot be covered in a comment. I just want to offer a couple points:
    1. not all solutions scale. In fact, generally, they do not. Finland runs a very clean and efficient – and free – school system that they’re happy with. But you can pick up the whole of Finland and drop it onto Manhattan, and people would not really notice. The same system will probably not work even for NYC, let alone for the US.

    It’s a bit like the highly efficient management of your Wifi compared to the Internet.

    So just how much Amish’s system, or bits of it, is relevant, is a big question.

    2. one of the biggest – in my system analytic’s lay opinion – reasons for our expensive system is that the users are price insensitive at the point of purchase. Insurance loves that because they get to manage more expense – and add more fees. Provides love that for obvious reasons. Users love that (after having paid the premium and forgotten about it).

    One of the feasible, policy-applicable ways out are the high deductible plans backed up by HSA accounts. Users get to see and feel the prices, while being protected against catastrophic events.

    • teageegeepea says:

      Finland has a population that wikipedia puts at 5,526,774, while Manhattan has 1,628,706.

    • JimJ says:

      High deductible plans don’t really provide health insurance; they provide asset insurance. Most Americans do not have enough money to pay for the deductible, so they either don’t (and *might* pay on installment later), or they don’t get treated. The choice between those two is not very well correlated with how serious the health issue is.

  39. eqdw says:

    Eighth, for the same reason, Amish try not to overspend on health care. I realize this sounds insulting – other Americans aren’t trying? I think this is harsh but true. Lots of Americans get an insurance plan from their employer, and then consume health services in a price-insensitive way, knowing very well that their insurance will pay for it.

    I’m a Canadian in the US, I have had exposure to both medical systems _and_ both cultures surrounding them. This take is 100% true. Americans constantly go to the doctor for things they don’t need to, constantly order tests they don’t need to, and generally use way more medical resources than Canadians do.

    One of the _major_ culture shocks I experienced moving to the US was that I was shocked at how frequently Americans seem to go to the hospital for minor things. Things like (this is a convenient example) the flu. Never in my life had I ever even heard of someone going to the hospital for the flu, unless they were elderly and in need of ER treatment. You just wouldn’t do that, because why would you go spend _six hours_ waiting to see a doctor at a clinic, when you know he’s just going to take your temperature, give you some advil, and tell you to drink water and rest. We have advil, water, and rest at home. Meanwhile the first time I got the flu in the US, everyone was telling me to go to the doctor. The conversation went something like this:

    Me: “I have a fever of 102 and feel like shit”

    Them: “You have the flu! Go to the doctor”

    Me: “Uh… why? There’s no medicine to cure the flu, what’s the point?”

    Them: “Well they’ll test you to confirm that it’s the flu”

    Me: “I already know I’m sick, what does it matter if it’s technically influenza vs something else?”

    Them: “Because then you’ll know it’s not something worse”

    Me: “I already know it’s not something worse because if it was something worse I would feel worse.”

    To this day I still don’t understand what the point of flu testing is, from an individual perspective.

    As a Canadian in predominantly blue-tribe circles, I am constantly called upon to act as an avatar of my people and to condemn Americans for being ignorant redneck conservatives. And, as the obvious point of difference, I am frequently called to comment on health care. The US healthcare system is terrible and way too expensive, isn’t free healthcare so much better? Well, I have a lot of thoughts about this, but one of my biggest thoughts is “maybe it wouldn’t be so expensive if you all didn’t go to the doctor every time you have the sniffles”.

    The really frustrating part is that, it’s as you say, it’s not poor people doing this. It’s mostly richer people doing this. And they aren’t doing it with their own money, they’re doing it with their cushy employer-sponsored healthcare, which they pay very little of their own money for. These are, generally, the same people who I see typically advocating for Canadian-style universal care. They don’t seem to understand that if they got what they were asking for, the quality of care (from their perspective, anyway; I don’t think this care was needed in the first place) would drop dramatically. It would be cheaper, yes, but it would be cheaper because of bureaucratic mechanisms kicking in that either explicitly or implicitly (because you don’t want to sit there in a waiting room for six hours) ration care. The cost savings comes from the rationing, and there’s no reason why we couldn’t just ration right now. Both at the individual level (_just stay home_ when you have the flu) and at the insurer level (insurers can, and frequently do, exercise exactly the same rationing that the government would. Except when insurers do it everyone gets mad, and when the government does it, everyone’s ok with it)

    • Nicholas Weininger says:

      If this is a major cost driver, we’d expect to see post-pandemic US health care costs drop significantly from people keeping up the habit of substituting telemedicine for things they would have gone to the hospital for pre-pandemic, no?

      • albatross11 says:

        This would be consistent with the findings that better health insurance doesn’t seem to improve health outcomes much, if at all.

    • sidereal says:

      I think it’s worth noting that this behavior is rational. I have no incentive to consume healthcare in a price-sensitive way because I’m not paying for it.

      I have some incentive to avoid diagnoses which will place me in a higher risk pool in the future – with the ACA that’s less of an issue, but that could change and as I understand is still an issue with medicare. To your point about flu, I will sometimes ask for a flu-test so I can get started on tamiflu ASAP. That’s at an urgent care though – even if I met my out of pocket max so that the cost was a non-issue, an ER visit takes hours.

      There actually seems to be some effort by the medical establishment to “ration” care, that is most doctors will put up some resistance to unnecessary tests. (I speculate there is some pressure from insurance companies to keep costs low). But, there are sometimes magic words you can use, and persistence, and if that fails just find a more cooperative doctor.

      • albatross11 says:

        I think a common thing is that sometimes the doctor or PA is just rushed and finds that ordering a flu/strep test and then giving you Tamiflu/antibiotics/both as indicated is the path of least resistance.

      • JimJ says:

        Insurance incentivizes extra tests, because then it is easier to put a reimbursable label on what the Doctor visit was for.

    • Garrett says:

      Americans pay for their healthcare in money. Canadians pay for their healthcare in time.

      When I have family visiting from Canada I drive around to the major hospitals and show them the ER waiting rooms. They can’t believe that they are that empty. There’s a lot of patient flow – it’s just being handled quickly.

      • teageegeepea says:

        And I suppose the difference is that insurance makes it someone else’s money, whereas spending your own time gives you “skin in the game” (even if you’re also wasting other people’s time and thus aren’t fully internalizing externalities).

        • Garrett says:

          That’s a good argument. The counter is that a lot of insurance has been instituting much higher co-pays for ER visits.

    • Matt M says:

      Thanks for this post. I think it pretty well maps to my general frustration with the health care debate in the US. I think a lot of the same people who argue for the US to have a system that more approximates the Canadian one would not agree with the premise that “Americans use too much health care.” Generally speaking, I think most people in this debate do not understand the relevant tradeoffs involved.

    • Doug S. says:

      That “flu” could be a bacterial infection you should get antibiotics for. It’s happened to me plenty of times. Or you could confirm that, yes, it’s influenza and go home with a prescription for Tamiflu or one of the other relatively new antiviral medicines. So it’s definitely a good idea to see someone if you’re running a 102 fever, but going to the ER is almost certainly overkill; an appointment with a general practitioner or a trip to an urgent care center is enough and costs everyone a lot less money.

  40. ksdale says:

    My father recently passed away from pancreatic cancer. The final bill for the month that he spent in the hospital for all of the tests and whatnot prior to coming home and dying a few weeks later was something like $300k. He wasn’t the kind of guy to go to the doctor excessively, and I can easily imagine the final hospital bill being at least 90% of his entire lifetime medical spending.

    My intuition, based on this anecdote and nothing else, is that this is not a particularly uncommon situation, where the thing that finally kills someone consumes the majority of their lifetime medical spending.

    If a community had a norm of “God’s will be done” once people reach a certain age or level of frailty, it seems plausible that it could drastically reduce medical costs overall (while simultaneously reducing life expectancy by a few years, perhaps?).

  41. The Amish have had a life expectancy in the low 70s since colonial times, when the rest of us were dying at 40 or 50.

    This sounds pretty hard to believe.

    • Matt M says:

      Agreed. How exactly could one explain that?

      In colonial times, their lifestyles, habits, and religious beliefs weren’t very far off at all from what the average “English” person was up to…

      • teageegeepea says:

        My understanding is that in all of Amish history, there has been a grand total of 1 homicide. That was quite different from the colonial “English”.

        • Le Maistre Chat says:

          Whoa. That sounds like one of those semi-idealized cultures from science fiction.
          “Ape shall never kill ape!”

          • Matt M says:

            Wait… isn’t this the plot of DS9?

          • Nick says:

            @Matt M
            Not quite. The Changelings who run the Dominion were a bunch of narcissistic fascists who subjugated one quadrant of the galaxy and tried to subjugate another. The Amish are pacifists.

          • Matt M says:

            All I’m saying is that if the Amish suffer a severe COVID outbreak, they better do everything they can to locate the only Amish who has ever killed another Amish…

        • matkoniecz says:

          https://en.wikipedia.org/wiki/Edward_Gingerich

          He is the first Amish person to be convicted of homicide.

          Murder happened in 1993. And as an additional note, he was apparently at least partially mentally ill and “somewhat of a rebel to the Amish way of life from an early age”.

          So the first Amish person to be convicted of homicide was unusually not-Amish anyway.

          I am still suspicious about “grand total of 1 homicide”. Is it reasonable to suspect more, not reported, murders?

          • Matt M says:

            I think it’s reasonable to suspect that if there were a murder in Amish country, the Amish would look to handle it themselves internally rather than involve the English authorities, which might be described as a “cover-up” or something like that.

            Edit: More literally, I suspect what the Wikipedia entry is saying is that he was the first Amish person to be convicted of homicide in an English court.

          • matkoniecz says:

            Or maybe Amish are really good at ejecting would-be-murderers from communities? Maybe they have their tolerance to violence so extremely low that they eject all potential murderers and also people that are just more violent than average, but in 99,9% innocent and not dangerous?

            Or their upbringing is simply very successful at ensuring that Amish are not becoming murderers?

          • SamChevre says:

            It’s hard to describe just how completely the Amish world is opposed to violence*. Children are forcefully taught, from the time they are tiny, not to push or hit back. The whole worldview–the hero stories, the way the adults relate, the way children are taught–emphasizes self-control and non-violence to others.

            *Spanking children isn’t considered violent.

    • Lambert says:

      And we weren’t dying at 40 or 50.
      We were dying at either 5 or 70.

    • Akhorahil says:

      This sounds pretty hard to believe.

      It’s likely that they had a life-expectancy comparable to other people at the time who lived in ordered, peaceful, communal, somewhat isolated farming communities. It’s just that a lot of people didn’t.

      And it should probably be read as “they lived to their early seventies in general”, not that it was the life expectancy counting infant mortality.

  42. JoeCool says:

    I’m suspicious about the claim that the minister handing out aid well cause he knows everyone. I suspect it is true for the median Amish person, (so maybe I’m not so suspicious) but for women trapped in abusive marriages, gay people, women who want more rights, people who are bad at conforming to the Amish way, generally, the minister is likely not going to use money on them effectively. They’ll be labeled trouble makers.

    I mean imagine if you were under the power of the minister and completely dependent on him for your healthcare expenditures. A weirdo rationalist like yourself? You’d be screwed.

    The median health may look good, but I bet (and I guess that would be difficult to study, and perhaps I’m mistaken) you’d find subsections of really poor health outcomes. It seems to me Amish religion is likely to be a bit like the happiness version of the trolley problem, would you have the median member of your community much happier at the price of the wierdo’s in your community being much less happy?

    My own home faith community, Mormonism, seems like Amish light in a lot of ways, you have the really good health outcomes, except for the outcasts with really bad health outcomes, but the median is the one thats usually talked about.

    The big thing that is cool about the Amish though, is that you can often vote with your feet and immigrate to various different level of strictness in Amish communities. Mormonism is pretty strict, unless you move to a new geographic location your stuck with your bishop, who has great authority over you.

    I will say its much more difficult to study whats wrong with religion than what it gets right, because whats wrong with religion usually comes in the form of a few good men code red style cloak and dagger stuff, which is inherently difficult to study.

    That is to say, strong social norms on secret anti social activity that you deploy against dissenters and don’t tell anyone about. At least in Mormonism….

    • Matt M says:

      Generally agree with this.

      In closed societies such as this, the “outliers” can easily be silenced, or in extreme cases, removed entirely.

      The US government has no such options.

      The answer to “What about all the chronically ill people I interact with every day? Where are they in Amish society?” is almost certainly “suffering in silence somewhere and you’ll never find out because there is no mechanism for them to ‘go to the press’ or whatever” or “kicked out of it and sent to you and your colleagues to deal with”

      • Alkatyn says:

        It reminds me a bit of those reports that say people in China or another authoritarian country report higher levels of satisfaction in their governments than western countries do. Well of course they do, they know what happens to the people who don’t

    • John Schilling says:

      I agree with this in the general case, but I think the effect is much weaker when it comes to acute medical care (or chronic care where treatments are obviously effective, e.g. insulin for diabetics). There are a lot of crappy people in the world, and most of them are particularly crappy to their outgroup. But we touched on this an OT or two ago with the bit where religious groups were doing charitable medical outreach efforts in response to the pandemic, and the number of people who are such egregious shits that they will deliberately withhold lifesaving medical care where there is a norm of providing such care is pretty small. And it’s tempting to think that our outgroups are all full of that nasty sort of person, but it’s rarely true.

      Partial exception when the medical care is for the obvious consequence of sinful behavior, e.g. snowmobile-riding in the Amish case. But I’d wager that the Amish woman whose arm is broken by her abusive husband, gets about the same standard of care as the Amish woman whose arm is broken in a carriage accident. She may then be ostracized as a disobedient wife, which would be a legitimate and plausible criticism of the Amish community, but she’s not likely to have her arm amputated because the community wouldn’t pay to have the bone set.

      • JoeCool says:

        yeah, I was thinking more psychological holistic style medical care.

        Acute care, I’m sure works.

    • ana53294 says:

      I don’t think they’d treat abused women less than those that get an equivalent injury.

      Even in communities where it may be OK to hit your wife, as she’s expected to obey and whatever, hitting a farm worker hard enough to incapacitate them is not something that will endear a husband to the community. So while the community may ignore bruises or whatever, I expect that they won’t show the same level of understanding for a husband who breaks his wife’s ribs.

      And in most cases, where domestic violence is concerned, the wife says she fell from the stairs/horse/whatever, and everybody pretends to believe her. The husband doesn’t go around saying he hit her, and everybody quietly pretends they don’t know what’s happening, because what happens in somebody else’s family is not your business. So the pastor would pretend to believe the wife and pay her bills if her ribs get broken.

      She won’t be denied medical care; she’ll be denied a divorce and a safe home, which is what would save her.

      • theredsheep says:

        One assumes the Amish don’t have “rough music” or equivalent; I wonder if they have some gentler corrective? Perhaps shunning? Given their intense abhorrence for violence, I suspect that anything as overt as pummeling your wife has a fair chance at reprisal, at least if you aren’t the bishop or something.

      • SamChevre says:

        On thing to note: in general, the church pays for medical care–they don’t approve it beforehand. If someone is hurt and goes to the doctor, they’ll get what treatment the doctor thinks appropriate.

        Shunning is the ultimate penalty, but hitting your wife would definitely get you excommunicated and shunned* if known. (So would child abuse-although the lines of where abuse starts are probably in some cases too allowing of abuse.)

