Our failure to find a non-insurance 'hammer' with which to hit all of the health 'nails' is a larger issue.

You can offer insurance against an event that has already occurred, but the insured price will always be slightly higher than the uninsured price. One could call that insurance, but it'd be a degenerate case, so at that point it's more of a word puzzle than a real question.

> Our failure to find a non-insurance 'hammer' with which to hit all of the health 'nails' is a larger issue.

Exactly - the problem is that people talk about health insurance like it's supposed to solve the problems that a wealth redistribution program would. Except, insurance is not a wealth redistribution mechanism - it has a completely different goal - and trying to turn it into one just results into the worst of all worlds (expensive and ineffective at achieving either goal).

Pretend for a moment that everyone receives their annual physical, as medical guidelines recommend. (They don't, but it makes our example simpler.) And let's say that the fair-market price of providing the physical, accounting for all costs borne by the provider and their practice, is $100. (That is an arbitrary number I have chosen, also to make our lives easier). What will be the co-pay for the annual physical for an insured patient?

The answer is that it will be $100 - there is absolutely no risk involved in this situation, so the expected payouts of the insurance company will be $100, and therefore they will incorporate that into their price. (The consumer will actually pay a bit more than $100 in total, because the insurance company has overhead costs, which are ultimately paid by the consumers as well). But of course, that's not the case, because the expectation is that health insurance will reduce these costs, and that people who can't necessarily afford $100 will still be able to have their physical. That's why health insurance isn't really insurance, except in name - we talk about it as insurance, but in reality, it's a wealth redistribution program tacked onto a risk smoothing product.

By definition, insurance is literally not intended to save the insured person money, in expectation. The expected value of all claims will always be less than the expected value of all money paid to the insurer by the insured entity. (This does not hold for every individual, but it does hold in the aggregate - that's where the risk smoothing comes in). The insured person pays the insurer a premium[0] in order to reduce the uncertainty in how much they would have to pay on any given month without insurance.

[0] Not as in "monthly premium", but as in "a premium on top of the expected value"

You fundamentally misunderstand the concept of insurance. Group insurance is, by definition, a way to spread the cost of rare catastrophes around the group so that the affected individuals don't bear the full brunt. There is the full expectation that in the event of a major medical event you will save money.

Try reworking your example around major medical events (e.g. a $500K hospitalization) rather than preventive care (the function of which is to reduce the risk of certain controllable medical events).

> The answer is that it will be $100 - there is absolutely no risk involved in this situation, so the expected payouts of the insurance company will be $100, and therefore they will incorporate that into their price.

No. Even pre-ACA, insurance companies offered low co-pays. How did they do that? Because your monthly fees will over the course of a year add up to far more than the cost of an annual physical. The purpose of insurance is to protect yourself against rare, but catastrophic events: you will probably not experience a wide variety of expensive medical ailments, but if you do they will likely leave you in financial ruin if you're uninsured, so you pay an insurance company money to protect yourself against that risk.

(Or, alternatively, your employer pays a health insurance company money to protect you against that risk, and offers that benefit to you as part of your total compensation package. Which is essentially the same as you paying for it, with some amount of risk differences and thus potentially lower costs due to the pooled employee health insurance policies, but that's outside of the scope of this discussion. TL;DR: it's still regular market economics.)

You might wonder why insurance companies offered low co-pays at all — was it just some marketing gimmick? But no, you can explain that with regular economics too: insurance companies are incentivized to make annual physicals affordable and attractive, because they can catch potentially-expensive medical issues when they're still much less expensive, thus lowering costs for the insurance company.

Yes, this is what I am saying. However, that has nothing to do with the price of co-pays for routine care, which is by definition predictable.

> But no, you can explain that with regular economics too: insurance companies are incentivized to make annual physicals affordable and attractive, because they can catch potentially-expensive medical issues when they're still much less expensive, thus lowering costs for the insurance company.

See, this is another pervasive myth. For healthy individuals, the annual physical is not cost-effective - it is very unlikely to result in long-term benefits to the patient, and it is far more likely to result in unnecessary care (such as follow-up tests and differential diagnoses for false positives): https://sciencebasedmedicine.org/re-thinking-the-annual-phys...

Also, I don't know why you're drawing a dichotomy between what I'm saying and "regular market economics". Everything I have said is standard, textbook economic theory. There's nothing obscure or even controversial about it about economists.

> Everything I have said is standard, textbook economic theory. There's nothing obscure or even controversial about it about economists.

