Hacker News new | past | comments | ask | show | jobs | submit login
Uncertainty and the Welfare Economics of Medical Care (1963) (fermatslibrary.com)
69 points by mgdo on Jan 13, 2017 | hide | past | favorite | 29 comments



Libertarians and conservatives believe that health care is or could be a well-functioning market, like buying a refrigerator.

But there are many basic differences. You know when you need a refrigerator, but often are not sure if you need a given test or health treatment. Refrigerators are bought one at a time, but tests and treatments are often whole bundles. Refrigerators of a given brand and model are basically identical, not so health treatments. If you get a bad refrigerator you can exchange it under warranty, a bad health treatment can cause you permanent damage.

You can get reliable information on refrigerators from Consumers Reports, not nearly so much for health care. You can generally get along without a refrigerator for a while, taking time to get information before making a decision which one to buy, often that is not the case with important health care.

There are many sellers of refrigerators with public prices and lots of price competition, not the case for health care. The prices for refrigerators are standard around the country, not at all the case for health care. The majority of health expenses are in the last few years of life, when people are in poor mental shape to make decisions. You can buy a refrigerator at a reasonable price without insurance, but not individual health treatments.

In these and i would guess other ways, a purchaser of health care services is, by the very nature of things, in a far weaker position than a buyer in a normal market. And of course, the health care providers like it that way.

Can anyone suggest any other basic differences?


I was under the impression that under the current US medical system that it is impossible to get a pre-purchase price on a lot of medical care. If you are going into hospital for a knee replacement you can't get anyone to tell you the total bill in advance.

One major difference is the actual price paid for a medical service by an insurance company is much lower (often 10%) of the cash price.

Another is that the you don't know what the insurance company will cover until months afterwards. None of these apply to normal consumer items where markets work.

The most efficient USA healthcare system would probably be to just expand the VA system to cover everyone and put a tax to pay for it.


> One major difference is the actual price paid for a medical service by an insurance company is much lower (often 10%) of the cash price.

It is often much lower, but it also varies by insurance provider and the specific plan you are on, which can itself (AFAIK) vary depending on when you joined said plan, your group, et c.

The hospital can't even guess at a price until they figure out which procedure codes they're going to bill, which itself isn't always clear, then to know what you'll actually pay they have to apply the effects of your specific insurance. It's often quite difficult to get them to do this. Often it will end up being wrong anyway.

Bonus: you can't be really really sure your insurance will pay their part until the bill goes out and they do. Sometimes they deny claims. They may be wrong to do so (it's sometimes hard to tell) but even if they are you're in for 100+ hours of work (much of it waiting on hold) to get them to do what they were supposed to.


All I can say is I am glad to be Australian. We might have some nasty wildlife, but at least I know that if I am bitten or stung I am not going to end up bankrupt.


I don't have enough certainty to want to take either side here, but wouldn't food make a better analogy than a refrigerator? Food is more similar to healthcare. It's something that you need to purchase regularly and it's vital to your immediate health.


Routine health care, like your yearly visit to your doctor, are more like food, but the important, expensive ones like getting cancer, which are the vast majority of health expenses, are occasional and unpredictable, so are more like buying a refrigerator when yours unexpectedly breaks down.


> Refrigerators of a given brand and model are basically identical, not so health treatments.

Heh, wonder if they would take the following compromise: no public healthcare option, but medical patents are hereby abolished. You know, getting the government out of healthcare... Not that I really advocate for this policy, but I mean, libertarians are not ok with being "coerced"[1] to provide medical treatment, but are ok with being "coerced" to not manufacture life saving treatments because someone else owns the IP to them, usually standing on the shoulders of decades of tax funded university research.

[1] scare quotes because I believe there is a difference between taking a particular individual and actually forcing them at gunpoint to be a public doctor, and simply taxing people into paying for a public health system that is attractive enough that enough people chose to work for it of their own free will (e.g. the NHS)


This is a little long to commit to close reading. Summary, anyone?


It's a classic in economics that explains why health care doesn't fit the free market ideal. The basic problem is that, unless you know as much about medicine as your doctor, you can't evaluate the quality of advice you're getting.

It's heavy reading, I don't know why it's getting upvoted here.


This is a bad argument.

Does anyone know as much as the contractor who made their house? Then why don't houses crumble, long after they are built?

A free market would have various accrediting bodies, and doctors would advertise the quality of their accreditation as evidence of their skills and qualification.


Houses can and do crumble, which is part of the reason building regulations were introduced. And builders/plumbers/electricians certainly aren't immune to the temptation to charge for unnecessary work. But for the average person, medical care is substantially more difficult to evaluate (your doctor is smarter and more heavily qualified than the average builder, right?) and a person worried about their symptoms has substantially more incentive to procure additional medical care that just might work than a person unsatisfied with their house procure speculative building work. I'd run a mile from a builder that conceded that despite their best efforts their structure might fall down of its own accord, and be nearly as worried by a doctor who insisted a remedy was guaranteed to work.

Accreditation bodies may be able to determine whether a doctor is knowledgeable; whether they overprescribe placebos to an individual patient to increase their revenue is significantly more difficult and resource-intensive to evaluate. And homeopaths and faith healers have accreditation bodies too.