        And don’t underestimate shunning–it’s a mind-breaking penalty. I’ve been beaten until I couldn’t stand, and I’ve been shunned; I wouldn’t even have to think to choose being beaten.

        *There’s a separate problem of how to handle people who have sinned but repented–shunning is a penalty for unrepentant sin. There are theological reasons that make inflicting consequences after someone is genuinely, demonstrably sorry and has been punished difficult in the Anabaptist world.

        • Garrett says:

          > it’s a mind-breaking penalty

          That’s why the threat is effective.

        • tcheasdfjkl says:

          If this isn’t too personal a question, I’m now very curious about what shunning is like – both how it works practically and what makes it particularly bad – and what kinds of things can get one shunned. (Also very curious about the rest of your experience tbh, and reasons for leaving etc.)

  43. rahien.din says:

    This is all very fair.

    If can add, though : the Amish have a very high relative incidence of genetic disease, as a consequence of their intense and isolating social structure.

    • teageegeepea says:

      It’s also known as a “founder” effect of starting from a small population bottleneck.

  44. block_of_nihilism says:

    I agree, I think the elimination of direct price discovery through insurance is a major driver of healthcare costs. If individuals were directly responsible for their healthcare costs prices would necessarily have to be transparent, especially for non-emergent services. In a world without insurance covering these services, providers would only be able to charge what their patients could afford. The challenge is that many areas now have oligopoly/monopoly situations where most of the healthcare services are controlled by a single entity. Insurance has driven this concentration and lack of competition as well, as large entities are better positioned to negotiate prices and navigate the high costs of billing and compliance.

    • jmo says:

      I agree, I think the elimination of direct price discovery through insurance is a major driver of healthcare costs.

      Most vet care is paid for out of pocket and prices are relatively transparent but the growth in spending mirrors that of human healthcare. That undermines your theory.

      • Matt M says:

        It’s an interesting question. My guess is that the growth in veterinary spending largely correlates with improved outcomes in veterinary care (pets are living longer, suffering less), which is seemingly not all that true among humans in the US.

        I don’t have pets but my understanding is that veterinary prices are, in general, not rising nearly as fast as human health care prices are. Increased veterinary spending would be a factor of the increased priority of pet care.

        • jmo says:

          I don’t have pets but my understanding is that veterinary prices are, in general, not rising nearly as fast as human health care prices are.

          You would be wrong. Vet prices are rising even faster.

          But as the paper’s authors document, animal medical spending appears to be increasing at a faster rate than the human equivalent. Data from the federal Consumer Expenditure Survey shows that between 1996 and 2012, our own health care spending surged by almost 50 percent while pet medical spending jumped by 60 percent. During the same period, the percentage of physicians increased by 40 percent while the supply of veterinarians all but doubled.

          https://slate.com/business/2016/09/the-cost-of-pet-health-care-could-be-rising-faster-than-the-cost-of-human-health-care-study-says.html

          • Matt M says:

            That doesn’t address my point. I don’t dispute that veterinary spending is up. But that could be explained by people having more pets, or seeking more complicated pet care than they did previously. Hence, my comment about pet outcomes.

          • jmo says:

            or seeking more complicated pet care than they did previously.

            That’s the same with humans. IIRC increasing healthcare costs aren’t primarily being driven by the cost of the same procedure/drug going up. It’s that we’re doing more.

        • Lambert says:

          Are we talking vets or pet vets?
          I’d not be surprised if most vets in the US worked with livestock.

      • block_of_nihilism says:

        This is a good point. I would guess increasing costs for educating a veterinarian and increased services are driving the rise in cost. Anecdotally, people definitely seem more willing to spend on their pet’s care, especially on life-extending treatments like chemotherapy, surgery, etc.

        Also, it is fair to say that even a “transparent” healthcare pricing structure would not lead to information symmetry given the wide gap between a provider’s level of knowledge and a patient’s. If a doctor tells someone they (or their pet) need expensive treatments/diagnostics, they are probably going to be more willing to go along with it due to lack of knowledge and the fear factor…

        • jmo says:

          I feel that many of those talking about price transparency don’t have an accurate view of how most healthcare dollars are spent. The majority of spending is for things that are episodic and emergent. The average person isn’t spending $5000/year on healthcare. They are spending a few $100 a year for 30 or 40 years and then they need $300k in care in the next six months or they are going to be in a very bad way. Price transparency isn’t all that valuable in that situation.

      • John Schilling says:

        There has been a very noticeable shift in my lifetime from “our dog is very sick; dogs die and I guess it’s time for us to have a new one” to “our dog is very sick and the vet says it will cost $6000 to keep it alive for a while longer, I guess we have to come up with $6000”. If veterinary costs are only mirroring that of human health care, that’s probably a win for veterinary medicine on the cost front.

        • Matt M says:

          Right – this is the kind of thing I was getting at.

          When I was a kid and our dogs got sick, my dad took them out back and shot them.

          20 years later, I had a close personal friend who took a second job to raise $10,000 to pay for surgery for one of her dogs.

          This will produce some pretty big changes in spending that have little to do with the actual cost of any given procedure…

          • jmo says:

            The timing is a little off but it’s the same for humans. They used to call pneumonia “the old mans friend.” Prior to WWII when you got old and sick you died and you did it quickly and cheaply.

          • Garrett says:

            People get mad at me when I point out that children are technically a renewable resource.

  45. jmo says:

    . And yes, they got medical care without health insurance, and avoided the kind of medical bankruptcies we see too frequently today. How did this work so well?

    Before WWII there wasn’t all that much healthcare available. If you had a heart attack, cancer, a stroke, etc. there was nothing to be done. When Ike had his heart attack as president he was prescribed bed rest. That’s all that was available. With the advent of antibiotics, open heart surgery, chemotherapy after WWII there was suddenly an explosion of healthcare available.

    You’ll notice that as soon as chemo, open heart surgery, antibiotics, etc. exploded onto the scene after WWII it was only a few years before Medicare and Medicaid came along.

    • borsch4 says:

      Also, during WWII there were wage controls and so many employers started offering health insurance as a benefit to attract workers. This wasn’t taxed at the time (as the government had more important things to focus on) and when the war ended the benefit of tax free insurance became entrenched. So not only was there an explosion of things to buy but there was a giant subsidy to buy those things.

  46. bisonbisonbison3000 says:

    Thanks for the insightful article. I appreciate your comments on the moral hazards and perverse incentives inherent in US healthcare.

    The greatest misconception regarding the US healthcare system is that it is a “market” system. That is a misnomer. Not all privately-run systems are market systems.

    In the purportedly “market-based” US system, money is taken from people’s paychecks and put into giant pools of money controlled by bureaucrats, and those bureaucrats control the terms by which goods and services (and accompanying payments) are distributed in the sector. Does that sound at all like a market system, or a socialized system?

    Socialized medicine is not a policy proposal in the US. It is the status quo. We have a balkanized, privately-run socialized medicine system, but it is socialized medicine nonetheless.

    By calling it “market-based,” Blue Tribe leaders get the joy of ascribing the problems of the broken system to Red Tribe principles. Red Tribe folks assume that because the system isn’t publicly run, it must be market-based. No Red Tribe leader advocates for an actual market-based system.

    • albatross11 says:

      The US health care system is like some demented version of the socialist calculation debate where we continue to have a profit motive, but eliminate prices and cover every market transaction in lies (fake prices nobody pays, surprise bills the buyer couldn’t know was coming, hidden cross-subsidies, etc.). It’s a wonder it continues to function at all.

  47. George3d6 says:

    To be honest I think that all this talk of healthcare is missing the big elephant in the room, which is the fact that people have replaced their priests with doctors, but the Amish still have priests.

    It seems that often enough people rely on hospitals to fulfill the role of the priesthood, to give certainty during uncertain times.

    For an example, look at the ongoing pandemic, people want to make sure we give the medical system as much time as possible to “develop a solution” and flatten growth to “not fill in the ICUs”, but nobody is looking at or discussing the actual prospect for a solution, or discussing death rates in ICU vs normal care vs home for conditions of the same severity (and it can’t be that much, considering invasive ventilation 1-month death rates are at 9x%).

    This smells of fear, of an inability to cope with the idea that “Yeah, no matter what we do a significant amount of people are going to die”, an inability to cope born partially out of the lack of religion.

    Not to mention dying, which is an activity people undergo in hospitals, because, again, they are unable to cope with the idea of death, they are unable to cope with the idea of death without religion.

    Or look at the finer details, the confession booth is now the psychologist or the psychiatrist, your priest telling you to keep the religious fast is now the nutritionist… etc

    By no means am I saying religion is THE only solution here, I’m a fairly staunch atheist (or at least agnostic), but to me that seems to be the underlying cause here more than anything else.

  48. BBA says:

    What does the life expectancy for devout Christian Scientists look like? They reject all medical care and believe prayer alone can heal all maladies.

    Maybe all healthcare is just a luxury good. Death is inevitable, what’s the point of going to such extremes just to put it off a little longer?

    Or maybe I’ve just been alone in this tiny apartment for too long.

    • John Schilling says:

      What does the life expectancy for devout Christian Scientists look like? They reject all medical care and believe prayer alone can heal all maladies.

      I don’t believe that is the case; Christian Scientists for example do not believe that prayer can heal a compound fracture and are perfectly willing to go to a hospital for that. I’m not sure exactly where they draw that line, but it is possible that they are mostly benefiting from modern medical care for the things that kill people when they are still young and rejecting it mostly for the things that kill people when they are old. That would weaken (but not eliminate) the life-expectancy signal.

      • Garrett says:

        There are a lot of things which are done in medicine which are both expensive and of questionable value. A lot of back and orthopedic surgeries may fall into this category. Same with generalized chemo for late-stage cancer. A lot of the stuff which is cost-effective is either pretty obvious (setting/plating a bone), or pretty cheap (generic antibiotics).

    • teageegeepea says:

      In 1989 it was reported that their women die four years sooner than other women and men die two years sooner:
      https://www.upi.com/Archives/1989/09/21/Christian-Scientists-have-shorter-lives-study-finds/9540622353600/

    • Simon_Jester says:

      I have a friend who lost a close relative (a Christian Scientist) to an extremely easily treatable tumor (trust me, WAY more easily treatable than normal for cancer). The tumor incapacitated them starting in their fifties. Said Christian Scientist then proceeded to be a drain on their family for years before finally expiring after refusing conventional medical treatment until, after several years, the tumor finally grew out of control and killed them.

      If, at any time during a window of several years, this person had chosen to receive simple medical interventions of a relatively straightforward type, they would likely have enjoyed several more years of productive life, significantly improving their own quality of life, and vastly improving the short and long term quality of life of the family members who had to support the Christian Scientist while watching them slowly and for lack of a better term willfully die for lack of basic medical care.

      Medical care is a ‘luxury’ in the sense that a roof is a ‘luxury.’ Man can live without a roof over his head. That doesn’t mean he should try, or that it is on net beneficial to ANYONE to expect him to.

  49. Nikitis says:

    RE: Insurance companies scamming providers: It is incredibly naive to assume that “patients are on the side of the insurances” or that “If occasionally scamming providers meant insurance companies had to pay more money total, then they would stop doing it”.

    First, insurance companies, like any other company, have zero incentive to lower their prices if they lower their costs. They only have to (barely) beat their competitor’s prices to stay competitive. If they can lower their costs even more, that means more profits to themselves.

    Second, an insurance company that always scams can be seen as a defectobot, to oversimplify a bit, and I don’t need to explain to you why this is a more costly strategy in the long term than the Amish cooperate-cooperate. As for why they don’t change if it is that costly, well, why would they? Companies past a certain size are not rational actors. They are eldritch Molochian abominations with no agency of their own. And the incentives are all wrong for any single cog in the machine trying to reverse the trend. Scamming a hospital is a concrete short-term benefit that can easily be used to advanced a career, while the long-term costs of increased vigilance are hidden and not easily attributed to the actions of a single employee.

    • Edward Scizorhands says:

      Insurance companies operate on the cost-plus model. The more expensive health care is, the more money they can earn.

      Making them cost-plus was explicit government policy.

      • Nikitis says:

        I would very much like a source on that. I’m not an American, but I thought the cost-plus model was ONLY for Medicare, while hospitals and private insurers were free to set their prices however they wanted.

        • AKL says:

          Insurance companies must spend 80% – 85% or more of revenue on health care, any less and they must pay a rebate to their members, so their profits are hard capped at 15% – 20% of healthcare spending (roughly).

        • Edward Scizorhands says:

          first hit on “PPACA profits limited”

          https://thehealthcareblog.com/blog/2012/02/04/does-obamacare-limit-profits-for-health-insurance-companies-in-your-state/

          The ACA requires health insurers in the individual and small group market to spend 80 percent of their premiums (after subtracting taxes and regulatory fees) on medical costs. The corresponding figure for large groups is 85 percent.

          Of course, when you frame it as “they must push through 85% of their income on medical costs” it looks nice and populist. But it’s cost-plus: the way for the insurance company to have more money for itself is to drive up the costs of everything underneath them.

          • AKL says:

            Except that insurance companies compete for business too.

            Obviously the market is not frictionless, but no employer or individual will sign up with an insurance company charging 2x its competitors.

            Insurance companies as a whole arguably have an incentive for healthcare costs to increase, but that’s not functionally different than saying that sellers in ANY market have that incentive, and there’s a reason we have antitrust laws.

          • acymetric says:

            Driving up health costs generally impacts all insurance companies (approximately) the same. There’s no competitive disadvantage there.

          • Ouroborobot says:

            Respectfully, as someone who has worked for multiple insurance companies, the idea that any insurance company would intentionally drive up costs in order increase the future hypothetical maximum absolute value of their share of the pie sounds insane to me. My company is very much concerned with limiting med expense to lower their MLR, and spends a great amount of effort to do so, much of which is directly beneficial to the insured. Even at a large for-profit insurer, they would have to bank on being able to capture a correspondingly larger amount of premium, which most certainly not guaranteed.

          • TheObeseDog says:

            There is no mechanism to drive up costs intentionally in a market like insurance. Prices are so competitive that any company trying such a strategy would quickly go bankrupt.

          • JimJ says:

            I’m surprised at how many responses to this missed the point. No, an insurance company doesn’t want to drive up (just) its own costs … but it does want to drive up the costs of medical care that are paid by everyone, including its competitors. That is different than normal sellers in a normal market (which is why you mentioned cost-plus). Normal sellers want to cut their costs, even if their competitors get the same cost reduction, because lower costs that get passed on in lower prices at least lead to a larger market. With cost-plus, driving up costs for everyone (perhaps by inflicting administrative costs) actually increases the market.

  50. Epistemic_Ian says:

    Above, I said that the most important factor is that the Amish comparison shop. Everyone needs to use other people’s money to afford expensive procedures. But for the Amish, those other people are their fellow church members and they feel an obligation to spend it wisely. For the English, the “other people” are faceless insurance companies, and we treat people who don’t extract as much money as possible from them as insufficiently savvy. But there’s no easy way to solve this in an atomized system. If you don’t have a set of thirty close friends you can turn to for financial help, then the only institutions with enough coordination power to make risk pooling work are companies and the government. And they have no way of keeping you honest except the with byzantine rules about “prior authorizations” and “preferred alternatives” we’ve become all too familiar with.