Claiming that "health insurance isn't really insurance, except in name - we talk about it as insurance, but in reality, it's a wealth redistribution program tacked onto a risk smoothing product" is definitely non-mainstream. An insurance company takes in payments to insure you against high-risk events, and that's exactly what a health insurance company does: you pay them a monthly fee regardless of whether you need medical care (or your employer pays them for you as part of your compensation package), and when you do need expensive medical care, they pay for it. In fact, many plans explicitly only pay for expensive medical care: anything under your deductible, aka inexpensive medical care, you're required to pay for. You might be able to make some sort of weaker claim about wealth redistribution post-ACA, but — co-payments existed pre-ACA.

Actually, some of the largest insurance companies and the vast majority of risk-bearing providers (hospitals that act as insurers) are non-profits. Though I never said that their pricing was based on charity; I said it was based on the fact that the product they offer is not really insurance (and that it exists in a marketplace in which prices are incredibly distorted by the existence of other factors, which would have been too long to explain in that simple example).

> Claiming that "health insurance isn't really insurance, except in name"... is definitely non-mainstream

On the contrary, that's the overwhelming consensus among economists. Which isn't surprising, because the economics of insurance are generally covered even at the undergraduate level, and the ways in which health insurance differs from a true insurance program are pretty glaring.

Equating risk-bearing providers and health insurance companies for the purposes of arguing that health insurance isn't insurance might as well be tautological. Sure, blueberry pancakes aren't really pancakes if you include blueberry muffins in the category of blueberry pancakes, but who cares?

It also ignores a number of other selection criteria and behavioral issues, which you are honest enough to note in your pretend for a moment intros.

However, people who proactively care for their health carry "upside risk" as well as downside, which your scenario does not account for.

And they screwed that part up. As a healthy young person, there is no incentive to buy massively overpriced, shitty high-deductible ACA plans, when one month of coverage costs more than the IRS penalty.

True. And the ACA now levies a tax penalty for those that don't sign up for insurance (2.5% of income, max ~$2K).[1]

[1] https://www.nerdwallet.com/blog/health/how-much-is-the-obama...

What are you going to do? Not let me die?

That is not insurance. That is a wealth redistribution. Those are not the same thing, although the current discourse around health insurance conflates those two concepts, for obvious political reasons.

> Could you imagine a system in which car insurance wasn't mandated but car insurance companies were required to repair your car regardless of its preexisting condition? No smart person would preemptively buy insurance. You just wait until your get into an accident and then bring the already damaged car to the insurance company and demand a policy to repair it. No insurance company can work like that.

Actually, this would be completely feasible to implement. The insurance company would price you based on your risk. In this case, P(risky event | all available information) = 1), so there wouldn't be much of a point to buying the insurance, but it absolutely would work.

The problem is that the ACA also forbids insurers from underwriting plans based on anything other than age, income, zip code, and whether or not they smoke. That's guaranteed to increase the cost both for healthy and for unhealthy patients, because they have to make overly conservative estimates when evaluating the risk level of their patient pool.

Yes, they are the same thing. Insurance is fundamentally a redistribution of risk with money being shifted from those who have not fallen victim to that risk to those who have. If you have a problem with that aspect of the ACA, you have a problem with the general idea of insurance.

>The insurance company would price you based on your risk (in this case, it would be 100%), so there wouldn't be much of a point to buying the insurance, but it absolutely would work.

You do realize that this is functionally the same as getting rid of the preexisting conditions protection, right? Whether someone pays tens of thousands to a doctor for care or to an insurance company for coverage are in practice exactly the same. The whole point is that many of us think that is a fundamentally unfair system to force someone to face in that situation.

This comment is so profoundly misguided that I have to comment.

Insurance does not work the way you describe. When companies do a fundraiser where someone gets to take a half-court shot with a basketball and win $1M if they make the shot, an insurance policy sells the company doing the fundraiser a policy that reflects the odds that a person chosen at random from the audience will sink the shot. That policy might cost $7K, since the shot will very likely be missed.

The insurance company makes a profit by charging a bit more than the actual odds reflect, so that over time if 200 shots are taken, it pays the $1M once and profits $400K. In a competitive market, the price of insurance will approach the probability.

Similarly, an insurance company might offer insurance that it will not rain on the last weekend in July. Perhaps an outdoor wedding facility wishes to buy that policy, but a farmer wishes to buy the other side of that risk. In such cases, the insurance company can charge less because there is a market for both sides of the uncertain event. Futures markets are also used for this purpose.

Most of our modern health care is not really risk-driven, it's based on markets that are highly regulated and prices that are influenced by lobbyists from various industries etc.

The key problem with your assertion is that at the time insurance is purchased, nobody knows who will be the victim or whether there will be a victim. Purchasers of insurance would rather spend a little bit of money just in case a bad outcome occurs, so they don't bear the full brunt of that bad outcome. Those who don't end up with a bad outcome don't get their money back, which is why the system works.