It doesn't work out that way though. Asymmetric information inevitably leads to market collapse [1]. The only reason healthcare "works" is because we're not rational actors. I mean those more as mocking quotes rather than scare quotes. Perhaps it's more accurate to say, the only reason healthcare works is because many actors are not motivated solely by money. Which, a decent MBA would consider a sucker.

[1] https://en.wikipedia.org/wiki/The_Market_for_Lemons


Sure, but forcing healthcare on everyone makes the market failure a government failure--if "failure" means someone paying much more for insurance than it actually costs for that person.


There is a pretty easy calculation, but the numbers are hard to come by.

People pay about $3000 for health care in the us, per year. people pay $750 for medicare. To be fair, half is by the employee, half is by the employer, so let's say $1500, because theoretically you would make more if the employer didn't have to pay that tax. On the flip side, high paying jobs, the employer pays the lion's share of health insurance premiums - but let's just ignore that.

Old people (medicare patients) get sick all the time. Heart attacks, cancer, whatever. Young people rarely get sick, but there are more young people - perhaps 15,000,000 70-74 year olds, and perhaps 21,000,000 20-24 year olds. If the cost of medicare for everyone is less than $3500, it's cheaper to just cover everyone at some basic level via medicare.

Very few 22 year olds get cancer. It's expensive. lots of 72 year olds get cancer. That is also expensive. Medicare pays for the expensive part of people's lives (the end of their lives). I'm pretty sure if the medicare tax rate doubled, 3% to 6%, $1500 to $3000 it'd still be cheaper than what we're doing now.

I dunno. It's late. It's a complicated topic. I have no problems with well off people paying a whole lot to get special care, extra insurance, or whatever. I think it would be a good thing for people who get in car accidents, have heart attacks, or get cancer to know they're not financially ruined.

It's nothing more than a linear programming problem. Right now people pay X. If everyone was taxed more they would pay Y. If Y < X, we should do Y. If not, what we have now is the best we can do. So be it.


> If Y < X, we should do Y.

This depends on whether you're fine with some people subsidizing other people. The average cost by forcing everyone to get insured is less because you're forcing some people to buy plans for whom it's not worth it. See adverse selection.


> some people subsidizing other people.

This is called insurance. Historically some deeply religious groups were opposed to insurance, because they felt it was gambling.

> he average cost by forcing everyone to get insured is less because you're forcing some people to buy plans for whom it's not worth it.

It's really a timeframe question. A 20 year old, generally, needs less care than a 50 year old, who needs less care than an 80 year old. one approach is to provide lifetime care, and hope that technology improves (as it has for all of human history) and assert that care today for an 80 year old will be less expensive in 10 years than it is today.

Now, of course, actual payments for health care is more expensive year to year, because, as you say, adverse selection.

In any case, as you point out adverse selection (the problem of information asymmetry) means the payer must be as knowledgeable as the payee. An obvious solution (but maybe not the best solution) is something like medicare.

I'm not sure i understand what you're advocating. Perhaps you mean each person should set aside some portion of their cash for whatever expenses they might have, and the odd 20 year old with leukemia should take on a few hundred thousand dollars of debt. of course if you're unlucky enough to have one of those 10% survival variations, the cost will of course be multiplied out by 10.

The average american only makes about $2,000,000 over their lives, and they have to eat and have a place to live and stuff, and only half will beat the average, what sort of return could an actual care provider expect? perhaps $200,000? So your model becomes pretty clear, if cost to treat @ 18 > $200,000, the person shouldn't be treated. That number will change with education, but of course decline with age.

Seems like not a good solution, but i see where you're coming from.


> This is called insurance. Historically some deeply religious groups were opposed to insurance, because they felt it was gambling.

No. I'm saying it's a subsidy NOT because randomly you roll the dice of risk and you get lucky; but because the expected value or utility of health risks is less than the cost of insurance (for example, you wouldn't buy health insurance for 1 billion dollars a month).

I'm not saying it's a subsidy because some people don't end up "using" their insurance plan; I'm saying if you repeat the scenario a billion time healthy people will be on average worse off so it's a subsidy. They wouldn't take the insurance unless government forces them.

> I'm not sure i understand what you're advocating. Perhaps you mean each person should set aside some portion of their cash for whatever expenses they might have, and the odd 20 year old with leukemia should take on a few hundred thousand dollars of debt. of course if you're unlucky enough to have one of those 10% survival variations, the cost will of course be multiplied out by 10.

I didn't advocate anything; I'm just saying the issue is a normative question; it doesn't have an absolutely objective answer.

Sidenote: maybe the 20 year old should be in debt. If we want to save as many lives as possible, it's cheaper to put money into clean water, stopping open defecation, vaccines, etc (depends on whether you value someone outside your country as much as someone inside).

The leukemia case is convenient for your viewpoint but what about nonsmokers subsidizing smokers? Should I have to pay for the treatment of the cancer of a smoker or the AIDS of someone who injected drugs or the measles of someone didn't take vaccines?

It's a normative question on altruism.


