Information, Please

Why do we need a national health policy, any more than we need a national grocery policy or a national automobile policy or a national matchmaking policy?

Over on another recent thread, one of our commenters keeps pointing to allegedly unique “information issues” in the market for health care. So let’s see how unique those issues really are.

First, there are issues like adverse selection. The very fact that you’re buying insurance makes sellers suspect you’re sick, and they charge accordingly. Therefore if you’re not sick you overpay, and because you overpay you’re likely to underinsure.

That issue is not unique to health insurance. It also plagues the markets for car insurance and homeowner’s insurance, along with plenty of other markets. The very fact that you’re selling a used air conditioner makes buyers suspect there’s something wrong with it, and they lowball their offers accordingly. Therefore, you can’t get a good price even for a perfectly good air conditioner, and because you can’t get a good price you’re less likely to list it for sale in the first place. That’s exactly the same adverse selection problem (with buyers and sellers reversed), but there’s no general clamor for a national used-air-conditioner policy.

That’s not to say that adverse selection is unimportant, or that we shouldn’t try to address it, and it’s not deny that it might loom larger in some markets than others. But it’s far from unique to the market for health insurance.

Another information issue — one being flogged endlessly by a persistent commenter in that other thread — is that providers generally know a lot more than their customers do about the merits of various medical procedures. This is presented as if it were a reason for providers (e.g. doctors, insurance companies, or federal program administrators) to make key decisions, as opposed to presenting the customer with a price list from which to choose — the same method that seems to work perfectly well in restaurants, auto repair shops and lawyers’ offices.

But all of this overlooks the biggest information issue of all which is this: Only the customer knows whether he’d prefer, say, three weeks of pain relief to, say, a new car stereo, or whether he’d prefer, say, a slight lifelong reduction in heart attack risk to, say, an extra five restaurant meals every year.

The whole problem is to combine the specialized knowledge of the providers (what works, what doesn’t work, what works with various probabilities, what the alternatives are, what’s potentially life-threatening, what’s potentially painful, and how much so, etc. etc.) with the specialized knowledge of the customer (what risks and what pain levels are tolerable, whether a little pain relief matters more than a new washing machine, etc. etc.) and bring all of that information to bear on the problem.

This, too, is a problem that is not confined to insurance markets. When the radio stopped working in my car, I had no idea what the problem was or how to fix it. My mechanic was able to tell me that it might be the amplifier or it might be the head unit, gave me his best guess of the probabilities, explained the reasoning behind that guess, and gave me prices for the alternative procedures. My mechanic, by contrast, had no idea how much I cared about having a working radio. I could have tried putting all that in words for him, but I didn’t have to, because our conversation — plus the price list — gave me all the information I needed to make the decision myself. (And of course if I hadn’t trusted my mechanic I’d have sought out another one.) The same procedure works perfectly well when I purchase legal services, of which I am once again as ignorant as I am of auto mechanics or medicine.

This is exactly the sort of information problem that the price system, if it had been designed, would have been designed to solve. It’s the sort of information problem that the price system solves every single day. I’m quite unclear on why people are unwilling to let the price system work its magic in the market for health care.

It does no good to keep repeating the mantra that doctors know so much more than customers about medicine. That mantra is true, but no more true than the counter-mantra that customers know so much more than doctors about what they’re trying to accomplish. Customers, absent crucial information from their doctors, are totally incompetent to make medical decisions. Doctors, absent crucial information from their customers, are equally incompetent. The question is not who is ab initio competent to make these decisions?; it is how to we get information to flow to a decisionmaker? All human experience suggests that the only workable solution to that information problem is embodied in the price system.

I do, incidentally, think that there’s a good case for a national health care policy, for reasons beyond the scope of this post. But the goal of that policy should be to make things work better, not worse, and they can’t work better without prices. When I wake up with chest pains that are almost surely indigestion, and my doctor wants to do an electrocardiogram “just to be sure”, neither I nor my doctor is initially competent to decide whether the ECG is worth the cost. But I can be made competent a lot quicker than my doctor can. For him to make a wise decision, I’d have to acquaint him, for starters, with a rundown of my financial situation, the extent of my preference for restaurant meals, and the condition of my washing machine. For me to make a wise decision, ll I need is a brief discussion about probabilities and a price. That’s what any reasonable health care policy should make it easy for me to get.


58 Responses to “Information, Please”

  1. 1 1 Bennett Haselton

    Well, I’ve never thought this was the best argument for a national health policy. I think a simpler argument is that some people can’t afford the health care they need, and we have a moral obligation to provide it. (Although as you said previously, maybe this could be achieved with subsidized health savings accounts.)

    But since you mentioned that providers know more about the merits of various procedures than patients do, I’d also be fine with a regulation requiring providers to know (and tell you) the current available information about the probability of different outcomes for different procedures. If you ever ask a doctor directly about probabilities, they’ll just say, “I don’t know”, although of course they pronounce it, “It depends.” (Whenever someone says “It depends”, I just say, “Whatever it depends on, fill in some sample values for those variables, and then tell me the answer in that case,” in which case it usually comes out that what they meant to say was, “I don’t know.”) There’s no reason that the medical information can’t gather, and doctors can’t provide, information about all the times that procedure X has been carried out (anonymized, of course), and what usually happens in cases similar to mine.

  2. 2 2 The Original CC

    I think the main reason people want some sort of national health care is that the suppliers (doctors & hospitals) work with the gov’t to make sure that health care is scarce. (Why do I need a $150 doctor’s appointment to get a prescription for birth control when a pharmacist or nurse could probably ask me a couple questions and then tell me what to buy? And would the Rx really cost that much if more companies were allowed to manufacture generic drugs?)

    As Arnold Kling has pointed out, once we have policy in place to restrict supply, the gov’t then “helps” out by subsidizing demand. The result is much higher prices.

    This is why SL’s arguments fall on deaf ears. I’m not saying that people’s reasoning here is correct, but they read your stuff and say, “Yeah, makes sense except there’s no way on earth anyone can afford to pay for medical care out of pocket. It’s too darn expensive. It’s not like groceries at all.”

    You can try to explain to people how we got here, but people lose interest at that point, and they probably wouldn’t believe that there’s any way to go back anyway.

    Lord help us when the AMA starts fighting back against robot doctors…

  3. 3 3 Jimbino

    I don’t get the argument that people are driven to health insurance because they can’t afford to pay out of pocket. Obamacare, for example, requires insurance companies to maintain a “loss-ratio” of more than 80%. This means that a premium of 25% over the cash value is built into the cost of healthcare insurance.

    It would be more accurate to say that people are driven to pay out of pocket because they can’t afford to pay the high costs of healthcare insurance. The market helps these people, like me, who are then free to find domestic alternatives to health care or to shop as medical tourists beyond the USSA borders, where they can find prices for procedures and drugs posted on the web for most all afflictions and where they have a worldwide network.

    What makes refusing insurance difficult in the USSA is the successful effort by industry and gummint to hide prices of everything. I am in Rio de Janeiro, where I can search on “colonoscopia,” for example and find pricing of R$1500 (=$500) for an exam that costs a fortune in the USSA.

    While I complain about the forced high cost of health care in Amerika, I would really hate to see what Obamafood or Obamasex would cost.

  4. 4 4 iceman

    I am very much inclined to agree with this; I do wonder though if Arrow would?

  5. 5 5 Wherels17

    There’s another big downside to the opaque system that currently exists. I’m looking at a very major voluntary surgery, and can’t get good information on the quality of the surgeon that my GP recommended. In a more open market, we’d be able to get this information and make better decisions on the provider.

  6. 6 6 Tim F

    Insurance, health insurance or whatever, is supposed to protect an individual from large and unforeseen expenses. Mixing routine health care with health insurance is an expensive and unsustainable idea.

    A return to the major medical insurance concept would do much to improve healthcare consumer choices and to introduce price discipline to the healthcare marketplace.

  7. 7 7 Lee Hibbs

    Health care is 1/6 of our economy, but nobody wants to spend 1/6 of their income on it.

  8. 8 8 blink

    I love the line about “competence” versus “information” questions. This is an important point, well put!

    I think the ending, “For him to make a wise decision…” is too quick, however. For me to decide whether the ECG is worthwhile, I need to compare the price the doctor describes with my valuation of learning the outcome. This includes all of those ancillary comparisons, so it should also be possible for *me* to express *my* willingness to pay for the information. The doctor need only make the simple comparison of cost and reported benefit.

    Certainly I expect it to be easier for the doctor to determine the cost of the ECG than for me to quickly specify my valuation, but they are they same in principle; I do not need to divulge to the doctor my preferences over meals, etc., since *my* price will summarize the relevant trade-offs.

  9. 9 9 Advo


    granted, you get some inefficiency if you give people a heliobacter antibody titer instead of a lobster sandwich (assuming they prefer the latter), but how much does that matter?
    How big is the utility loss from overriding that preference?
    The medical test has utility (hopefully), so how high would you expect the utility loss to be typically? 10%? 20%?

    I must admit that I don’t have a lot of faith in people making utility maximizing decision on complex issues they know little about, so that’s the reason why I’ve been somewhat dismissive of this point.

    In addition, don’t you also sometimes dismiss subjective preferences?
    You advocate structuring the tax system in such a way as to discourage consumption and promote capital investment, or have I misunderstood that?

    I think I’m doing the same, aren’t I? I promote investment into health (= continued productivity, human capital) at the cost of consumption (also admittedly at the cost of other investments).

  10. 10 10 Roger

    I agree with most of what you say, but note two contrary views:

    (1) In discussions with friends, I have found that maybe half of them deny that they would ever go against medical advice for anything. Some say the same about lawyers, dentists, and other professionals. They trust those professionals to make certain decisions for them, and often they do not understand why anyone would want to make those decisions for themselves.

    (2) Some egalitarians deny that our society should allow using greater amounts of money to buy better medical care. They may concede that rich ppl can buy better cars, but medicine is somehow different, they say.

  11. 11 11 Zazooba

    I’m convinced.

  12. 12 12 Advo

    They trust those professionals to make certain decisions for them, and often they do not understand why anyone would want to make those decisions for themselves.

    For the average person, this represents a realistic assessment of their own abilities.

    I think it’s a better idea to properly structure incentives for the professionals, rather than force the amateurs to make decisions they’re ill-equipped for.

  13. 13 13 Advo


    Customers, absent crucial information from their doctors, are totally incompetent to make medical decisions. Doctors, absent crucial information from their customers, are equally incompetent. The question is not who is ab initio competent to make these decisions?; it is how to we get information to flow to a decisionmaker?

    The problem with that line of reasoning is that a big part – perhaps the most important part – of the cost-relevant decisions an individual is forced to make under an HSA system is the decision of whether or not to go to the doctor in the first place.

    A decision that, as you say, the individual is not equipped to make without help from a professional.

  14. 14 14 Steve Landsburg

    Roger: Allow me to suggest that those of your friends who “would never go against medical advice for anything” might rethink their positions in a hurry if they faced different prices for different procedures. The fact that people don’t bother to weigh these decisions is a terrible side effect of the fact that we’ve made costs invisible to the customer.

  15. 15 15 Steve Landsburg


    I must admit that I don’t have a lot of faith in people making utility maximizing decision on complex issues they know little about

    Yes, that’s what you say in about 50% of your comments. In the other 50%, you do have a lot of faith in people (doctors, insurance companies etc) making utility maximizing decisions on complex issues they know little about (such as the risk-tolerance or pain-tolerance of a given customer, or that customer’s financial situation, or whether that customer needs a new kitchen).

    Constant repeitition is not usually a good debating tactic. Constant repetition of one thing alternating with constant repetition of its exact opposite is generally even less effective.

  16. 16 16 Advo

    In the other 50%, you do have a lot of faith in people (doctors, insurance companies etc) making utility maximizing decisions on complex issues they know little about (such as the risk-tolerance or pain-tolerance of a given customer, or that customer’s financial situation, or whether that customer needs a new kitchen).

    That is only relevant insofar as it results in signficant utility losses.
    Let me repeat the question: how high do you suppose are the utility losses because of this is issue (as a percentage of healthcare spending)?

  17. 17 17 Teddi

    Thanks for clarifying, SL, it makes more sense.

    BTW, mind going into the reasons why you are warm to the idea of universal healthcare?

  18. 18 18 acarraro

    In equity markets, traders are compelled to show trades to all other parties. They are also required to disclose large positions and other material “proprietary” information. The market doesn’t just happen to work: there are a lot of rules that make it work properly. You could require doctors and hospital to provide similar transaction data: e.g. how much they charge for each procedure and what the outcomes are. I think in general they have strongly resisted this.

    I think this would greatly improve the healthcare system, but it’s unfortunately suboptimal for each participant to provide this information in a prisoner dilemma kind of situation: each player has an incentive to keep his own information secret.

    I personally don’t think health care consumption should fully be a function of personal income. I strongly support the UK system: any treatment which passes some kind of cost-benefit analysis should just be provided, with more expensive or unproved or marginal treatments being subject to personal resources. I think health savings accounts are problematic due to the distribution of health care spending: it can be very lumpy and erratic. Even situations one could consider predictable like maternity, have incredibily long tails, with expenses balloning if something goes wrong. The idea that someone would beed to buy maternity specific insurance to cover this tails seems troubling to me. It is a more or less direct child subsidy, but I am in general in favour of those to be honest. I generally see them as an investment.

    I also have another problem with the standard health case policies is that they are never true insurance: they don’t cover the full loss. If I get diagnosed with cancer or high blood pressure, the insurance will pay for immediate treatment. But in reality, my true loss is much larger: my expected lifetime healthcare spending has likely gone up. In many cases this second loss is massively higher than the cost of the immediate treatment, and yet I don’t think it’s possible to buy this kind of cover as an individual. A home insurance which paid out one year rent in case of accident wouldn’t really be attractive and yet that’s what health insurance mainly is…

  19. 19 19 iceman

    It’s not always that “complex” to save real $ when it’s *your* $. 2 examples:
    1) my dad was having recurring nosebleeds and a doctor referred him to a specialist. (Note my dad is a very frugal guy but actually made an appointment!) On the way out the lady at the little pharmacy store recommended a $2 bottle of saline spray which solved the problem.
    2) my dentist said I was grinding my teeth at night and recommended an occlusal splint for $200 (about half insured); I got a sports mouthguard for $2 at Target which she later said was just fine in my case.

    Roger 10 – of course #2 is an illusion right?; an example of how it seems “egalitarians” are happiest if everyone is equally deprived of innovations which just create pesky distributional problems. Never understood how a utilitarian could rationalize the much greater preventable death and suffering over time?

  20. 20 20 Daniel

    I have to say, if people were given menu options of probability and procedure costs then I think Steve’s idea would work, universal high deductible plans and subsidies for health savings accounts.

  21. 21 21 anonymous

    My two cents as a primary care physician who likes to read economics books written for laymen (including The Armchair Economist!):

    I would love to practice under this type of system. For a long time now, I have felt that my true strength of educating patients is not rewarded under our current system. In fact, I’m doing the patient a great disservice when I’m only allotted eight minutes per patient.