We all know there is a need for social services to provide healthcare for those who can't afford it or who have really bad luck. That's not insurance, however, it's social services.

You say that my comment was profoundly misguided, but your comment right here is just a rephrasing of mine Someone with risk spends a little bit of money to help absolve themselves of that risk. If a bad outcome occurs and they fall victim to that risk, they don't bear the full brunt of that bad outcome because money is shifted from people who didn't fall victim to that risk.

Regarding you point about probability based pricing of insurance, we have collectively decided that we don't want health insurance to function that way. If it did, it would lead to the preexisting condition problem in which a person cannot afford insurance because they have a condition that requires expensive care. We instead subsidize their policy with a price increase on everyone else's policy.

> we have collectively decided that we don't want health insurance to function that way.

I agree this is the case, the problem is that it's not really insurance any more it's a bundle of insurance, prepayment, etc.

No, they are not the same thing. The fact that a transfer of money happens is not a sufficient criterion for defining insurance.

Take two people with different risk profiles but who are both insured. If you can tell a priori which one is expected to have lifetime claims that exceed their lifetime premiums, then you don't have insurance - you have a wealth redistribution scheme[0].

Note that I didn't specify which person had the greater risk profile, or whether they both purchased the same "tier" of plan, or even whether they purchased their insurance from the same insurer. This property of insurance still holds even if the two people have completely different risk profiles, if one purchases a gold plan and the other a bronze, and if one person purchases from MegaInsurance in New York and the other purchases from AcmeInsurance in California - as long as they are both insured at risk-adjusted rates.

> You do realize that this is functionally the same as getting rid of the preexisting conditions protection, right? Whether someone pays tens of thousands to a doctor for care or to an insurance company for coverage are in practice exactly the same. The whole point is that many of us think that is a fundamentally unfair system to force someone to face in that situation.

First, nobody is paying tens of thousands of dollars to a doctor, because there's an out-of-pocket maximum cap. (And that cap could still exist under a risk-based pricing world.)

Second, it's not, functionally the same, because that doesn't mean that you can't separately provide income- or wealth-based subsidies if you're aiming to redistribute wealth. But that happens at a completely different layer from the risk underwriting - and because the underwriting process is allowed to properly account for a person's risk profile, you end up with lower aggregate premiums (pre-subsidy). Lower unsubsidized premiums means that you don't need to subsidize as much money in order to achieve the same sticker-price premiums that consumers see - in other words, the entire process is significantly cheaper for what appears to be the same result to the patient.

The reason we don't do this, even though it would be significantly cheaper, is because it's politically infeasible.

[0] Which, you may note, is currently the case - and that's because health insurance as it stands is a mishmash of two completely unrelated products ("insurance" and "wealth redistribution") that we happen to try to stuff into the same box.

>The reason we don't do this, even though it would be significantly cheaper, is because it's politically infeasible.

I agree with your point here, but it is the whole perfect being the enemy of the good thing. Like the original article states, this system isn't perfect. Even Obama admits this. I do however believe the current system is unquestionably better than the system we had previously. I therefore think it is a bad idea to return to the previous system while we hope to eventually come up with a better one. It is not hyperbolic to say lives literally depend on it.

Probably more than anyone, Obama understands that politics is the art of the possible. The fact that the ACA was passed by a margin of one vote is some evidence that, in the face of raging blind opposition, they didn't leave anything on the table. If congress weren't in thrall to gerrymandered hyperpartisanism and effectively unlimited donor money, the law would have had at least some bipartisan support, been revised numerous times, had bad parts improved, good parts enhanced, and concerns of both parties and various constituencies addressed in light of the empirical evidence gathered over the last seven years. I don't think there will be a return to the previous system. Instead, for several million Americans, they'll try to demolish the first four floors of a building and disingenuously point to the doctor's office on the fifth.

If they have insurance. Without guaranteed issue they may not. With guaranteed issue they may have "access" to insurance but can't afford it, or the insurance they can afford may exclude the procedure they need. Retroactively. And even with insurance they may pay 25K a year in premiums plus that 10K max. Then 28K and 15K the next year... It's a system designed for optimal profit, not efficient (or moral) distribution of resources that every person will require. Debating whether it's insurance or insurance-like, or how the underwriting works, begs the question. Insurance companies should not be involved.

It sounds like you are trying to respond to a different sort of discussion altogether.

This whole subthread is in reference to the (implied) statement "requiring insurers to cover pre-existing conditions requires a mandate [and it will necessarily increase premiums to the extent that we have seen]"

Your responses is tangential to that, addressing either (a) what would happen if we didn't require insurers to cover pre-existing conditions, or (b) other potential failure modes which could potentially occur, and which already occur under the ACA.