    There is a third way—Robin Hanson’s proposal to buy health, not health care (see also here). Instead of mediating the conflict of interest between patients and insurance companies with a complex set of government regulations and terms in the insurance contract itself—rules which are difficult for the customer to understand and often create suboptimal outcomes—Hanson’s plan aligns the interests of insurance companies with patients so that no such rules are necessary. In short, insurance companies have no legal or contractual obligation to pay for any healthcare, but they do have a financial incentive for their customers to be healthy. This has a number of advantages over the present insurance system in the US:

    – Insurance is cheaper because it won’t pay for procedures which aren’t worth the cost—the customer decides what tradeoff they want to make between their health and their insurance premiums, and the insurance company will only pay for procedures which are worth it given that
    – Insurance won’t nickel-and-dime customers or try to get out of paying for procedures which are worth it, because their financial incentive is for their customers to be healthy, not simply to pay as little as possible
    – Right now, insurance companies pay for what they have to pay for—they have no reason to care about what does and doesn’t work. Hanson’s proposal mobilizes capital to figure out what what actually makes people healthy. If you’ve ever wished there were more research into something, if you’ve ever wished it were easier to tell what treatments would actually work, if you’ve ever wished that babies didn’t die because the system didn’t care enough to give them the right nutrient fluid—imagine what would happen if the same amount of effort and intelligence and research that is currently applied to predicting asset prices were applied to making people healthier.

    The current system is not inevitable.

    • rahien.din says:

      Robin Hanson is an idiot. His proposal is isomorphic to what we already have.

      Health insurance companies could be LMO’s already. Every person is a collection of “lives.” Each “life” is the absence of a certain preventable disease process. For instance, if your heart is healthy, your “healthy heart life” is alive, but if you get heart disease, your “healthy heart life” dies. The insurance company must spend more money on you now than they had to spend on you before. Hanson says as much when he describes multiple types of insurance (life, health, pain, disfigurement). Implied therein is an insurance based on organ or disease state, and in fact we already have that kind of insurance.

      So who is the beneficiary, should your “healthy heart life” die? An MDC would basically not show up in town unless you died, but consider what would happen if a MDC was not paid their benefits upon a client’s death. They would cause problems for the insurance company, for the client’s family, for everyone involved, until they got their payment. Of course, we have this already – the symptoms of the disease itself, and the adverse effects of treatment! In order not to have the symptoms, you have to expend some kind of money and/or effort – you are literally paying your symptoms a variable annuity.

      People in general are willing to let their health get away from them, rather than follow sound advice. This just means “I will buy the cheapest LMO so that they won’t interfere with my life, but in order to get that LMO I have to put up with a kind of nasty MDC.”

    • rahien.din says:

      Here’s another way Robin Hanson is an idiot.

      His preferred use case is “contract with an LMO and faithfully follow their health advice.” This seems fair – it’s the best way to forestall any preventable disease that could lead to disability, disfigurement, pain, or death.

      But the people who will seek and follow sound advice are kind of rare. The reason why people are unhealthy in the first place is often because they want to feel and seem healthy more than they want to actually be healthy.

      The rare health-driven person will independently seek information and follow it to their benefit. They don’t need the LMO’s advice, so, they don’t really need what the LMO can provide. The only way it benefits them is if the LMO restricts access to important information – this is called paternalism, and it is a bad thing, and we don’t do it anymore. These people shouldn’t buy Hansonian health insurance. So the LMO is stuck with all the people that won’t follow sound advice, and die sooner. This is called “adverse selection.”

      His method would work in a perfect world, but in a perfect world it would be unnecessary.

      • Andrew Hunter says:

        I’m pretty sure the idea here is you let LMOs select on conscientiousness. If I sign a pledge to never drink or smoke, I get far cheaper rates…and if they catch me out at a bar, my insurance is voided. If I’m not willing to sign that pledge (and get tested, etc) I have to go with Jim Bob’s LMO with higher rates.

        Obviously it’s harder to verify things like diet and exercise, but the idea is sound in principle. (Also, I get you don’t like Robin Hanson, but he’s a well-known and careful economist. Do you think he’s never heard of adverse selection?)

        • rahien.din says:

          Hanson is certainly a well-known and careful economist. He also mentions being a puzzled economist – how is it that there is no interest in this idea? Presumably not even from other well-known and careful economists? Probably because he isn’t grappling with the real issues.

          If people would follow health advice and/or if insurers could feasibly select on conscientiousness… we would live in that world already because it would be a nicer world for everyone. Everyone would be healthier, health care would be cheaper, and almost any payor system would work. Not even our jacked-up insurance system would prove ruinous. If people won’t follow health advice and/or insurers cannot feasibly select on conscientiousness, Hanson’s LMOs die by adverse selection. Unfortunately the latter case more closely resembles the world we live in.

          Hanson’s proposal is all fine and good and maybe would work. But the only world in which it works is the world in which any system works.

          it’s hard to verify things like diet and exercise

          See, this is the problem exactly – you are taking entirely the wrong perspective. Think like a business. The question isn’t “How can we possibly verify our clients’ exercise and diet?” That would be too hard. Instead, the question is “How will we restrict our clients to easily-verifiable exercise and diet?” That would make your clients do the work for you, and if they fail at that task, you recoup your loss via financial penalties.

          This brings us to another question. CMIIW but I believe you’re a libertarian – how is Hanson’s proposal even palatable to you? It would require insurers to examine your daily activities in order to score your conscientiousness, and to influence/coerce you toward the way of life they would prefer you to have. Whether you think that does or does not resemble our current situation, the proposal seems antithetical to libertarianism.

          • Edward Scizorhands says:

            The standard libertarian response to “how could you possibly let an insurance company check on your health???” is that “you are free to do business with that insurance company or not.”

            This is, like, libertarianism 101.

          • Andrew Hunter says:

            I am not a libertarian, except insofaras that I’m skeptical of a lot of particular government inverventions. I would describe my political philosophy as Singaporean, though I guess Tyler Cowen has started calling this “state capacity libertarianism”–I want efficiently thought through and economically well-designed authoritarian states that don’t do all that much but do what they do do crushingly well. (I’m also the card-carrying founder and only member of the Let’s Do The Things That Every Single Credible Economist Agrees Are Obviously Free Wins First And Think About Literally Anything Else After, Because If You Can’t Manage That Why Do I Trust You To Do Anything Right Or Well Party.)

            And I’m not sure how any of your comment opposes Hanson’s plan: yes, this only works if you let insurers select on conscientiousness and other rules. Letting him do that is what he is proposing. Casting it as life insurance is a fairly minor tweak on incentives (though I think a good one.)

            By the way, the idea of insurers or other financial agents with an interest in my life inspecting my behavior is neither new nor particularly surprising. They routinely do. I got a much better rate on my mortgage because I took a bunch of actions that banks watched that increased my credit score. Above and beyond that, in fact, I showed one particular banker, who specialized in loaning to tech industry workers, a bunch of details of my Google compensation package, and he quoted me an absurdly good rate, because he’d taken the time to realize what GSUs mean. (Seriously, most bankers are bad at understanding tech comp and quote too-high rates as a result. There’s literally a single guy at who has made an enormous book of business via word of mouth by being better than that.)

            If I wanted life insurance (I don’t), I’d get better rates if I let them do an extensive exam. Many car insurers will let you put a device in the car that measures how fast you drive in exchange for a discount; I have never been offered such a device, but I’d happily take one.

            I wouldn’t accept a discount on my health insurance in exchange for not eating animal products, because I disagree with that insurer’s investment thesis (and also value eating animal products at more than the discount.) But if an insurer gave me a bonus in exchange for coming once a year and demonstrating that I scored X on the army PFT or something, I’d probably jump on it. As Edward points out, this is perfectly libertarian if you care about that (which I don’t): everyone trades with who they find it +EV to do so.

          • rahien.din says:

            ES,

            Then I guess the standard libertarian response is dumb.

            Libertarianism’s M.O. is opposing coercion. If libertarianism 101 includes the idea that a choice between the awful and the catastrophic is still a meaningful expression of liberty, then libertarianism is not even an ideology. At best it’s a slogan.

            Liberty has to be more than performance art.

            AH,

            My bad, I mischaracterized you. Thanks for your explanation.

            yes, this only works if you let insurers select on conscientiousness and other rules. Letting him do that is what he is proposing.

            But as you described they are already doing that.

            And what is more important is that people aren’t following health advice.

            To the extent that Hanson’s plan is “give people health advice and let insurers select on conscientiousness, with some fancy accounting,” hey guess what we are already doing that.

            Edit: to de-ninja myself

          • Edward Scizorhands says:

            If you are completely unaware of a given school of thought, you should probably stop saying what it does and doesn’t believe in.

          • rahien.din says:

            Being forced to to choose between the awful and the catastrophic equals coercion. Being forced to choose between ceding one’s liberty and the catastrophic seems the exact form of coercion that a libertarian would find the most noxious.

            Your answer seems to be “it’s still a choice, therefore everything is okay.” But every human action is a choice! By that standard, we already live in a perfectly libertarian world.

            I get that I have stated things provocatively. And that there is sometimes tension between intuition and practicality.

            But come on. We are always offering each other choices. Coercion is a particular subset of choice-offerings that encroach upon liberty. “Your wallet, or your wallet and you life” is coercion even though the victim has a choice. So if libertarianism is about anything, it must be about what choices I am permitted to offer to others.

  51. emdash says:

    The ideas:

    If occasionally scamming providers meant insurance companies had to pay more money total, then they would stop doing it.

    and

    Someone invents health insurance, which seems like a good idea. But this creates a series of perverse incentives, which other actors figure out how to exploit

    Seem to conflict with one another.

    If the incentive structure is broken, it might not send the corrective signal assumed in the first quote. It’s not at all clear to me that insurance companies are (mainly) incentivized to keep prices low. Companies are competing with one another for subscribers, which means it doesn’t matter to them AT ALL if the price for some surgery goes up 10x as long as that increase applies to everyone. But if I can get my hot shot lawyers to come up with some scheme (scam) where we are only paying 5x more on that surgery while everyone else pays 10x then I’m actually winning comparatively.

    Bonus points if my schemes lead to further price increases (because hospitals spend more on administration), because then my lawyer-investment strategy is getting more opportunity to gain more of an advantage over my competitors. Theoretically this could eventually backfire if my firm’s reputation got bad enough, but a lot of health insurance companies seem to survive despite atrocious reputations.

  52. vaniver says:

    Second, I don’t think this is actually how our grandparents did things. I asked my literal grandmother, a 95 year old former nurse, how health care worked in her day. She said it just wasn’t a problem. Hospitals were supported by wealthy philanthropists and religious organizations. Poor people got treated for free. Middle class people paid as much as they could afford, which was often the whole bill, because bills were cheap. Rich people paid extra for fancy hospital suites and helped subsidize everyone else. Although most people went to church or synagogue, there wasn’t the same kind of Amish-style risk pooling.

    This makes me think that the Amish method, even though it works, isn’t the method that worked for past generations. It’s an innovation intended to cover for health care prices being higher than anything that traditional societies had to deal with.

    I have heard it said that in our grandparent’s time, the largest expense a hospital had was laundry. Today, we have huge magnets to see inside you, and computer-guided particle beams to knock out your tumors, and so on, and an incentive system that rewards people for coming up with more expensive treatments with superior success rates while ignoring cheaper treatments with slightly lower success rates.

    [Siderea’s series of posts on this seems like a solid hypothesis here.]

  53. SamChevre says:

    The Amish-Mennonite world I grew up in has a very similar system of mutual aid, and my brother is a deacon, so is one of the key people who handles financial issues (including healthcare) for his community, and co-ordinates with other communities so has a sense of what’s typcial. So I called him and got rough numbers. My sister (not Mennonite, but lives in the area) is a nurse, and provides some informal healthcare, so I have some insights from her as well.

    My brother’s estimate of costs, for a church with 25 families, 150 people, of whom 100 are under 18–about $300,000 a year would be typical; about $100,000 paid out of pocket, about $200,000 paid by the church (about $2000 per person per year in total). Rough guess at total income would be about $2 – $3 million, so about 10% of income is going to health care.

    There are a few additional points I’d note:
    1) Adults who never joined the church aren’t part of the insurance pool, and often (depends on the community) are allowed to join Medicaid even as children if they will obviously never be metally capable adults: thus, severe congenital defects still are covered by Medicaid.
    2) End of life care makes a big difference in costs. Almost no one gets aggressive chemotherapy, for example, and a lot of nursing care is provided by family members.
    3) One place where the increasing costs are very visible is in childbirth-related costs: an uncomplicated childbirth costs about 10 times (inflation-adjusted) what it did 40 years ago. There’s a strong shift toward homebirth, and birthing centers that are only semi-medical, driven largely by costs.
    4) Paying out of pocket means people are cognizant of costs. One thing my sister is frequently asked to do is advise people on whether they should go to the doctor, and if so, is it an emergency. Imagine that everyone asked a nurse “do I need to go see a doctor” before going to urgent care or the ER–it would save a significant amount of money.
    5) There are a fair number of providers who specialize in treating Plain people inexpensively, and the networks of the Plain world make them discoverable. This pushes the average prices down a lot.
    6) On the “increasing cost” spectrum, a lot more Amish/Mennonites work at jobs where serious on-the-job injuries are common – farming, construction, etc.

    ETA: if people have quesitons, I’ll answer as best I can–and I can ask my brother more questions too.

    • John Schilling says:

      Imagine that everyone asked a nurse “do I need to go see a doctor” before going to urgent care or the ER–it would save a significant amount of money.

      Note that you also have to imagine that the nurse is confident she won’t be sued for malpractice or stripped of her credentials by the state if one of the “stay home, it’s probably just the flu” patients turns out to die of Not Just The Flu. That’s another advantage the Amish can call on that would be hard for the English to duplicate.

      • johan_larson says:

        Here in the province of Ontario we have a 24-hour help line staffed by RNs who provide basic advice and try to direct callers to appropriate care. I don’t know how forthright they can be; I’ve never called.

        https://www.ontario.ca/page/get-medical-advice-telehealth-ontario

        • Randy M says:

          My insurance has advertised the same (I don’t recall if current or previous, probably both). I haven’t used, but have a bias against using hospitals anyway.

          • Garrett says:

            My insurance does as well, but they can never answer my question and pretty much default to “we recommend you be seen in the ER within a few hours”.

          • etheric42 says:

            Gerrett’s experience is exactly mine. The insurance or generic nurse help lines always recommend the hospital. The nurse help line associated with the local mega-practice either recommends the hospital or coming into the practice (which is open well past normal business hours).

            When calling the doctors directly for advice, they usually downplay things it seems (or maybe just compared to the nurses).

            Telehealth so far has always either said come into the clinic for testing (flu) or what Rx do you want me to call in.

          • johan_larson says:

            Who is running the help line may matter. The one I described is run by the provincial government, which is the party that gets the bills for our health care spending. As I understand it, this service was set up specifically to lower the load on the ERs. So to the extent they have an agenda, it’s to get you to not go to the ER, but rather take care of the problem yourself or go to a regular doctor or nurse practitioner instead.

            I suppose it’s still possible that the help-line is ineffective. It could be poorly run, or doctors as a group could could have manipulated training or practice standards so the RNs staffing it are unable to offer effective advice. I’m not sure. But the incentives of the party paying for it — the provincial government — point in the direction of getting you to accept cheaper rather than more expensive health care for whatever issue you bring to them.

          • Garrett says:

            My insurance certainly wouldn’t want me to go to the ER unnecessarily because they’d be paying a significant part of that cost. But everything falls into regulations, liability, etc. If the nurse follows a script overseen by a doctor, the nurse isn’t on the hook as long as all unknowns point to prompt medical attention. Obviously, insurance companies have the extra resources to set up complex decision trees for this sort of thing. Small practices probably don’t so they always recommend that someone come in.
            If you get a doctor on the phone, they have the professional authority (and liability if wrong) to tell you to stay home or whatever. Add in incentives to not have to deal with another unscheduled patient…

      • sconzey says:

        In the UK we have something similar to the system johan_larson outlines below called NHS 111 (so-called because you reach the service by dialling ‘111’ on a phone). It’s supposed to be the non-emergency equivalent of 999 and give guidance before you head to the ER.

        Within 111 you can be escalated to a nurse or a duty doctor to do a phone-consultation and make a medical decision. If you do get referred to ER the notes from the call go with you and you get to skip some of the ER triage steps.

        With that said, they tend to be extremely trigger-happy about sending people to ER! I’ve been sent to ER for a small cut on my finger by 111 before (to be fair it was deep, but not that deep!), and I’ve been sent over multiple times for various kids’ fevers. Most of the time I end up on the end of a rant by the duty doctor or ER triage nurse about how trigger happy NHS 111 is. On the most recent occasion, the duty paediatrician advised me to get a pulse oximeter and walked me through how to triage a childhood fever myself rather than calling 111.

        That base-level of trust between strangers is an extremely valuable social technology. Huge respect for cultures and societies that can maintain it.

    • Scott Alexander says:

      Thanks. This is amazing to have this kind of access to a primary source, and I’m glad that the real number ($2000 per person) isn’t too far off my estimate. I’ll definitely highlight this comment on the next open thread.

  54. AJD says:

    In essence it seems like a combination of transparency, civic mindedness, good living and social accountability are mostly responsible for the Amish advantage. Not a bad lesson for us English.

    WRT pricing, it seems like the lie of “non-profit” hospital status, the insertion of the profiteering middleman and various artifacts of “pure capitalism” (pharma pricing, patent shenanigans, lack of regulatory vigilance, frivolous and/or exorbitant malpractice suits, inapt distribution of medical specialists because of skewed incentives) all work together to keep pricing high and escalating. In all, a tightly coupled and complex system… it feels very insoluble without wholesale reinvention.

    Now I feel encroaching despair. Better call a shrink! 😉

    • Scott Alexander says:

      Any links on what makes nonprofit hospitals a lie?

      • AJD says:

        https://www.statnews.com/2017/11/27/hospitals-tax-exempt-status/

        https://www.theepochtimes.com/study-finds-huge-revenue-asset-growth-for-non-profit-hospitals_2978214.html

        https://www.commonfund.org/news-research/blog/nonprofit-hospital-ethical-consideration/

        In most parts of the country the mergers and buyouts of local hospitals create monopolistic systems. Plus, they are all opening deceptively titled standalone “emergency rooms”.

        In one county in my local region I sit on the regional EDC. The major local hospital system has been opening these emergency rooms. But, they lack much of the standard equipment that a hospital equivalent has. That engenders a bizarre reality in which patients are brought to the standalone center by ambulance, and then must be ambulanced to the hospital because they need an MRI or catscan, or trauma treatment. The billing codes are hospital codes, but the offering is urgent care grade. Patients are then billed both for hospitalization and multiple ambulance trips. These are giant hospital systems not outlying independents.

        All this takes place while claiming to be nonprofit. This accounts not just for price inequities, but also for vast tracts of rural areas without hospitals. I don’t know how else to think of this product “diversification” and multiplication of revenue streams as other than as profiteering.

        • Don’t forget certificate of need laws.

          • Telomerase says:

            Oh yeah, “certificate of need”… the NH hospital board decided the state didn’t “need” a specialized cancer hospital in Manchester, so we don’t have one.

            Every week some poor sap flies to Houston for his cancer treatments…

      • Edward Scizorhands says:

        If I start a for-profit and am 100% the sole owner, all surpluses go to me as returns to the shareholder.

        If I start a non-profit doing the same thing, all surpluses go to me as wages.

        • Doug S. says:

          In theory the IRS should have something to say about that. In practice who knows?

          • Wency says:

            I think the IRS is there to make sure you filed the paperwork correctly and aren’t paying yourself above a “market wage” (with some freedom to define and justify that). Though that might also fall to the board or other parties, but it seems to me non-profit CEOs, while certainly well-compensated, aren’t paid wages that are unfathomable in the private sector.

            Most small companies aren’t highly profitable once the value of labor is subtracted, so paying 100% of profits to the employee-owner(s) as wages falls within the realm of plausibility. This isn’t necessarily true at large companies, so I think large non-profits just have to allocate “profits” to administrative bloat, perhaps creating a network of favors to the CEO, or employing the founder’s good-for-nothing children and nephews, etc.

          • Edward Scizorhands says:

            What do you think is tax fraud? You don’t have any profits to be taxed because you make sure that payroll grows just enough to make sure there is nothing left. You hire up friends and relatives who don’t have to show up to work. Who is going to check?

            I’m not talking about pretending to be a charity with tax-deductible contributions (although those are certainly vulnerable to the “boss hires every family member to do no work” problem). And I may have fucked up the distinction between non-profit and not-for-profit, which is my bad.

            You can’t scale well with this scheme, but if you are offering something plausibly community-serving, you can ride on a lot of freewill of customers who think you are on their side. About 20 years ago there was a scourge of “credit counseling services” that never turned a profit.

      • Randy M says:

        I think he is referring to the connotations of the staff making very little, when in reality the term means no share-holders profit, but doctors and admin may still be collecting handsomely.

  55. trevor11 says:

    > Second, I don’t think this is actually how our grandparents did things. … there wasn’t the same kind of Amish-style risk pooling.
    >
    > This makes me think that the Amish method, even though it works, isn’t the method that worked for past generations. It’s an innovation intended to cover for health care prices being higher than anything that traditional societies had to deal with.

    I think actually that something like the Amish method was exactly what a large fraction of our great-grandparents (and older) did for inexpensive health care. Have you looked at all into the “mutual-aid society” (i.e. “fraternal lodge”) health-care systems of the past?

    https://mises.org/library/welfare-welfare-state

    The author of that article (unsurprisingly for a Mises Institute author) argues that this affordable system of healthcare was gradually killed off by a series of government regulations a couple of generations ago.

  56. sarr says:

    A couple of little things:

    The Amish: avoid the highest sources of injuries. Cars.
    Our ancestors: avoided many expensive forms of care because they didn’t exist yet. Eg cancer treatments.

    • RC2020 says:

      While they don’t own or drive cars, Amish are still vulnerable to them. Motorists can underestimate the high speed differential between their own vehicles and a horse-drawn buggy, which has a top speed of around 8 mph. This danger can be increased if a car is cresting a hill and the driver has only seconds to react to a previously unseen buggy on the downward side, or when driving at night, where a buggy’s distinctive shape may not be visible, and its rear lighting may be mistaken for that of a faster-moving car or van.

  57. John Schilling says:

    Everyone needs to use other people’s money to afford expensive procedures.

    Not everybody; some people don’t need expensive procedures at all. Or “need” them only to secure an extra few months of misery in a hospital bed.

    And over the long term, only the people whose needs are significantly above average and/or wealth significantly below average need to use other people’s money. Since the needs are highly variable and unpredictable, it’s good to have a way to pool and transfer the money when the need arises – but this comes with the moral hazard that people will be incentivized to exaggerate their needs and hide their means when the decisions are being made. Having a high-trust society is a big help here.

    • sidereal says:

      Once people have paid their premiums (and perhaps, deductible), their marginal healthcare consumption isn’t out of pocket – so isn’t it effectively using other peoples money? It’s like an all you can eat buffet, why would you ever trust someone will eat as if they were paying per pound?

      • jmo says:

        so isn’t it effectively using other peoples money?

        If you’ve been paying $6000 a year for 20 years and suddenly have a heart attack and it costs $100k to treat. Is that “other people’s money” or are you just cashing in on premiums you’ve paid?

  58. Scoop says:

    I think the entire mystery is solved by lifestyle. People have a huge tendency to overrate the importance of healthcare to longevity and underestimate everything else. Excepting antibiotics, vaccines and emergency surgeries surrounding childbirth and severe injuries (car crashes, gunshot wounds), healthcare is a pretty small contributor to lifespan and even less of a contributor to healthspan.

    The biggest factors by far are
    1. Diet
    2. Exercise
    3. No smoking and minimal alcohol and/or drug use

    After that
    4. Sleep
    5. Good social network

    If we were all on-point on those five things (or even just the first three) but had no healthcare beyond vaccines, etc., we’d live longer and better than we do now.

    If you look at the top ten causes of death in the US — heart disease, cancer, accidents, chronic lower respirator disease, stoke, Alzheimer’s, diabetes, and influenza pneumonia, kidney failure, suicide — good lifestyle reduces the incidence of most of them by well over 50%.

    It also eliminates the need for a trio of other medical interventions that do extend the lives of people who choose to abuse themselves: statins, blood pressure medication and diabetes treatment.

    • Randy M says:

      A lot of that number 1–diet–is in contention. Does dietary saturated fat contribute to heart disease? To obesity? What about transfats? How much protein is optimal? Does the source matter? What level of salt?

      Though of course a lot of what is eaten and especially drunk here is obviously harmful, I’m not confident of getting good advice to prevent heart disease, stroke, cancer, or obesity by diet.

      • Scoop says:

        People way overstate dietary uncertainty. There may be some arguing at the margins as to which diet is optimal, but it’s pretty easy to put together something that’s better than what 90% of the country eats.

        I’d say the easiest is the modified Michael Pollen: Eat unprocessed food, including plenty of veggies.

        • SteveC says:

          Indeed, while vegans, paleos, and carnivores might disagree on what on what is “right” to eat, they agree a lot on what constitutes “crap”: highly-processed, unnatural “food” made from Big Ag products.

          • Scoop says:

            Yes, you can get do pretty much any diet from vegan to paleo with the seven-word rule I put up.

            You can’t do carnivore, obviously, but I’ve still seen plausible defenses of carnivore.

            You can get the micro-nutrients others get from veggies by eating eggs and organ meat. Muscle meat that we humans typically stick to has plenty of protein and fat but few micro-nutrients. Organ meat, on the other hand, has a nutrient count that makes broccoli look like junk food.

            Presumably that’s why carnivorous animals all go for the innards first thing after the kill.

          • SteveC says:

            About that muscle muscle meat, searching on carnivore+deficiencies produces interesting results, and two main pieces of advice:
            -eat the products of grass-fed / free-range animals
            -organ meat, especially liver
            The first might be expensive, but the latter is incredibly cheap. I was appalled to learn that cod livers (at least from Pacific cod) are mostly discarded. So strange that the “quinto quarto” nowadays is dog food. I’ll bet the Amish still eat a lot of Leberwurst and Blutwurst.

    • 205guy says:

      I think stress is a major debilitating factor in modern life.

    • SteveC says:

      Good list, but I’d tweak it:
      Social network should go on top, and the Amish do have closer social ties (as did all Americans in times past). Skip to about 7:00 here:
      https://www.ted.com/talks/susan_pinker_the_secret_to_living_longer_may_be_your_social_life

      Amish men do smoke.
      I find their diet suboptimal.

  59. 10240 says:

    What is actually the legal situation regarding waiving your rights to a malpractice lawsuit? If you sign a contract with your doctor that you understand that not doing this test/treatment implies a somewhat higher risk of certain complications than doing it, and you agree not to sue for malpractice if it happens, is such a contract not enforceable?

    • Scott Alexander says:

      I think it’s not enforceable by law, but Amish people seem generally trustworthy so doctors believe them.

      • Matt M says:

        Yeah… any scenario in which you’re forcing someone to sign a piece of paper not to sue you is one in which you clearly don’t consider the other party to be generally trustworthy, and the piece of paper they’re signing is literally of zero value unless you’re absolutely certain it would hold up in court (which, as Scott suggests, is probably not the case in scenarios like this)

  60. Swami says:

    I have some real questions where this came from…

    “This separates them from insurance companies, who are constantly trying to scam providers however they can. Much of the increase in health care costs is “administrative expenses”, and much of these administrative expenses is hiring an army of lawyers, clerks, and billing professionals to thwart insurance companies’ attempts to cheat their way out of paying.”

    “Scam” and “cheat” are two pretty powerful words which in this case are being applied to the industry as a whole (not some rare bad apple). I understand none of us like insurance companies and certainly we can find an occasional bad apple, but this seems to require a bit more documentation. I would almost go so far as to say it is self contradictory.

    Giant insurance companies are deep pockets, and if they are routinely or “constantly” cheating and scamming then they can easily be sued or closed down for doing so, thus the army of lawyers and administrators would be requiring they play by the rules.

    I really wish Scott would provide evidence for this. It comes across almost like sour grapes in his past dealings.

    • Scott Alexander says:

      I’m mostly hearing this from talking to other doctors and people in the medical field – my company has billing staff who insulate me from it.

      I agree it would be surprising if insurances literally scammed people in illegal ways often. I can only report that every doctor and billing specialist I have talked to says this happens to them and that you have to be prepared for it.

      Part of the dynamic is that in areas with only one or two insurers, those insurers are monopsonies. If you are a doctor there, you either take that insurance, or you have no patients. That means insurance companies can make you sign contracts saying whatever, and some of those terms might be that insurances can refuse to pay for any reason.

      My own interactions with insurance companies have been that their actions probably make sense to them and come from some kind of consistent logic, but that nobody else understands it, and if they make a mistake, then anybody who can correct it is behind ten layers of phone trees that you will never be able to navigate, so it’s easier to work around it than get it dealt with. I expect a lot of billing complaints come from this.

      • Swami says:

        Yeah, I agree completely with your longer explanation, Scott. I think a better term for it is “BS insurance company fineprint and legalese” or something.

        I bet if we had the insurance claims adjusters give their side of the story they would say something similar about being scammed by the hospitals and doctors and customers.

        I am not defending either side, just pointing out that the nature of the relationship is extremely adversarial.

    • etheric42 says:

      During one of the pregnancies we tried to get our employer-based insurance to cover home birth with a midwife, or at least some of it. The midwife employed a medical billing/protest firm (seemed like some kind of law firm from their letterhead and initials) which we paid for. They went to the insurance company and made the case that midwife care and home birth was a covered procedure for the insurance company, but there were no in-network providers in our area or surrounding areas. The insurance company responded with a letter saying they would make an exception and treat this as in-network.

      Then after the birth the insurance company valued the “usual and customary” payment for the midwife at close enough to zero that with a little additional accounting magic they paid less than what we paid for the protest (even though we were supposedly on a zero-deductible plan). The midwife went back to the protest firm and the lawyers basically said the insurance company was enough within their rights to do so that it wasn’t worth the fight (and then billed her for the additional research).

    • matkoniecz says:

      Many types of scamming and cheating are perfectly legal.

      Or litigation costs are higher than recovery. For example, bank failed to notify my properly about new payment and took some of money. AFAIK this was illegal. But I am not going to sue bank for this amount of money.

  61. deluks917 says:

    I thought the rationalist consensus was settled at least ten years ago? Expensive healthcare isn’t that useful. The most useful health interventions are usually extremely cheap (especially vaccines). There are tons of caveats to the preceding statements but it holds in general. Robin Hanson wrote about this in 2007: https://www.cato-unbound.org/2007/09/10/robin-hanson/cut-medicine-half . The evidence is extremely compelling.

    • Scott Alexander says:

      Since then many papers have come out that contradict the RAND study, though I’m not going to search for all of them at the moment.

      • Telomerase says:

        Expensive (or at least high-tech) health care would certainly be “useful”… Bowhead Whales live over 230 years, and their genetics are still trapped in evolution’s programmed-death game.

        The problem is that you can’t sell most anything that would be “socially disruptive” under the FDA system. There aren’t even drug approval categories for anti-agathics or nootropics. I was fixing Duchenne’s MD with dystrophin gene vectors back in 2003 or so, and had already been immortalizing mammal cells (including Bowhead, yes, I personally SAVED THE WHALES [cells]) for years before that.

        But you can’t use anything from the labs in the clinics.

        http://www.strike-the-root.com/51/walker/walker6.html

    • Swami says:

      Yeah, that and the idea that nobody spends other people’s money efficiently. Insurance is not an efficient process to pay discretionary expenses. During the Obamacare debates, most insurance savvy people suggested that insurance needed to focus on catastrophic and routine checkups, with the discretionary and routine costs below that funded out of people’s own pocket.

      • Telomerase says:

        True. Insurance is for outliers that you can’t afford. It works fairly well for that, and terribly for small things.

        • North49 says:

          In fact, “insurance” to cover routine annual exams, or prescription drugs, or any number of things currently covered by “insurance” for which the cost is known and agreed to in advance can’t rightly be called “insurance” in any meaningful sense. It’s just pre-paid for.

          Netflix isn’t “movie insurance”, even though in effect you’re paying a monthly premium because you might need to watch a movie some time. A gym membership isn’t “exercise insurance”, etc, etc…. Most of what people do with their medical “insurance” is really a medical pre-pay plan. Perhaps if we figure out why gym memberships are so expensive we could gain some insight into some portion of the medical costs.

          • albatross11 says:

            Right. The actual insurance is called something like a “high deductable plan.”

          • etheric42 says:

            I thought the theory was that the routine exams lowered the overall cost of the catastrophic thing later, so even catastrophic insurance had financial reason to cover them. And even not-preventative-but-not-catastrophic care helped prevent the worsening of conditions to prevent the catastrophic care. Better an FP visit than sepsis and a hospital visit.

            Of course the math behind that is probably contested, but I thought there was at least an argument for it.

          • Garrett says:

            > And even not-preventative-but-not-catastrophic care helped prevent the worsening of conditions to prevent the catastrophic care.

            The problem here is that insurance is on the hook for worsening of chronic conditions which do have treatments. For example, if someone is an insulin-dependent diabetic and they fail to properly take their insulin, they can end up in the ICU for several days, or worse. But someone failing to do so adds a lot of cost to the system. So there’s this hope that by covering lots of primary care the ICU stays will be avoided.

            Same with heart disease, etc. But there’s no willingness to apply responsibility to the patient for their risky actions. Hell, stop covering medical care for people driving a car without a seatbelt or riding a motorcycle without a helmet.

  62. Telomerase says:

    Could you please show me that “$411 per month health care plan”? My company plan (no, it’s not “employer paid”, you know more economics than that), costs about $21,000 per year.

    I’m in New Hampshire, which is sort of part of the United States.

    (warning: if I could get health care for “411 per day”, I’d just sit around and blog all day. But I WOULD link to SSC 😉

    Seriously, how does one even find out how much health care would cost for a given individual? Are there some carefully hidden quote web sites? You see those for car and home insurance, but where are the health care prices? Thanks in advance for any pointers from anyone.

    • Swami says:

      California has web sites like that easily googled (or Duck Ducked). Mine comes to about $18,000 per year for a couple with about a $6000 deductible/out of pocket additional max. IOW, we spend between $1500 and $2000 a month depending upon whether we actually use it or not. This is the competitive rate for a bronze (not silver or gold) package. The better packages cost more per month but tend to have lower out of pocket maxes.

      • Telomerase says:

        Ok, that ain’t no “$411 per month”. WTF did that come from?

        Your figures sound close to what my company is paying (we’re a little higher because company pays most deductibles).

        • Swami says:

          I am assuming a real young person in a rural area. But it is nothing like what I paid in Illinois or California. And in Illinois I was locked in to a pre Obamacare package.

    • etheric42 says:

      HDHP, 3 individuals currently, $700/month total for 1 adult and 2 children. The plan is optimal in my area for non-premium healthcare spends above 7k and below 4k when I last checked, not including the tax benefits of the HSA. My employer doesn’t offer health insurance thankfully so I only pay $200/month total. I paid even less when my wife was on the plan due to the weird way the subsidy works (the $ of the subsidy is equal to a silver plan, so if you go with a bronze plan the subsidy scales faster than the premium does). If the cost of health insurance is your only concern, go get your blogging on.

  63. Jeff R says:

    A big part of the conundrum, I’d think, lies in the fact that the US spends a lot on end of life care via Medicare. I quickly dug up a Richmond Fed report on the subject:

    in the U.S., average medical spending from all payers during the last 12 months of life was $80,000 (in 2011, measured in 2014 dollars) and spending during the last three calendar years of life was $155,000.

    https://www.richmondfed.org/-/media/richmondfedorg/publications/research/working_papers/2018/pdf/wp18-18.pdf

    Obviously, if you’re just steering clear altogether of intensive and expensive interventions, like what Scott describes, you’re probably going to drive down your average healthcare spending per annum.

    Another factor is that the Amish may just be healthier than the average American, in terms of having lower incidences chronic conditions like diabetes. You can’t sit at a desk all day getting metabolic syndrome when you eschew post 1900 technology (mostly).

    This is just a guess, but I would guess there’s a behavioral economics aspect to it, too. When you’re spending your neighbors coin on medical services, you’re likely to be a little more judicious, whereas with Medicare being financed via payroll taxes, many older people I think implicitly have the attitude that they paid into the system for decades, and now that they’re on the plus side of 65, dammit, they’re entitled to take some money out of the system and spend it on whatever services they see fit. I’ll bet people act a lot differently when they think of healthcare services as “spending my neighbor’s money” vs “the system.”

    Anecdotally, I would add that it also seems like some old people enjoy going to the doctor, when they know they’re not paying. Someone has to listen to them babble, for a change, and it gives them an excuse to get out of the house. Yes, I’m speaking of my Grandma here. She doesn’t drive anymore because her eyesight’s terrible, so she doesn’t get out much, but a doctor visit is an excuse to get my aunt to drive over, pick her up, and then they usually go to lunch or get coffee together afterwards. This is one of the few social opportunities she has on her calendar in a given week.

    • Telomerase says:

      Yeah, just went through taking care of an old relative. They blew through all their money in the last year just paying for 24-hr aides (they had great insurance, but all it would pay for was doctors and nurses, not aides or housekeepers).

  64. craftman says:

    Rich people paid extra for fancy hospital suites and helped subsidize everyone else.

    This feels like the airline model, and I approve of it. But, if you have been in a hospital lately, where exactly are the “economy seats”? I worry that we all want the fancy suites and the insurance companies have slowly adjusted to paying for them. And like a frog in water brought up to a boil, we forget when we go to the hostipal that we’re sitting in the equivalent of a first class suite on a trans-pacific flight.

    • Matt M says:

      Agreed. “Budget” health care is basically illegal in the US. Start suggesting that what we really need is a “Wal-Mart of health care” and watch as 80% of the population shrieks in horror.

  65. ConnGator says:

    Blame doctors and the AMA. Reason has a great article on the history of how we used to have a near-perfect system that avoided the twin evils of rationing and over-treatment: mutual aid societies and consumer cooperatives (with no health insurance companies).

    Doctors were not making enough money so the squashed them. Spoiler: they are (sort of) coming back today as Direct Primary Care (DPC).

    https://reason.com/2020/04/05/how-doctors-broke-health-care/

    • Telomerase says:

      To fair, pharma and its cartel-enforcement agency helped break health care too. I actually wish the AMA was still reviewing drugs for safety instead of the FDA that just banned all commercial AND hospital PCR test kits at the beginning of a pandemic:

      https://www.technologyreview.com/2020/03/05/905484/why-the-cdc-botched-its-coronavirus-testing/

      Yeah, it would be nice if CDC could make a test kit as well as a random Mayo grad student… but if FDA hadn’t banned Mayo (and everyone else including Co-Diagnostics) from making them, it wouldn’t have been such an utter disaster.

  66. ConstantConstance says:

    I am a member of a Mennonite church. Some things to note:

    1. It’s not Old Order Mennonite, so it’s not that similar to the Amish.
    2. I read rationalist blogs, so I’m obviously a very bad Mennonite.
    3. Everything I say here is mostly stuff that I’ve heard from church friends who have Amish relatives. I don’t have any statistically solid research to back up my claims.

    So pretty clearly you shouldn’t pay attention to my remarks. But here they are anyway.

    There is a strong “stop whining and get back to work” ethic in the Anabaptist tradition. I don’t think this is based in theology; it’s cultural, and comes from the members being relatively poor and doing a lot of agricultural work, which leads to getting bruised, banged up, and often gives you really bad back problems. You deal with these problems using home remedies and by sharing your pain in “prayer and praise” sessions with your fellow church members. If, for instance, you use a hand tool to scrape the hair off of a pig you’ve just slaughtered, you’re likely to get a bit of a rash from leaning against the prickly-haired pig, and you might cut yourself with the tool. You don’t go to the doctor about that. You just complain to your friends who sympathize and pray for you. You also wash up thoroughly, which helps.

    In the Mennonite world (not directly the Amish world), lots of older women are nurses. They became nurses at a time when nurse training was mostly on the job, in a hospital setting. You became an RN by being shown how to do a procedure, then doing the procedure, then getting the “she knows how to do the procedure” checkbox checked on your record. Eventually you got the “Okay, she’s a nurse now” checkbox checked. Even today many younger women are nurses; if you look at the course offerings of the Mennonite colleges, you’ll find that they almost all offer nursing degrees, and those programs are well-attended.

    Consequently, a lot of basic medical knowledge finds its way to the friends and relatives of the Amish. This is mostly of the “advanced first aid” variety, but still, it keeps people from going to the doctor for basic stuff.

    There was a time in Colorado, in the early part of the 20th century, when many of the hospitals were Mennonite. I don’t think any of them are now. But Mennonites have a strong tradition of health care, because it is in keeping with Jesus healing the sick. To be a health care worker was to “follow Jesus” in a very direct way. You can read more about Mennonite hospitals in this article: https://gameo.org/index.php?title=Hospitals,_Clinics_and_Dispensaries

    Distribution of monies from religious-based health collectives is very unfair. I actually can probably dig up some studies on this. The more people like you, the more money you get. Men’s pain is more important that women’s pain. It’s a pretty serious problem. My husband is a lawyer (and hence also a very bad Mennonite) and he deals with the fallout of these “mutual aid” systems when they fail. Please do not get starry-eyed about these systems. They often fail in terrible ways.

    The link above, to Gameo, is an excellent resource for all things Anabaptist. Gameo stands for Global Anabaptist Mennonite Encyclopedia Online. I highly recommend it.

    • Telomerase says:

      ALL things Anabaptist? Am I going to find an online wargame for the Siege Of Munster? 😉

      >”Men’s pain is more important than women’s pain.”

      (!) I was expecting to find a bunch of studies showing that women are tougher… but guess I should have known, we XY-syndrome people are just too dumb to feel anything. So apparently they should be spending more on XX pain, not less:

      https://www.sciencealert.com/do-women-tolerate-pain-better-than-men

      • ConnGator says:

        The Hardcore History podcast about the Siege of Muster was one of the most interesting/disturbing things I have heard in the past few years. Not for the faint of heart.

      • ConstantConstance says:

        ALL things Anabaptist? Am I going to find an online wargame for the Siege Of Munster? 😉

        Ha! If I’m lucky I could scrounge up a jigsaw puzzle of it. Maybe I should make a jigsaw puzzle of it. I could sell it at the next fundraiser alongside the many dozens of plates of Pfeffernüsse.

        And by the way, if a low carbohydrate diet is essential to good health, then the Anabaptists ought to all be dead by age 25. Mmmm, Pfeffernüsse.

        • Telomerase says:

          You can eat more carbohydrates when you farm by hand. Though I had an uncle die of Type 2 at age 35, and he had farmed along with me pretty much by hand… our tractor was from 1918 😉

          That Munster puzzle would be an innovative Amish tourist product 😉

          My mother played Family Farm for several years before she died… is there an online Amish competitor for WoW? Maybe World Of Simplicity?

    • eqdw says:

      I haven’t gone to church in like 12 years, so in the religious sense I’m a bad Mennonite, but I thought I was the only one here. What up, my probably-fourth-cousin?

      • Laura Alexander says:

        There are at least three bad Mennonites here! Hello from PA. I was raised Mennonite, but not a Mennonite descendant, so probably no relation.

  67. Telomerase says:

    MIRI was barking up the wrong clade. The “AI Risk” we’re underfunding isn’t Artificial Intelligence, it’s Amish Increase!

    “according to a new census, the Amish are growing faster than ever. There are nearly 251,000 Amish people in America and Canada, according to Ohio State University researchers. That’s more than double the estimated population in 1989 of about 100,000. Researchers estimate the population will double again to half a million within about 21 years.”

    Only one man tried to warn us, and all we did was persecute him with copyright suits:

    https://www.youtube.com/watch?v=lOfZLb33uCg

  68. Doug S. says:

    There’s another reason why everyone hates health insurance companies. We’re not their customers! People who get health insurance through a group plan with their employer get stuck with whichever insurance company their employer picks. If you think they suck, well, too bad, the alternative is to pay for your own insurance on the individual market with no tax subsidies, which costs a fortune. If my father’s auto insurance company gave him the same headaches that his health insurance did he would have switched insurers years ago, but since his employer chooses the insurance company, he can’t. (Part of the problem is that he works in a different state than he lives, and most local doctors and hospitals are not in-network providers for his out-of-state insurance.)

    • Telomerase says:

      We only buy health insurance through our employers because providing health insurance was a way to get around WW2 price controls.

      Stupidity is far too “sticky” in our system… smart things are evanescent, but idiot programs and agencies go on forever.

    • etheric42 says:

      I made (I feel) a convincing case for my previous employer to stop offering health insurance. Most of our employees were <$15/hr workers (many much less) and the insurance plan we offered was horrible for what it cost. I argued it was cheaper for us if we just stopped offering health insurance and paid the ~$2k/head penalty and then it would be cheaper for our employees to then just get a subsidized plan on the marketplace (losers in this situation would be individuals whose spouses had insurance through their work but their spouse-coverage plans were worse cost/benefit ratio than our employee-only plans due to the family glitch, but from the employees I surveyed that affected almost no-one because our plans were so bad, and we could have easily found a way to funnel money to the employees affected and still have saved money).

      I lost the pitch because of concerns that not offering health insurance would make us look bad and make hiring difficult (because it's amazing how much companies will push "we offer health insurance" to applicants but then look confused when you ask them for their plan terms).

  69. Prussian says:

    Am I seeing things or is Scott edging in the direction of realizing that government control of healthcare is a bad idea?

    Sorry to be sarcastic here, but all of this was predicted well in advance. You can go read Ayn Rand on this, or Leonard Peikoff’s essay about the death of a profession, or go right back to Isabell Patterson who spells this out.

    • eric23 says:

      He says that Amish live ~5 years less despite their healthier lifestyles. Maybe Ayn Rand considers that a win, but I don’t…

      • craftman says:

        5 less years at the end of your life in exchange for what, though? According to the back-of-the-envelope caluclations ($2,000 per year vs. $11,000 per year and a $50K average income) you have 20% more of your yearly income back in your pocket.

        For a lot of families in the $40-60K income range, it doesn’t seem like a stretch to argue that it’s a decent trade off. The Amish don’t really take advantage of it but it probably means a less stressful life, more leisure, etc. during the years you do have on this earth.

        • eric23 says:

          For some people, it’s 5 years less at the end of life. For other people, it’s a lifetime of not suffering with a curable disease.

          For most middle-class people, 20% higher income allows them to buy what, a bigger house, or a fancier car? These are not things which really make one happier. A neighborhood with less crime, or with better schools for the kids? Everyone would be 20% poorer and so would be in the same position in these arms races.

          (Also, the number should be 14% not 20% – US health care spending is 18% of GDP and the assumption is that Amish health care would save 80% of that)

    • alchemy29 says:

      That’s a very strange conclusion. If you consider the Amish to be a self governing society, which they certainly seem to be at least along this axis, their system is literally communism. He concludes that health insurance is possibly bad for society, but in countries where the government controls most of healthcare, health insurance is rare or nonexistent.

      I don’t know what system is best. I don’t know if you have a specific proposal for a system – just no regulations or government spending on healthcare at all?

      • RC2020 says:

        I don’t really see how Amish society can be seen as communist. Property and businesses are owned by individuals. In some settlements, there is even something of a class system, composed of farmers, who may be wealthy in terms of land but not necessarily in terms of money, day labourers, working for Amish and non-Amish businesses, who have a steady, but often not particularly high, income, and finally business owners, who can have a lot of disposable cash. If you treat the Amish as a self-governing society, you could, I suppose, view their healthcare funds and other mutual aid programmes as a form of socialism, though I suspect few Amish would see it that way. A better example of a communist Plain Anabaptist society would be the Hutterites, who live on communal farms with little personal property.

        • Simon_Jester says:

          They have private property, but they’re running their health care system on a basis of “from each according to his abilities, to each according to his needs.”

          They’re not a Randist paradise, in other words- quite the opposite. So they aren’t supporting evidence that a Randist or otherwise anarcho-libertarian approach to handling health care works.

          • IrishDude says:

            Scale matters. Voluntary socialism on the scale of Amish farms is perfectly compatible with anarcho-libertarianism.

    • TheAncientGeeksTAG says:

      He should not be concluding that,since there are healthcare systemd which are much more government controlled than the US one,and much cheaper.

      • sharper13 says:

        Serious question:
        Which health care systems managed to spend less than before after switching from a more market-based to a much more government controlled system?

        • thisheavenlyconjugation says:

          Which healthcare systems didn’t?

          • sharper13 says:

            As far as I can tell, all of them, but I haven’t looked at 100% of the health care systems in the world.

            Certainly, the UK and Canadian systems dramatically increased their costs/spending after their changes toward government-provided systems.

            It seems to me that anyone who argues that Action X will cause result Y based on other people’s experiences (In this case, switching to a government system in the U.S. will reduce health care spending) should be able to point to some examples of when that actually happened at some point in the world sometime in history.

            Until then, I think it’s reasonable to conclude that the opposite will happen, because that’s what appears to have always happened before when the same action was taken by others elsewhere (and, to be fair, in the U.S. States which have done something similar).

            But I’m wide open for counter-examples to at least open up the possibility of this time being different. Maybe Singapore is a good bet? I haven’t researched their transition yet.

          • Edward Scizorhands says:

            If either of you would supply examples, it would help a lot.

          • thisheavenlyconjugation says:

            @sharper13

            Certainly, the UK and Canadian systems dramatically increased their costs/spending after their changes toward government-provided systems.

            Source? There’s no huge discontinuity at 1948 in the graph here; there’s an approximately straight line increase from 1915 to 1982 with a dip for WWII. You can’t be referring to the absolute increase since 1948, since AFAIK that has occurred in basically all comparable countries regardless of their healthcare systems (and in fact if we did a comparison including the US I expect government control would correlated with a *lower* increase).

            @Edward Scizorhands
            I’m not claiming that “switching from a more market-based to a much more government controlled system” typically reduced costs historically, but rather that there are approximately no instances of that switch that don’t have obvious huge confounders (like being the implementation of universal healthcare). But actually from looking at that data for the UK, it looks like there wasn’t a large increase even though you might expect one (since treating more people costs more money).

          • sharper13 says:

            The source you’re citing is actually measuring two different things, GDP and health care spending in the UK, but even then it’s obvious from the graphs that when the NHS was implemented, spending increased significantly, 2-3x within a few years, depending on how you measure and from when. The UK actually cut benefits right away from what was originally implemented, eliminating dental work and other “nonessentials” over the short term, because the costs so exceeded their expectations in the first few years.

            Anyway, here’s a direct quote from the source you linked: “After the war, with the creation of the National Health Service in 1948, healthcare spending began a sustained increase…”

            Not exactly a cost savings.

            (On my phone, but if your own source isn’t sufficient, I’ll dig up others from my computer tonight.)

          • thisheavenlyconjugation says:

            I don’t see any reason why we would be interested in talking about absolute spending rather than as a proportion of GDP.

            it’s obvious from the graphs that when the NHS was implemented, spending increased significantly, 2-3x within a few years, depending on how you measure and from when

            It certainly isn’t. Spending as a proportion of GDP was 2% in 1947 and 1.8% pre-war. It didn’t reach 2x that level until the 70s. But even if you consider 20 years to be “a few”, as I said before that growth was the continuation of a trend that began several decades before the NHS.

          • sharper13 says:

            Here’s the constant-pounds-per-capita graph of the early NHS time period from the same site. Presumably, you can see the change in direction at the time the NHS took over. It’s also on the %GDP graph, they don’t differ that much.

            Using constant per capita pounds, it took four years to double at the start of the NHS. If you want to be most generous in your time selections, it took 9 years to double.

            In terms of trendlines, it absolutely changed upward with a steeper slope at the time of government socialization of health care in the UK.

            In terms of some people’s misinformed idea that the switch to the NHS reduced spending, at least you appear to be conceding that didn’t happen, so kudos there. After that, there’s no real need to debate the finer points of exactly how much and how fast the NHS system increased health care spending.

            Note, here’s a basic summary of the timeline of legal changes in the UK. The initial legal changes to a more government system began with the National Insurance Act in 1911, the NHS was just a major expansion and switch to tax-and-spend for all rather than a required wage-deduction insurance based scheme for the employed.

          • thisheavenlyconjugation says:

            Obviously absolute spending per capita increases over time. Again, why should we be considering that rather than proportion of GDP? The whole reason we’re discussing this issue is because the latter has risen. If absolute spending had risen at the same rate as GDP no-one would be interested in healthcare prices.

            It’s also on the %GDP graph, they don’t differ that much.

            It certainly isn’t. Spending as a proportion of GDP was 2% in 1947 and 1.8% pre-war. It didn’t reach 2x that level until the 70s. But even if you consider 20 years to be “a few”, as I said before that growth was the continuation of a trend that began several decades before the NHS.

            The initial legal changes to a more government system began with the National Insurance Act in 1911

            Yes, it’s certainly true that many healthcare systems became “more government” over the course of the 20th century, and also that spending increased. But the existence of a causal link is the conclusion you’re supposed to be providing evidence for. You can’t just assume it.

  70. MasteringTheClassics says:

    Just to log a data point, I got my wife (low 20s, no preexisting conditions) catastrophic coverage through a Christian health sharing ministry a couple years back and the price was $125 a month. About a decade ago I saw the same ministry advertise that coverage for a specific accident-prone young family of 4 never exceeded $240 a month. I don’t understand the economics, and the coverage was very sparse, but such prices do exist outside the Annabaptist communities.

  71. vmsmith says:

    Some years ago — decades, actually — there were two follow-on publications to the Whole Earth Catalog. I think one was the Whole Earth Quarterly and the other was called the Co-Evolution Quarterly.

    In any case, at some point in the late-70s or in the 80s there was a fascinating article in one of them about the difference between English farming and Amish farming. At the time English farmers were having some serious financial difficulties, and bankruptcies were soaring, but Amish farmers were doing quite well. It was a really interesting look into similar sorts of practices mentioned in this medical analysis.

    One thing I remember was that tractors were not used because they would induce people to take out loans to buy more land. They felt that whatever a farmer could farm with horses was just about right. One downstream effect was that the farmer always had enough horse manure to fertilize his fields, and did not need to buy chemical fertilizer. So the Amish farmer never went into debt to buy tractors or additional land, and never had to buy fertilizer. It was as though every single practice was part of a virtuous circle.

    And really, the pillar in the middle (which you touched upon) was the strong sense of community and helping each other (a barn raising being a perfect example).

    Having been raised not too far from Lancaster, PA — Amish capital of the world — I grew up with Amish sensibilities lightly seeping into my consciousness from time to time. There was something nice and gentle and soothing about driving slowly on country roads because you never knew when an Amish horse and buggy would be around the bend. We could do worse than study how the Amish have crafted their communities and life styles.

  72. LGS says:

    I feel like I’m going crazy here. The Amish have some of the healthiest lifestyles (low smoking and obesity rates, good amounts of exercise), and their life expectancy is ~5 years lower than regular fat desk-job Americans, and you consider this a success story? What the hell?

    The Amish should probably expect to live 10 years longer if they had access to proper healthcare. This is not just about end-of-life care, either: for one thing, very few people actually spend 10+ years in a nursing home; the average number of years seems closer to 4 depending on how this is counted. For another, unless the Amish kill their elderly, they can’t avoid all costs of elderly care and must keep helping their elderly walk and get dressed until they die naturally (which could be a couple of years).

    It sounds to me like being Amish might save you $300k in healthcare costs over a lifetime, but likely costs you as much as 5 QALYs. Would you pay $60k to buy a QALY for yourself? This is roughly what Americans regularly spend on life years, so I think the answer for most people is “yes”.

    • eric23 says:

      I’m not sure he considers it a success story. Even if it’s not a success story, there might still be things we can learn from it.

    • SamChevre says:

      I don’t think obesity is particularly low in the Amish world–certainly it’s not anywhere I have connections. Gross obesity is rare, yes–but BMI in the low 30’s is common.

      Another thing to note is that people who do heavy physical work tend to have lower life expectancy, even controlling for observable factors. Part of this is much higher death rates relating to on-the-job injuries–a couple 25-year-olds a year dying in construction and farm accidents in a population of 10,000 pulls overall measured life expectancy down – and part of it we don’t know why, but observe it. But neither of these is amenable to better medical care.

      • albatross11 says:

        Yeah, there’s so much different between Amish and non-Amish communities that it seems like it’s going to be hard to untangle how much is health care. (Consider the differences in life expectancy between blacks and whites in the US–that might be primarily economic, cultural, genetic, or some other thing, and it’s hard to untangle.).

        From what Scott said, Amish life expectancies have stayed about the same, while non-Amish life expectancies have grown somewhat. The natural question is whether this is mainly due to better health care, or some other thing like better emergency services or fewer people working in dangerous jobs.

  73. Carl Milsted says:

    One factor to always keep in mind when looking at health care expenses: EVERYTHING that requires domestic labor is more expensive relatively. Legal services, education and general government are all expensive. The reason is that monetary inflation has been much higher than official rates. The official rates have been pulled down due to cheap imports.

    Not only are imports intrinsically cheap, they are taxed at a lower rate than U.S. made merchandise. If we replaced income and labor taxes with a universal consumption tax, we would be taxing imported consumer goods at the equivalent of a 30% tariff.

    Joe Sixpack has a hard time paying his doctor because he has less to trade. In the old days even a high income doctor would spend his income buying Joe Sixpack’s services indirectly. Today, high income doctors spend a greater fraction of their income on foreign made merchandise and vacations to Costa Rica. The loop is open.

  74. yossarian says:

    From my experience with doctors, half or more of them don’t even require any particular intelligence or acute skills. Some neurosurgeon? Hells yeah, that does take skill and intelligence. An average therapist/general practitioner – if it’s a condition usually known as “common cold”, write a sick leave note for a week. Else – send them to a professional in a particular field (you are sneezing from your arse? Go see a proctologist). Work that could be automated via a 10-or-so lines of bash script.

  75. deciusbrutus says:

    I was going to ask why Amish population wasn’t exponentially increasing with a rather large exponent, given that each family has 5-10 children… Then I did a minute of research.

    Amish population is exponentially growing, with a doubling time of about 20 years, and has been since 1900. The reason they haven’t already become a majority is that they started with a very, very small population. A reason that they might not become a majority in the future is land use limits and category problems- without a centralized “Amishness”, when new congregations form in new geographic areas, they can have value drift and stop being Amish.

    • RC2020 says:

      While new Amish settlements do sometimes end up ceasing to be Amish, usually through becoming car-driving, often evangelical, Amish-Mennonites, rather than abandoning conservative Anabaptism altogether, this is usually prevented by the strong links, both religious and social, between the new settlement and the one (or more) that its founders came from, and with other settlements they are in communion with. Visiting ministers will preach at services, and visiting bishops will hold baptisms, wedding and communions, until the new settlement ordains its own, or can entice a minister to move there, which may take years – and in any case ordination requires the presence of visiting bishops to be legitimate. Even after the settlement has its own clergy, they may still maintain links with those in other settlements by attending ministers’ meetings. Social links to other settlements are maintained by visiting them for social events, particularly weddings. These events are also an opportunity for the new settlement’s young people to find marriage partners, which can further strengthen relations between settlements.

      • deciusbrutus says:

        I predict that if Amish birthrates continue at the current rate, the rate of children of Amish being themselves not-Amish will drastically increase.

        • RC2020 says:

          What are you basing that prediction on? And are you predicting that fewer children will join the church, or that more will leave it after baptism?

        • Gnecht says:

          Current retention rates are in the 60 to 90%, I think. Pennsylvania’s Elizabethtown University has a “center” that specializes in studying Amish-type communities, their population surveys are probably among the best of their kind. They say average 85% retention.

          groups.etown.edu/amishstudies/frequently-asked-questions/

          Perhaps retention will become less, as you say.

          On the other hand, as Amish and mainstream cultures each continue their own paths, the effort required to cross the gap between them increases. It’s already tough enough that at least 2 organizations have been formed to help Amish leave their communities.

          mapministry.org
          amishheritage.org

          Try doing that with any other religious or cultural minority, and see what happens.

          • SamChevre says:

            The thing is that the 60% to 90% is a “stay Plain” retention rate; the number who stay in communities like the one they were born in is lower. A lot of car-driving Mennonites have an horse-and-buggy church grandparent.

            I grew up in one of the strictest of the car-driving groups; I think 80% of the youth I grew up with are still Plain, but less than a third are still in the churches I grew up in.

          • John Schilling says:

            On the other hand, as Amish and mainstream cultures each continue their own paths, the effort required to cross the gap between them increases.

            And as the Amish/Mennonite/”Plain” community grows absolutely, it will probably be easier to find a place to fit within it. As David Friedman noted in his book, each group makes its own rules. And as SamChevre notes above, moving within the broader community is more common than leaving it altogether.

  76. m.alex.matt says:

    This Econtalk is an interview with someone who wrote a book (partially) about what the healthcare system looked like before the rise of insurance. It’s also got a lot of polemic against the AMA and the insurance companies themselves, but the base data/descriptive aspect is very useful.

    A modern healthcare system that relied a lot less on insurance but instead on subscription fees for horizontally integrated doctor’s practices like the pre-insurance practices Ford-Chapin describes doesn’t seem too crazy, especially paired with some kind of health insurance that is true insurance — designed to cover large cost, low-but-predictable risk events. Maybe a public catastrophic insurance plan.

  77. ajfirecracker says:

    Hey, this voluntary price-based system outperforms our heavily regulated system. I guess the one aspect of our system which is voluntary and has price signals is the reason our system works so badly.

  78. smack says:

    To what degree might the difference in life expectancy be due to the prevalence of Maple syrup urine disease among Amish, which causes child mortality? This is a genetic condition particularly prevalent among Amish.

    • SamChevre says:

      I’d expect the impact to be trivial – that’s an issue in a few small subgroups within the Amish world. I’d guess (with no research) that the impact is not much greater than the impact of sickle cell anemia in America as a whole.

    • Gnecht says:

      Maple Syrup Urine Disease is actually manageable with a no-leucine diet, and a prescription supplement formula of some sort. I know a fellow with the condition who’s about to have his 26th birthday.

      Also, in recent years, it’s been discovered that a liver transplant solves MSUD, and somehow the MSUD patient’s own liver is fully functional for someone without that disease. This allows for “domino” transplants.

  79. dbenamy says:

    The average Amish family earns about $50K (the same as the average [American] family, somehow!)

    Anyone have more info about how this works?

    • Akhorahil says:

      Maybe having ten kids to help out at the farm helps? It’s per family, after all…

      Also: zero unemployment. And these days, you can surely get a massive price markup when selling local Amish organic food to middle-class liberals.

    • SamChevre says:

      It’s a mixture of things, but primarily very hard work and lots of small businesses and tradesmen. Children work routinely–I was working 2000 hours a year when I was 9. Also, people routinely overestimate how much of the income in Amish communities comes from farming; most people in most places are tradesmen (carpenters, bricklayers, mini-barn builders) and do some small-scale farming on the side, if at all.

  80. tcheasdfjkl says:

    Reading some of the comments here got me curious about ex-Amish experiences, and in an article about one ex-Amish woman I came across this, on the topic of health care expenses:

    Even the simple life lost its appeal when her parents refused an MRI after she began having severe headaches; instead, she endured medieval-sounding medical treatments that involved inserting balloons up her nose and inflating them.

    The article doesn’t say if this is a normal experience for her community, but it does make me wonder if this kind of thing is part of the story.

    • SamChevre says:

      If you want a really good book-length picture, Levi’s Will is remarkably accurate. (It’s roughly based on the author’s father’s story.

  81. Carl Pham says:

    If the life expectancy of the Amish is indeed still in the low 70s, then that right there might be enough to explain the difference in costs. Life expectancy for Americans in general is now 78, which is not a trivial difference from 72 or 74, say, and those extra 4-6 years may very well be enough to explain a huge difference in costs. Don’t we read that most of a person’s lifetime healthcare costs come in the last few years of life, as the medical system pours effort into staving off death?* Since those costs are amortized over everyone, of whatever age, it follows that we hugely overpay for health care when we are young and healthy, so that we can be massively subsidized when we are old and sick — more or less the same thing Social Security does.
    ————-
    * Nor do I blame anybody for this: it’s actually hard to be anywhere near 100% sure when death is inevitable, there are enough random miracles and dark surprises in anyone’s experience, physician or patient, that it would take an iron discipline to make a cold call that in this particular case — yourself, your spouse, your child, a patient you know and like — the odds are no longer reasonable and you should just quit trying. It would not surprise me if there were some bitter Amish who felt the community had made a call like that, or the community had pressured a family into making a call like that, a bit too early.

  82. Thomas says:

    Two comments. First, I worked in medical claims for a national health insurance company. We did NOT try to scam the hospitals. It was pulling teeth just getting basic documents out of them. I would call and ask them to please please please fax me ONE DAMNED page from their medical records so we could pay a bill. Hospital records suck. Small practices were far worse, since they usually hired their spouse or dim cousin to handle paperwork.

    Two, those doctors in Tijuana? They do submit paperwork to US insurance, and we do pay. Not out of the US insurance market at all. So the cost saving must be somewhere else. In many cases they gave the paperwork to their patients who submitted it, and we paid the patient.

    • Garrett says:

      Regarding billing insurance, I get to be on both sides of this. As an EMS provider, I think I’ve had a total of one chart come back as not-fit-for-billing, due in this case to legitimately forgetting to attach a single page. It’s not hard. But’s it’s also easy in the sense that there’s only about 5 pieces of information which are actually required to make an ambulance chart billable. Whether insurance will *pay* is another matter because there are weird legal rules which prevent billing people from telling us what constitutes having wording to make something reimbursable.

      Of the few I’ve managed to trick out of billing people: writing up a chart for a patient who calls 911 because they have flu-like symptoms with severe nausea, vomiting and diarrhea will likely not be reimbursed. Adding that they were so weak/dizzy/nauseated that they “were unable to ambulate without assistance” suddenly means that insurance will likely pay, even though nothing about the patient presentation has changed.

      Now imagine having to figure out the “magic words” required to be documented for payment for every procedure which is performed. I can see why they’d all be cranky.

      On the reverse side, I have no idea why or how different coverage criteria are selected by different insurance plans. My insurance changed this year, again. I’m on 2 injectable medications. The first is a (comparatively) cheap, generic medication with a long history and is considered a first-line treatment. The second in an on-evergreen-patent medication which costs so much I blow through my annual out-of-pocket max in the first weeks of the year. Guess which one requires a complex pre-authorization. If you guessed the cheap one, you’d be right! Why? Who knows! The idea that insurance companies would even allow pathological insurance terms like this to be set up just boggles my mind.

  83. phoenixy says:

    This will get buried, but I was reminded of this really interesting article in Granta by David Feuer, Let There Be Light!, about his experience working as a psychologist in a Hasidic community. Part of the takeaway was that people in that community with mental illness, even really serious mental illness, simply weren’t getting treated. Feuer’s first patient is a schizophrenic man who has been experiencing symptoms for over 20 years. He lives with his mother and has never gotten any treatment. Somebody like that, who lives in an insular, healthcare-avoiding community with an illness that destroys their quality of life but doesn’t outright kill them, is never going to show up on the healthcare system’s radar or be reflected in any statistics about health outcomes.

  84. Gnecht says:

    I applaud this article’s accurate understanding of a culture that’s distant from the author in various ways. I live in an Old Order Mennonite community, which has differing applications of the same principles on which Amish communities are built. (Pre-1693, Amish and Mennonites were the same people.) From my own experience with Amish-type church community mutual-aid healthcare, just as the article guesses, one of its superpowers is making medical spending feel like your own money even when it’s not. Contrast with mainstream USA systems that usually include goals of making medical spending feel like it’s not your own money even when it is.

  85. Gnecht says:

    Here’s the worst “miss” that I picked up:

    For people who rarely leave their hometown and avoid modern technology, a train trip to Mexico must be a scary experience.

    I think it’s actually the contrary, and many Amish would see a “medical tourism” trip as an exotic vacation. Lots of Amish are actually quite into travel, even if they don’t ride on airplanes as a matter of principle. I’ve seen at least one tour operator that offered an Anabatist History tour of Europe (visits Traschelwald Castle, the Tauferhole, the Felix Manz spot on Zurich’s Limmat River, etc) with connections by ship. (USD 5K via air, USD 12K via water.) Once in Europe, there’s at least one hotel that explicity markets to Amish and Mennonites with their deep interests in family history.

    en.bienenberg.ch/images/stories/file/AnabaptistPOImapBienenbergInfo.pdf

    And it’s common for Amish to have extended family in distant communities needing obligatory visits for weddings, funerals, etc. “Amish Taxi” service is becoming an entire service sector of its own.

  86. Gnecht says:

    Second worst miss, also by several commenters:

    chronic diseases… I feel like your fellow church members would get tired of you pretty quickly. I’m not sure how the Amish cope with this kind of thing, and maybe their system relies on a very low rate of mental illness and chronic disease.

    People with long-term needs are actually seen as prime opportunities to demonstrate an Amish community’s capacity for care and support. Think “ongoing barn raising.”

    Look up Dr. Holmes Morton and the Clinic For Special Children in Strasburg, PA. It’s largely supported by donations from Amish and Mennonite communities. (clinicforspecialchildren.org) Somewhere I saw it described as “the only clinic in the world with (Ion-Torrent-something high-end DNA research gear) and a hitching rail out front for clients’ horses.” There is a TEDx Talk by Dr. Kevin Strauss from that place that describes what they do with researching founder-effect genetic disorders in Amish and Mennonite communities. The Amish and Mennonite populations are similar to Iceland’s, with a somewhat small and isolated people group descending from not very many founders. The afore-mentioned family history interests make the Amish and Mennonites quite attractive to genetics researchers. The CSC has studied Spinal Muscular Atrophy, a death-sentence congenital disease for which AveXis has made Zolgensma — the world’s most expensive drug at USD 2.125 million, says Wikipedia. Amish and Mennonites are at higher-risk for SMA, and somehow the CSC was able to work out a sweetheart deal with AveXis that as far as I understand it, basically can let many Amish and Mennonites get Zolgensma for free. (I can guess: “Hey AveXis, we’re really good with genetics research, we’ve earned the trust of a half-million population with great relevance to your new drug, and we’ll do lots of heavy lifting for you if you let them have it cheap. They support us, we’ll support you, you support them.”)

    Second, mental illness. The Amish absolutely care about mental health. But they are also concerned about mental health pros taking one look at them and saying, “Step 1 to solve your problem is leaving your (negative-adjective) religious community and entering the real world.” Amish would likely need to be treated with a cultural sensitivity on the level needed for indigenous tribes before they’d trust the mental health pro. There’s become quite a list of Amish and Mennonite “counseling centers” that provide short-term residential care for their people who seek counseling. Usually these are staffed by ministers or lay members from the same communities as those they serve. There have become enough of these that I’ve heard comments within Amish and Mennonite communities wondering why there are so many today, with waiting lists, while past generations didn’t have them. Standards of care vary among these places. For a mainstream mental health provider with a program for Amish and Mennonite communities, see…

    wellspanphilhaven.org/Programs-and-Services/For-the-Plain-Communities/Program-Overview

    Often a mental health patient will start out a place like that, under professional care, until any indicated medications are set up and the patient is stable. Then they’ll go to a more pastoral-care type place for help working through whatever issues they’re struggling with.

    Places at the other end of the Amish mental healthcare spectrum are unlikely to have websites, and may focus on nutritional approaches. Not completely unfounded, did I see something about gut-brain research once?

  87. argentus says:

    I think to answer questions about the emergence of insurance you would need to know the real rates at which people *weren’t* getting healthcare back before the widepsread adoption of insurance. Most of the explicit engineering in the social safety net was designed to address specific groups of people who apparently weren’t getting care: VA for injured vets, Medicare for the elderly, Medicaid for the chronically poor. Was it the case that these people were getting care before the dominance of insurance or were they really not getting care but they were “out of sight, out of mind” or disease and suffering were generalized enough that it just seemed natural and unavoidable?

    Given the litany of horrors that your average Progressive Era activist brought to public attention, I find it somewhat hard to imagine that people working in the Triangle Shirtwaist factory or living in a horrible tenement slum in Chicago were getting any kind of access to non-quack care. I also find it extremely dubious that freedman were getting any kind of reliable care after Reconstruction unless a particular black community was able to train their own doctors and healthcare people. Or American Indians many of whom live in tight communities rather like the Amish who theoretically should be more able to adopt a model like this but mostly haven’t for whatever reason.

    My hunch is that if you were poor and broke your leg, you are in a better position to pay out of pocket because of general cheaper prices of care. If your were poor and your leg was blown off in the war and you had chronic pain and PTSD, you were probably mostly out of luck.

    I’m very much a communitarian so I’m very interested in models that allow localities and communities to address their own problems. They usually have a much better understanding of what their problems are and how to fix them than some institution thousands of miles away trying to apply some one-size-fits-all policy.

    But what do you do with people who don’t really have a community to turn towards or with whole communities that are collectively stripped of their abilities to engineer solutions? Think a share-cropping settlement in the Deep South in 1889. Or a bunch of migrant workers picking strawberries for a pittance in the Central Valley. I think there were probably quite a few of these people always falling through the cracks and that this just became more visible and less tolerable for whatever reasons.

  88. Gnecht says:

    For further in-depth study on Amish communities, there is “The Journal of Amish and Plain Anabaptist Studies.”

    amishstudies.org
    ideaexchange.uakron.edu/amishstudies/

    Search journal for the term “health”:
    ideaexchange.uakron.edu/do/search/?q=health&start=0&context=15565073

    Hint to Scott Alexander: You find the Amish interesting? Also have questions about communism? Combine the two and check out the Hutterites, which originated from the same Anabaptist movement as the Amish, and have had the unique spectacle of a capitalist economy seeking market protection from a communal one.

    brandonu.ca/rdi/files/2014/03/SLIDES-Webinar-Presentation-Hutterites-in-NA.pdf

  89. borsch4 says:

    On your comments about the rise of insurance and the rise of costs our buddy Milton Friedman has written on that, and some commentators have pointed important things out. Essentially health insurance in the U.S. does not work like insurance anywhere else. It really arose as a quirk of WW2. Wage controls were put in and firms trying to attract workers started offer non-wage benefits such as health insurance. This went unnoticed by the IRS and eventually when the war ended and the IRS tried to tax health insurance provided by companies people had gotten used to tax free insurance and it became another stuck benefit. Firms competing for workers now have an economic incentive to give benefits more than cash (If I offer $1000 in wages to an employee they only see the wages-tax, if I offer $1000 worth of insurance the employee sees the whole benefit). This is essentially a giant subsidy to all health care consumption. While in certain cases there are positive externalities which justify subsidies to particular health care areas, a giant subsidy to all health care consumption makes no sense and raise all costs. In fact, this is one thing that almost all health care economists agree on: ending the tax exempt treatment of health insurance. Of course politically this is suicide, but when did economists ever care about politics.

    Two papers that are easy to read and interesting takes on the health care system in the U.S, one from Victor Fuchs the father of health economics and another from Milton Friedman:
    Fuchs V., 1996, “Economics, Values, and Health Care Reform”
    Milton Friedman, 2001. “How to Cure Health Care?”

    • Edward Scizorhands says:

      A lot of non-cash consumption wasn’t taxed before the reforms of . . . Reagan, I think. Company cars, company apartments, a whole crapload of it. I don’t know how health insurance escaped that one.

  90. Gnecht says:

    Somewhere in Southeast Pennsylvania, there is general-practice medical clinic run by and for Amish and Old Order Mennonite people. I forget its name, but in 2010 I spoke with one of the men on its board. The clinic organization put up the building, got whatever registrations the state needed, and hired the doctors and nurses.

    They had been unable to get costs much below any other clinic, he said. A doctor visit there costs much the same as elsewhere.

    HOWEVER, what an Amish family can have done with that doctor visit is a whole ‘nother matter, once other parties are out of the picture. The clinic board member’s favorite story came from a star doctor who’d since moved on, but said he really loved the freedom of working there. Once, an Amish-or-similar parent brought in a child with a severed finger. Time was running out to get it reattached. Star Doctor looked at it, figured he could handle it, and did the reattachment right there in the general-practice clinic. Everything worked out fine, everyone’s happy.

    And even if it hadn’t worked out, I doubt the Amish family would have blamed the doctor. I think they mostly don’t want medical pros to be god-like infallible authority figures, and are likely at peace with the idea that sometimes people make mistakes and best efforts can fail.

    Later, there was another patient with an almost identical severed-finger situation. But in this case, the patient wasn’t Amish and there WAS health insurance involved. Everyone said, “We’re not touching this one. Off to the hospital with you.” Time ran out, and the reattachment failed. Star Doctor was burdened, knowing he almost certainly could have saved the finger if he’d been allowed to have a go.

    This clinc can’t do major body-cavity surgery. That would need a much higher level of state registrations and compliances. But otherwise, it sounded like the clinic could do most anything a rural hospital commonly does. Set broken bones, etc.

  91. Anna says:

    Have you ever read the Atlantic article How American Health Care Killed My Father? It’s quite good, and it definitely makes that point about insurance companies.

    Short version: Using any third-party-payer system leads to rising costs. Solution: Have catastrophic-only health insurance, replace Medicare taxes with mandatory donations to an HSA, and when poor people can’t afford their care, just give them the money directly instead of trying to hook them up with insurance. I love me my Medicaid, but I think an EBT card with a reasonably sufficient “health” balance (like it has a “food” balance) would end up giving more people more power over their health choices, and help prevent health costs from rising in the long run.

    “Reasonably sufficient” varies widely with diagnosis, of course, so I think the balance would have to be actually diagnosis-based. It would be important to include a diagnosis of “something’s wrong, but we don’t know what, and we might spend the next ten years sending you to specialists before we figure it out”, and a diagnosis of “basically healthy person who still needs a yearly checkup and some occasional Tylenol or cold meds”. There’s definite potential for diagnosis fraud if some diagnoses give you more money than others, and I don’t have a good solution to that. The incentive for that could be minimized (but not eliminated) by keeping really good estimates of how much people actually spend for given diagnoses, and giving people that much, instead of trying to tell them how much they should be spending.

    • Edward Scizorhands says:

      From that article:

      Eight years ago, my wife needed an MRI, but we did not have health insurance. I called up several area hospitals, clinics, and doctors’ offices—all within about a one-mile radius—to find the best price. I was surprised to discover that prices quoted, for an identical service, varied widely, and that the lowest price was $1,200. But what was truly astonishing was that several providers refused to quote any price. Only if I came in and actually ordered the MRI could we discuss price.

      There are lots of people that hate the ideas of figuring out the price for a medical procedure. Like having to care about the price is something only low-class people have to do. Or in “other countries” no one cares about price (true for some) so we shouldn’t either. And frankly many parts of the health care system feel deliberately designed to make you feel helpless and distressed, so that you can’t spare any effort to talk about money.

      • Gnecht says:

        Related: Oklahoma City Surgery Center.

        reason.com/blog/2017/01/27/what-happens-when-doctors-only-take-cash

        surgerycenterok.com

        slatestarcodex.com/2018/06/20/cost-disease-in-medicine-the-practical-perspective/#comment-640948

  92. Gnecht says:

    Another consideration: it’s possible to have both church community mutual-aid health care… AND an organization that does deals with hospitals like heath insurance networks.

    immergrun.org

    The foundation serves Amish and Mennonites who because of their religious beliefs do not have medical insurance. Immergrün works as a patient advocate to obtain quality medical care at affordable rates, and to eliminate issues patients can experience when dealing directly with hospitals and physicians.

  93. kaleberg says:

    The era of rising costs was also the era where doctors started being able to do a lot more. Lewis Thomas, reflecting on his training as an MD noted that one of the few things doctors were taught to do really well was spot pernicious anemia which was one of the few diseases they could do anything about. There were no antibiotics, kidney transplants, SSRIs cochlear implants and so on. The 1947 PDR had 380 pages. The current one is over 2000 pages. Cancer only became a problem in the 1930s because life spans were increasing enough for it to show up in the death statistics. Insulin and sulfa drugs were pre-World War II, but penicillin was magic, and we treat an awful lot of chronic diseases – e.g. high blood pressure – that would simply kill one in the good old days.

    I’m sure we could have done a better job controlling the costs. A lot of other nations did. Still, the health insurance business, modern medicine and our insane US cost structure share a history.

  94. kaleberg says:

    It’s interesting how so much of this discussion is about MD salaries. Where is a discussion of the ROI and cost of administrators and CEOs? Why does a hospital room cost so much more per night than a nice suite at the Park Hyatt in NYC? Somehow or another it all seems to come down to overpaid working stiffs, and doctors sell their labor just likee any other working stiff.

    • justin1745 says:

      The CEOs, at the very least, are trivial when it comes to total costs because there are so few of them. I doubt their compensation combined equals even 1% of U.S. healthcare spending. If they all took a 50% pay cut, your average health bill goes down at most 0.5%. That’s nice I suppose, but $100 off my $20,000 bill just doesn’t move the needle.

      Physician and clinical services, however, was 20% of total spending in 2018, and if you cut that 50% you’d end up saving $2,000 on the bill above. Note I don’t think we can or should do that, but that’s why people talk about MD compensation more than CEOs.

      Administration is probably an area which would be more fruitful to find cost savings, but in a complex health care system like ours even that will be hard.

      The reason a hospital room is so much more expensive is that there’s far more expensive labor and capital involved. There’s likely far more employees per patient at a hospital and those employees cost more.

      I don’t know the exact numbers, but let’s assume for the sake of illustration that a hotel will have 1 person on staff providing service for every 4 guests, while a hospital has 5 people on staff providing service for every 4 patients. The hotel staff costs $100 per day whereas the hospital staff costs $400 per day. The hotel’s labor cost is $25 per guest and the hospital’s is $500.

      You could do the same exercise with capital expenditures. The Hyatt has very nice furniture and bedding and stuff, but the hospital has super-expensive medical equipment: MRIs, CT/PET scanners, X-Rays, and various other things. Everyone pays their Hyatt bill directly, the hospital has to negotiate third party billing with numerous faceless insurance organizations, it’s trying to cover costs from lower payers (Medicaid) or non payers (the remaining uninsured). I would suspect that hospitals have higher insurance costs than hotels, etc. Hospitals sometimes do more for you than what you need, through defensive medicine, to which the closest analog in the hotel space is probably the minibar which you can avoid with a modicum of self control.

      There’s also cost control via prices which is stronger for hotels. If the Hyatt raises prices too much, it will lose business to other high end hotels. If the hospital raises its prices, most of the bill isn’t paid directly by the patient, and while the insurance company does try to manage costs, it is limited in the sense that patients know which hospitals in their area are the good ones and will want them to be in network. Over time, this probably allows for more bloat in hospitals.

      Lastly, I suspect that because they are more complex organizations, hospitals probably end up overpaying for more things than hotels and that adds a bit to the incremental cost burden as well.

  95. justin1745 says:

    –“That makes even more people need health insurance, and the cycle repeats as prices grow higher and higher and insurance becomes more and more necessary. This syncs well with some explanations I’ve heard of rising college prices, where once the government made easy loans and subsidies available to everyone, prices rose until they consumed all the resources available. I have no idea if this is true or not.”–

    This is my view as well.

    Other countries have been able to limit this cost explosion because they’ve been more centralized and a higher authority can limit prices system-wide. It would be as if the U.S. had introduced college tuition assistance fifty years ago, but said in order to participate, total ‘sticker’ prices cannot rise more than inflation + 1%.

    I think the left critique that there are lot of administration costs is also real and part of the problem, and I like how you point out that small Amish societies can eliminate this by knowing people, by having a reputation for honesty, etc.

    I’m not sure how to fix this in the US. I worry that the costs cannot be cut too much from wherever they are. The administrative costs can be eliminated by single payer, but you’ll have an increase in demand due to less cost sharing and more access, and these will probably largely offset. From there, you can only get cost improvements from cuts to providers, and that’s really hard. People hate losses much more than they hate smaller gains. If health care spending had grown at a 0.5% slower pace since 1975, total spending would be over 20% lower than it is now.

    I worry that the amount of taxes required to finance a universal, single payer system would create enough deadweight loss such that it wouldn’t be worth it. The best path forward for the U.S. is probably a two-tier system.

  96. Synonym Seven says:

    I asked my literal grandmother, a 95 year old former nurse, how health care worked in her day. She said it just wasn’t a problem. Hospitals were supported by wealthy philanthropists and religious organizations. Poor people got treated for free. Middle class people paid as much as they could afford, which was often the whole bill, because bills were cheap. Rich people paid extra for fancy hospital suites and helped subsidize everyone else.

    Why did health care prices start rising? I’ve wondered about this a lot before – see here, here, and here. Looking into this issue, I noticed glimpses of a different possibility. The increase started around the same time that health insurance began to spread. In one sense, this is unsurprising – of course health insurance would become a thing around the time care became unaffordable. But I’ve never seen someone really try to tease out causality here. Might the two trends have been mutually self-reinforcing? The price of care rises due to some original shock. Someone invents health insurance, which seems like a good idea. But this creates a series of perverse incentives, which other actors figure out how to exploit (eg the Lucemyra example above). Insurance-based-health-care becomes less efficient, but hospitals can’t or don’t internalize this to the insured patients – they just raise the price for everyone, insurance or no. That makes even more people need health insurance, and the cycle repeats as prices grow higher and higher and insurance becomes more and more necessary. This syncs well with some explanations I’ve heard of rising college prices, where once the government made easy loans and subsidies available to everyone, prices rose until they consumed all the resources available.

    All this overlooks the single most obvious factor: early 20th-century treatments are cheaper than 21st century treatments. Just like how a 2019 Honda Civic is way more intricate than a Chrysler New Yorker or other cars from that era, or how it’s a far bigger investment to create a factory capable of creating Lego blocks than a factory capable of creating wooden alphabet blocks. Penicillin wasn’t even being prescribed until 1945, to say nothing of the even-more-complex specialized medicine we have access to today.

  97. Jules says:

    You mention a lot of ways that insurances raise prices for everyone, but the most ludicrous seems to have been left out:

    https://en.wikipedia.org/wiki/Chargemaster
    “In the United States, the chargemaster […] is a comprehensive listing of items billable to a hospital patient or a patient’s health insurance provider. In practice, it usually contains highly inflated prices at several times that of actual costs to the hospital.[1][2][3] The chargemaster typically serves as the starting point for negotiations with patients and health insurance providers of what amount of money will actually be paid to the hospital.”
    […]
    uninsured, and others who pay out-of-pocket for healthcare pay the full chargemaster listed price for the same services.[17][18]

    In a more fun format, as explained by Comedian Adam Conover: https://www.youtube.com/watch?v=CeDOQpfaUc8&feature=youtu.be&t=30

  98. dspyz says:

    In answer to the general question of “What causes cost disease?” have you considered that society-wide our productive labor is being re-purposed away from providing healthcare, education, and affordable housing.

    Consider that if 3 people own the same amount of wealth as 50% of the nation then from an economic perspective I can expect the approximately same revenue for catering to the needs of those 3 people as I can for trying to provide healthcare for the other 150 million (where “catering to the needs of” generally involves helping them find ways to further increase their wealth).

    So for instance, when I was working at YouTube on the experiments team, I was maintaining the infrastructure that other YouTube employees use generally to determine how to increase watch-time. I think it’s probably the case that the mechanism by which certain features increase watch-time is just as likely to be “addiction” as it is to be “better UI”. People don’t focus a lot on mechanisms so this is largely speculation.

    When YouTube is more addictive they increase ad revenue and in many cases decrease the productivity of users (because people watch YouTube at work). But even if YouTube isn’t actually _decreasing_ users’ productivity, it’s certainly not _benefiting_ users in any way. The real beneficiaries are Google and Google’s shareholders who make more money when more hours are spent on YouTube.

    Users don’t actually _want_ YouTube to be more addictive. This isn’t a quality or feature we look for when choosing a video streaming service. So I wasn’t doing a job which was _productive_ in the sense that it’s doing anything useful in the world. It was only productive in that it drives more wealth towards Google shareholders.

    But I took this job because I was well-paid for it and because it gave me prestige to be able to say I worked at Google and so now I have lots of other job opportunities. I think there’s a sense in which a large segment of the economy is geared towards unproductive or even counterproductive labor, but because this labor enriches a very small number of people, we all continue to pursue it because it’s just as profitable (if not _more_ profitable) than doing “useful” things. That leaves a smaller segment of the population “left over” to provide healthcare, teach schools, build reasonably-sized living spaces, etc. As supply decreases, prices go up. Note, I’m _not_ talking about the Baumol effect here. The price increases because of a decrease in _supply_ of medical care when large segments of the population are no longer interested in becoming doctors, _not because_ of doctors’ salaries having to be raised to match anyone in particular.

    You might think this increase in price _would_ accompany a salary increase, but various corporate actors such as banks, insurance companies, drug companies, student-loan agencies etc. find ways interject themselves and lap up the surplus revenue that comes as a result of these inflated prices so that the actual doctors/teachers/professors/construction-workers don’t see a corresponding salary increase.

  99. Calion says:

    I’m confused as to why you don’t even mention the standard libertarian answer that the problem arose with the government providing heavy tax incentives to businesses that offered health insurance, starting shortly after WWII, which disconnected healthcare expenses from healthcare consumers, and started the path to the ugly mess we have today.

    (Sorry if this has already been mentioned; I’m not going to read 547 comments to find out. On the other hand, you’re probably not going to read 548 comments in any case, so it’s unlikely that you’ll see or respond to this. As they say, There’s Got To Be A Better Way.)

    Edit: Perhaps it’s in one of the other three posts you link on this subject, which I’m not going to read right now. Still, I think it’s the best candidate.

    • MostlyCredibleHulk says:

      heavy tax incentives to businesses that offered health insurance, starting shortly after WWII

      IIRC it actually started during WWII, where there were salary caps to avoid employers – suffering understandably from the lack of working hands – driving labor prices sky high and thus create shortages at wartime since nobody can afford to hire anybody anymore. So, FDR, genius as he is, instituted wage caps. But employers aren’t stupid either – they still want to attract workforce. So they started offering fringe benefits. Like, for example, paid medical service. And then IRS sealed the deal by making it much cheaper for employer than for the employee to buy the medical service. And now it’s politically infeasible to decouple because anybody who tries is obviously wanting to let people die without medical service to make a quick buck, and thus is literally worse than Hitler.

  100. MostlyCredibleHulk says:

    I’m fascinated by how many of today’s biggest economic problems just mysteriously failed to exist in the past

    And then the state decided to help. College, houses, healthcare, anybody notices the pattern yet?

  101. wayanad Empire says:

    I like your website, my thanks for doing such great information
    Wayanad Empire