Medical stuff is much more complex than building a house. You can pretty well predict how well a house will perform but every human is different. One therapy works for one person but not for the other. In addition your accrediting bodies would be highly susceptible to corruption as the ratings agencies have shown.

One thing that could be done is to open up all data hospitals and doctors have about their success rates and other quality numbers, also prices. Then a lot of people could crunch these numbers and make sense of them.


It's not a bad argument, because the relatively minor discoveries in medicine in the 19th century in a hands off market already demonstrated what happens. The right wingers also provided more material by loosening controls over supplement marketing in the 90s.

Desperate people seek out cures. In the old days they gave money to the church for miracles. In the 19th century they bought patent medicines, which at best were cocktails of opiates and alcohol and at worst were cheap whisky. Google the "Duffy Malt Whisky" company. In the 21st century, we have homeopathic quackery and supplements.

The sober advice of accrediting boards is always defeated by the silver tounge of advertising and marketing. Only government has the power to address this.


Information asymmetry is the general subject for those who are curious:

https://en.wikipedia.org/wiki/Information_asymmetry


That seems like a pretty bad example. House construction is pretty regulated in most places, with governments dictating anything from materials and design down to the number and location of electrical outlets.

A better example might be something like... rock climbing gear, maybe? It's hard to come up with things that are both life-critical and unregulated, whether because regulations are necessary or because government likes interfering.


I don't think anyone that has dealt with housing contractors thinks it is an ideal free market. The "uncertainty" concepts that Arrow developed in this paper have been used in studying all kinds of other markets from health insurance to used cars. In fact, this paper has been used to argue that health markets are not that special from other markets because there is always some uncertainty.


> A free market would have various accrediting bodies

This is actually an example of market distortion.

For me, I view the use of "Free markets" to be implied uses of "competitive markets" and Perfect Competition is defined as follows:

> the situation prevailing in a market in which buyers and sellers are so numerous and well informed that all elements of monopoly are absent and the market price of a commodity is beyond the control of individual buyers and sellers.

An accrediting body (of any type) is a symptom of a market where buyers are not (and perhaps cannot) be well informed enough to make decisions that maximize their utility/benefit.

Now you might think, at this point, that's ridiculous! No one can be well informed about every potential market decision they make in their lives. And you would be right, this is why it's an unachievable ideal:

> Though there is no actual perfectly competitive market in the real world, a number of approximations exist [2]

Moreover, there exist market externalities whose burdens aren't appropriately placed on the responsible actor. The classic example is pollution, which can often be modeled as a social cost rather than a supplier cost [3].

Now accrediting bodies and government regulation can be good steps at correcting these distortions, but they aren't perfect...as anyone who has hired a licensed contractor and gotten poor results can attest.

> Then why don't houses crumble, long after they are built?

There is a lot of government regulation that tries to correct for this. But even then, a lot of homes have significant issues despite things such as home inspectors, building permits, contract licenses etc. My parent's fairly expensive home, for instance, had a chimney that was nailed to their roof and ended up falling over in a windy storm causing roof and water damage. The house was new construction, inspected, and part of a large development.

But a lot of government regulation is actually an argument in support of the points made in this publication...as it means that the market was not suitable to laissez-faire handling [4].

[1] https://www.google.com/search?q=perfect+competition&oq=perfe...

[2] https://en.wikipedia.org/wiki/Perfect_competition

[3] https://en.wikipedia.org/wiki/Externality

[4] https://en.wikipedia.org/wiki/Laissez-faire

Edit: I am moving fast and simplifying significantly, but I would be happy to address follow up questions/go into more detail if interested. But, as a disclaimer, I am far from an expert...I just have a limited understanding of some precious few Economic fundamental concepts.


> It's heavy reading, I don't know why it's getting upvoted here.

This is needed exposure for a lot of the libertarians that populate hacker news.


> It's heavy reading, I don't know why it's getting upvoted here.

Advocacy?


There's some stuff here: http://www.nationalreview.com/critical-condition/245118/gosp...

>... Arrow’s paper, which endorses the view that “the laissez-faire solution for medicine is intolerable,” is widely considered to have founded the field of health care economics, and continues to be celebrated, especially by those to the left of center.


That article gives a bit of a summary, but its own arguments are pretty weak - a lot of arguments about why Arrow is wrong, but no explanation whatsoever for the empirical observation that Arrow is right: just compare per-capita costs of health insurance with similar results between countries with mostly free-market systems like the USA with mostly single-payer/public systems like France.

http://voices.washingtonpost.com/ezra-klein/2010/01/america_...


The US is specially screwed up by its insurance system, which adds a lot of noise to the price signals.

Another thing that could drop the costs is the elimination or reduction of medical quotas and medical licenses. Those two are easier and smaller to solve than full welfare, which considering the us system, could be really expensive to implement.


If anyone wants an accessible summary, this is good:

http://hspm.sph.sc.edu/courses/econ/Classes/Arrow.html

This is the key paragraph:

" The basic problem is that, unless you know as much about medicine as your doctor, you can't evaluate the quality of advice he or she (mostly "he" in 1963) gives you. "




Applications are open for YC Winter 2022

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: