> consumers are constantly faced with a choice of "buy food" or "starve to death".

And funnily enough, food is heavily government-subsidized and regulated. It's not really a free market either.

Also starving to death isn't really the same level of urgency as dying of a heart attack. There's a couple orders of magnitudes difference in the amount of time available to make a purchase decision, and the level of physical and mental stress you're under while making that decision.

> And funnily enough, food is heavily government-subsidized and regulated. It's not really a free market either.

Sure, but by this metric, there exists no free market on the planet. If your argument is that the healthcare market should look exactly like the food market does today (with subsidies and FDA regulations), most adherents of private healthcare would agree with you.

Privatization != "no welfare".

> Also starving to death isn't really the same level of urgency as dying of a heart attack. There's a couple orders of magnitudes difference in the amount of time available to make a purchase decision, and the level of physical and mental stress you're under while making that decision.

Okay, but not all healthcare is "dying of a heart attack". Obviously it's impossible to shop around for healthcare when you're having a heart attack, and that's exactly what insurance is for. After EMTALA, emergency care is free if you can't pay for it. This isn't really controversial. What's controversial is whether the same framework needs to be applied for planned care, like MRIs, colonoscopies, annual physicals, tonsillectomies, vaccines, prescription drugs, etc.

Different types of food are fungible goods. If broccoli is expensive I can instead purchase carrots and still survive. If I think chemotherapy is too pricey, I can't take an aspirin instead to treat the cancer (at least not with the same prognosis).

This is very true, and not at all what I mean when I say "choice". In free market choice means alternative vendors. Do you have other choices for getting chemo besides the one hospital?

You may be surprised to know that 80% of hospitals in the US are non-profit. The evil profit motive isn't the reason their bills are so outrages.

In the 60's US healthcare expenditure was 5% of the economy and it is close to 20% right now.

I'm curious (you don't come out one way or the other on this from your comment, but this will help me make sense of where you're coming from) do you think there are any domains where a free market doesn't work? Or is your position that free markets work universally (or less absolutely, that free markets work for almost all practical purposes and fail only in very artificial environments).

Free markets don't work when the cost of the transaction isn't paid by the two sides of the transaction but by a third party. For example the environment where I can buy a gas guzzler from you and drive away polluting the environment. There you need intervention but the framework of intervention and has to meet some specific criteria. I don't want to digress.

Other than that I have yet to see a problem that isn't solved by this scheme: increase the set of choices available to the person, and in some corner cases give them money so they are free to choose.

You have to bear in mind that the alternative to the free market approach is for someone else to come in an constrain either the buyer and seller in some way. Ths may work for a limited time and for a specific set of buyers and sellers but it won't work beyond that. Given that people have diverse and evolving needs the forced solution causes long term harm. Then you'll need some kind of propoganda machine to either exagerate the good or down play the harm.

To be clear I do not believe that we have a free market in the US in a lot of areas and what people conceive of as free market -- or rather what has been shoved down their throat as free market is anything but that.

I think you and I probably would have different notions of what it means to "solve a problem." In this case I suspect you have a notion that a free market in many ways nicely sidesteps needing to even resolve this question in the first place. It frees one from having to commit to overarching, centralized value systems and instead allows for gestalt value systems to arise naturally from the behavior of people. This nicely avoids the issue of needing to impose a higher authority's will on a population and all the authoritarianism that that entails as well as the inevitable schism between a centralized value system and what people actually want. (I happen to disagree with this take and can expand on why if you're curious, but if this accurately reflects your views, there's enough commonality at least for me to make the next point.)

> increase the set of choices available to the person

This view abstracts behavior into that which is governed by "choice" and "coercion." I think this binary distinction is a fine model for a lot of domains, but a poor one for healthcare.

Choice feels much more like a spectrum in the domain of healthcare than it does in other domains. For a rough sample of points along this spectrum, you have "do this or die immediately," "do this or die in the next several months," "do this or suffer permanent disability," "do this or suffer great pain," "do this or suffer some probability of some amount of disability," "do this or suffer mild discomfort," "do this or be slightly annoyed."