    I happen to think your proposal would work well for most primary care TREATMENTS, particularly medications that are used to reduce risk of future disease. Almost all worthwhile clinical trials calculate an absolute risk reduction, from which one can determine the “number needed to treat”. If the NNT for a particular medication is 10, then ten people would have to take it (for the time period covered by the study) to save one from that disease. No one can tell ahead of time who is going to be that one patient saved. I often tell patients this and let them make their own decisions.

    I don’t have enough experience with other specialties to know if this applies in the situations they may come across. Small studies of different chemo combinations for example might not have enough power to calculate these numbers accurately.

    The difficulty I see is in DIAGNOSTIC TESTS. The whole reason for doing them is because we don’t know what they are going to show, or which one of the battery of them is going to have the positive result. And yet it is human nature to feel that the negative ones were a waste of money.

    One thought exercise since no one has brought this up yet: If I recommend a test but a patient decides not to do it, and it is discovered later that this decision harmed him, will he file a lawsuit claiming I didn’t do enough to convince him to do the test? Notice the question is not whether he will win, since simply filing the suit is disrupting and damaging enough under our current system. Maybe this new proposal needs to be accompanied by specialized malpractice courts with a panel rather than a jury, or perhaps a loser-pays rule.

    Other worthy books on this topic that will stimulate discussion:
    Catastrophic Care by David Goldhill
    The Undercover Economist by Tim Harford (Chapter 5)

  22. 22 22 Ken B

    @Steve 15
    And in 12 he manages both in one short comment.

  23. 23 23 Ron

    Part of the problem is severely inflated list prices for hospital
    services if you don’t have insurance.

    >If you were told that the list price for the surgery, based on
    >something called a chargemaster, is about $14,000, but that the
    >hospital has agreed to do the same surgery (with anesthesia and
    >everything) for $5,600 for HMO patients, $4,700 for patients with
    >Blue Cross/Blue Shield, $3,500 for Medicare and $2,800 for Medicaid,
    >would you agree to pay the list price of $14,000? Of course not; if
    >you offered $6,000, the hospital would likely agree, and you would
    >save 50 percent.
    >But the hospital will not give you this information.–20140114_1_hospital-billing-patients-chargemaster

  24. 24 24 RichardR

    The reason why almost every western country (if not every) has a national health policy is because almost nobody could afford health care otherwise.

    With regards to demand, there is perhaps no universal human desire but almost everyone demands good health.

    There are virtually no certainties in life but it is certain that everyone will get old and die (unless they are unfortunate to die before getting old).

    When people are old and in need of health care, very few people would be wealthy enough to afford the health care they demand. Therefore, the state must provide health care for old people.

    Since the state must provide health care for old people, and old people are the main recipients of health care, it makes sense to also provide care for other people too since a large proportion of health care costs are fixed (such as a hospital).

  25. 25 25 nobody.really

    Five considerations that may have not yet been mentioned in this discussion.

    1. The US has a law that requires emergency rooms to treat all comers who arrive with a serious condition, regardless of ability to pay. This has proven to be a wildly expensive way to provide health care to the poor. Thus, Obamacare was adopted not within the context of a libertarian fantasy world, but within the context of a world already filled with cross-subsidies. It was designed to create a politically viable means to improve the efficiency of the world as it then existed.

    2. Adverse selection does not harm most people’s efforts to get health care; it harms only a minority. That’s because most people get health insurance through their employers. Insurers rely on the fact that employers will likely hire people on the basis of some criteria that does not correlate with poor health, and that large pools of such people will have favorable health statistics. Insurers will also extend insurance to the worker’s spouses and kids, based on the premise that workers are unlikely to select spouses and kids on the basis of some criteria that correlate with poor health.

    People who are not in these categories, though, are screwed.

    3. Tying healthcare to employment arguably impairs the efficiency of the labor market. In particular, it makes people reluctant to leave an existing position that has healthcare to one that does not—including leaving to start a new firm. So whatever inefficiency is created via health care policy is offset in whole or part by improved efficiency in the labor market.

    4. Landsburg values a price mechanism to ensure that consumers can maximize their utility by, for example, choosing to allocate resources toward things other than health care. Where Obamacare is concerned, I suggest that Landsburg has identified the wrong party as the consumer. Rather, the relevant consumers are the members of the public who want to live in a world where all their neighbors have health care so they can stop going to spaghetti dinner fundraisers to raise money for Tiffany’s chemotherapy. Now, maybe Tiffany’s parents would rather spend the money on auto repair. Too damn bad; I’ve voted to maximize MY utility, not theirs.

    5. People such as Charles Murray agree with Landsburg’s focus on a price mechanism and trade-offs, but also want to subsidize the poor. Murray favors a policy providing poor people (or all people?) with a guaranteed minimum income in lieu of our current collection of poverty-fighting programs, and letting people spend the funds as they see fit. But until they can marshal the political muscle to achieve that end, they’ll go with second-best alternatives, such as socialized health care. (Note: I don’t know of Murray’s views on health care specifically.)

  26. 26 26 Advo

    What the debate between Landsburg and me boils down to, I think is that I believe that the utility losses due to ignorance on the part of the patient under an HSA system are higher than the utility losses due ignorance on the part of the doctor under an insurance system, whereas Landsburg thinks it’s the other way around.

    Is there any way to reasonably quantify the utility losses or is this a debate that’s limited to “gut feeling”?

  27. 27 27 iceman

    nr 25 – re: your #1, I’ve long been skeptical that this is nearly as “wildly expensive” as commonly believed. I suspected it has much to do with financial (not social) accounting for facilities designed to have “peaking” capacity. Then I heard someone knowledgeable say a very similar thing. Perhaps someone else who is knowledgeable could educate me further.

  28. 28 28 Steve Landsburg


    What the debate between Landsburg and me boils down to, I think is that I believe that the utility losses due to ignorance on the part of the patient under an HSA system are higher than the utility losses due ignorance on the part of the doctor under an insurance system, whereas Landsburg thinks it’s the other way around.

    No, I don’t think you’ve boiled this down correctly at all.

    First (and less importantly) you keep saying “utility” where (I hope) you mean welfare. I’ll interpret accordingly. If you really mean utility, then there are a thousand other things we’d have to settle before we could move on to a meaningful discussion of health care policy.

    Second (and very importantly) you phrase this as if I have acknowledged that there are significant losses due to the patient’s ignorance. I acknowledge no such thing. It takes a great deal of medical expertise to know how Lipitor works at a molecular level. Essentially none of that knowledge is relevant to my decision about whether I want to take Lipitor. All I need for that is something like: Among patients of my age with similar medical histories, the probability of a heart attack in the next 10 years is x% with the Lipitor and y% without it, plus a few more numbers for the 5-year and 20-year forecasts, plus (if available) a few more numbers about ancillary benefits and about the probability of side effects. You might need a lot of education to come up with those numbers, but you need very little to make use of them once they are conveyed to you.

    So I maintain that your constant references and re-references to patients’ ignorance are red herrings. I have no idea how my car works, but I do know how much I’m willing to pay to avoid a 20% chance that I’ll stall on the highway. I have no idea how my home generator works, but I do know how much I’m willing to pay to have it run my furnace in the event of a power outage. I have no idea how Lipitor works, but I know enough about its track record to know that (at the price I’m currently charged) I want to take it. I maintain, then, that the “information problems” that you keep pointing to, as well as the information problems that I keep pointing to, are in no way unique to the health care industry.

    If we’ve decided that these problems are best dealt with through the price system for auto mechanics and home heating, there’s no apparent reason to reverse that decision when it comes to health care. And if you plan to respond that health care is uniquely important, then I pre-emptively counter-respond that it’s therefore all the more important to use the only system that ever works.

  29. 29 29 Carney

    I could probably get behind the mechanic comparison if mechanics were bound to provide a basic standard of care to any broken down car brought to them regardless of the owner’s ability to pay. Or if the average American owned the same car for 80 years rather than less than eight. Or if your car’s radio problem was contagious and could potentially infect my car. In that world, the cost of fixing your radio would be higher because the mechanic would have to cover the cost of fixing penniless jalopies. And your choice of a less expensive less effective treatment could impact all the cars yours comes in contact with, assuming your radio problem is Contagious Acute Radio Syndrome (CARS), which your mechanic would have to run a test for.

    All that added cost might have you scrambling for an extended warranty, but alas, your car was denied because of a pre-existing condition, a flaw dating back to when it was manufactored, where the rear passenger-side door handle sticks. It has little affect on your life and doesn’t impact any of the car’s other functions, but it still gets you denied when you try to get an extended warranty.

    You might think that all the extra car work would increase the need for mechanics and the extra mechanics would help keep the prices down, but the six-figure debt and ten years in mechanic’s school keeps their numbers low and the price of their services high. Which opens up plenty of room for charlatans offering snake oil changes. Which is how your car health market gets flooded with naturopath mechanics, alternative car healers, Christian Science Mechanics (who will pray away your car problems), and car-o-practors(who think that minor alterations to your car’s alignment will cure any car trouble).

    Those folks are especially fond of car health savings accounts because putting spending power in the hands of the owners pretty much ensures more money for them. Naturopath mechanics know that whether or not their treatment is effective won’t matter when they’re offering discount prices, and one only needs to look at the multi-billion dollar car supplements industry to see that owners will shell out cash for products with no proven efficacy. Those owners’ cars will eventually end up at a real mechanic, who will have to deal with worse problems because the owner first went to a quack, and the owner didn’t want to spend money from their CHSA on early screenings because a negative test would feel like wasted money.

  30. 30 30 Steve Landsburg

    Carney: Your comment is terrific, and gets at many of the reasons why I said in the post that there is a good case to be made for having some sort of national health care policy (though it’s not entirely clear why “national” is better than, e.g. “state-wide”).

    It also highlights the fact that the good arguments for having such a policy have nothing to do with the sort of “information issues” that Advo keeps talking about, which is exactly the point I’ve been trying to make.

    Thanks for posting this.

  31. 31 31 Steve Landsburg

    Carney: Re your last paragraph, why is it any more difficult to prohibit using HSAs for naturopaths than it is to write insurance policies that won’t pay for naturopaths? In any event, HSAs are not the only way to bring prices into this. Another is to have the insurance company pay the customer, not the provider, and then let the customer decide whether he wants to buy the service — as we do for car insurance. One problem with this, of course, is that we’d have a lot of false claims. The car insurance industry seems not to be crippled by this problem. I can see where the problem might be a lot harder to control with medical stuff (where you might be the only person who knows for sure that you get severe migraines that call for expensive medication) but that doesn’t mean it’s necessarily unmanageable.

    (Why are such policies not currently offered? Is that a market outcome, or are they regulated out of existence? Maybe some commenter knows…?)

  32. 32 32 Carney

    Steve Landsburg: I’m glad you liked my comment as I rather enjoyed writing it. As to your question regarding HSAs and naturopaths, I drew the connection because the biggest supporters of HSAs in Congress also tend to be big supporters of woo practitioners. Seems likely that any bills you get which push the former will also get you the latter.

  33. 33 33 nobody.really

    Sinaporeans pay for much of their own care out of their own pockets, and their major insurance program is designed to cover long-term illnesses and prolonged hospitalizations, not routine care…. The island state has excellent health outcomes while spending, as of 2014, just 5 percent of G.D.P. on health care. (By comparison … the United States spent 17 percent.)

    However, there has never been a major Republican policy proposal that just imitates what Singapore actually does….