The far-"left" part of this spectrum cannot ever realistically expand its set of choices. The most extreme version of this is that you're literally incapacitated and so can never make a choice of e.g. what hospital to go to and what treatment to administer no matter how many hospitals or treatments exist.

However, I think the same problem persists in less extreme states as well. Health ailments can directly impact a person's ability to choose to begin with in a variety of ways apart from just physical or mental incapacitation or degradation. Various treatments and healthcare choices impose switching costs that reduce a person's choice even when they are nominally capable of making one. For example, if a patient chooses a single hospital for a bout of appendicitis (when they are in such pain that they cannot make a choice in that moment other than to dial 911), even once the acute problem of surgery passes, they are unlikely to be able to choose a separate hospital for their post-surgery hospital stay without jeopardizing their health due to movement and continuity of care concerns.

Even in non-emergency cases there is an extreme information asymmetry and unpredictable path dependence (certain choices lock into other choices down the line but the nature of how they lock in may not be apparent at the beginning) that make it hard to formulate what "choice" would even look like.

In some ways, I personally view the need for coercion in the healthcare space as precisely a way to return to a world where modelling things as a binary distinction of "choice" vs "coercion" makes sense again.

Any plan for regulation of healthcare always must deal with a distinction between "elective" and "necessary," "non-essential" and "essential," "covered" and "not covered." That line is drawn precisely where we have a best guess that the model of a binary "choice" vs "coercion" holds vs the model of a spectrum of choice; the ideal is that care is provided to boost a patient back into a universe where the binary "choice" model is a reasonably good approximation.

More generally there is the problem that healthcare has a weird squeeze of monopolistic and non-monopolistic needs.

At a base level, in almost all domains including healthcare you need some amount of a regulatory framework to counteract the problem that market participants generally have an incentive to decrease the number of choices to the other side. I think you probably agree with "coercion" at this level (stuff like preventing collusion among players, certain stances on breaking up certain kinds of monopolies, etc.).

But the problem is that in healthcare you do want powerful players because there are benefits we want to reap from large players. Large drug makers are the only ones capable of performing substantial R&D and regional hospital and transportation networks are really the only ways you can get the necessary infrastructure and expertise to treat a lot of things. On the buyer side you want large insurance pools to even out risk for people.

But those all have inherent monopolistic tendencies that are exacerbated by the problems of choice that I mentioned.

Different types of healthcare are also fungible. And not all healthcare is cancer! I think the biggest problem with having any discussion around healthcare policy is that we automatically assume that we should treat routine treatments and visits the same way we treat catastrophic accidents like cancer and brain surgery.

The vast majority of healthcare expenditure is preventive or planned care, which is largely fungible. MRIs are fungible. Primary care is fungible. Antibiotics are fungible.

To the extent that healthcare isn't fungible, it's in very specific cases like end-of-life care, cancer, catastrophic surgery, and rare patented drugs. They also account for a tiny minority of overall health expenditure.

We can use different tools across both of those problems.

> The vast majority of healthcare expenditure is preventive or planned care

I don't think that's true for the U.S. Preventive care and planned care (if understood to be stuff like physicals, blood checks, screening, etc. including your examples of MRIs and primary care) as far as I remember is actually a small minority of healthcare expenditures (< 20% is a number I recall). I can try to root around for sources if you're curious, but I'm also curious where you're getting the impression of "vast majority."

> Antibiotics are fungible.

Not really. Definitely not in the same way that food is fungible. I assume you're talking about generics here? But generics again actually make up a startlingly small minority of healthcare expenditure costs despite making up the majority of prescriptions IIRC (again I'm going off memory but I think it was something like 75% of medication expenditures are due to medicines with no allowed generic alternatives).

Basically the places that you're suggesting the free market should best apply to are already the smallest slices of the healthcare expenditure pie (and also already quite effective in that limited domain).

I think he includes things like hip replacements which are "elective" and not of the "pay now or be dead" kind.

Ah, I don't ever remember reading numbers for those so I can't comment on that (I'd be curious if anyone has a breakdown of surgery costs by elective, semi-elective, and emergency).

But even stuff like hip replacement kind of is on a sliding scale. How much choice do you have if the alternative is death? What about cognitive impairment? What about blindness? What about impaired range of motion? What about mild discomfort? What about pure annoyance?