    First …. [u]nder their Medisave program, they spend money saved in mandatory health-savings accounts, to which employers contribute as well. Second, their catastrophic insurance doesn’t come from a bevy of competing health insurance companies, but from a government-run single-payer system, MediShield. And then the government maintains a further safety net, Medifund, for patients who can’t cover their bills, while topping off Medisave accounts for poorer, older Singaporeans, and maintaining other supplemental programs as well.

    So the Singaporean structure does not necessarily minimize state involvement or redistribution. It minimizes direct public spending and third-party payments, while maximizing people’s exposure to what treatments actually cost. And the results are … extremely impressive….”

  34. 34 34 Ken B

    Interesting. Seems like a good approach, at a first look. It’s exactly the opposite of what single payer advocates seem to always insist on: “free at the point of delivery.”

  35. 35 35 Fred Anderson

    @RichardR 24
    You say, “The reason why almost every western country (if not every) has a national health policy is because almost nobody could afford health care otherwise.”

    I don’t understand. I’m assuming that “national health policy” is referring to “government health policy” and that “we, the people” are the “government” — at least in the western democracies. So at first glance, I think you’re saying, “we, the people, must have a national health policy because otherwise we, the people, cannot afford health care.”

    The most likely resolution is to note that key word, “almost” (as in “almost” nobody could afford it). Are you saying, “We must force the 1% to pay the health costs for the rest of us”? Since staving off death may be infinitely difficult (and therefore, infinitely expensive) and since none of us want to die, isn’t this a formula for loading infinite expenses onto that part of the population? And won’t they try to flee our predations?

    Alternatively, are we invoking “the government” to stop we lemmings from charging over the cliff, since this seems to be our desperate desire? If we will bankrupt our families, our neighbors, and (certainly) “that fellow behind that tree” in our desperate attempts to avoid death, are we asking the government to “build a wall” to prevent such access?

    curious thought: Did we, perhaps, elect Donald Trump precisely because he seems so inclined to build walls to prevent access?

  36. 36 36 David

    Loved your notion that you need to combine specialized knowledge of provider with the specialized knowledge (re: preferences, tolerance) of the customer. This observation is one of the most important reasons I launched a company in this area called Amino. We support price and quality transparency for every doctor and every hospital in America, publishing tens of millions of prices for everything from all surgeries to vaccines and diagnostic images. Our goal is not to make everyone a “price shopper” but to give each customer the information they need to make a confident decision. This is information we have always deserved. That is echoed very well in your positioning of the problem. Feedback welcome, by the way!

  37. 37 37 Neil

    To a large extent, price allocates goods according to ability to pay. We’d all love an ocean front home in South Maui, but most of us just can’t afford it. But health care is one of those goods many people think should not depend on ability to pay. Thus, price system rationing is considered inequitable (who should die simply because they are poor?), and we have a compelling poitical demand for universal coverage, which politicians can ignore at their own risk. I do not think it is any more complicated than that.

  38. 38 38 Dave

    I think you are both correct and incorrect when talking about the price mechanism as a filter.

    The ACA was supposed to be a safety net, but 30 million people were left out of the safety net, simply to save the insurance companies money. There isn’t anyone who argues that a single-payer plan would be more expensive, (it wouldn’t). Americans spend about $9400 per year on health care, and we get terrible results. Norway is a distant 4th at $6,600 per year.

    Singapore spends about $4,100 per year, and is considered to have the best health care system in the world. It is instructive to look at what Singapore does right to get the results they do. For one thing, they use cost as a filter. Everyone has some skin in the game, and no health care is ever free. But people’s health savings accounts are subsidized based on their income, so no one is ever bankrupted by a hernia operation. In addition, the savings accounts are managed in such a way that if to finish the year with money left over, you can keep some of it.

    When Democrats and Republican make speeches, they allude to some of these principles, as if they are going to propose some of them in the United States. But they never do. Most every time we hear arguments against single-payer health care, the arguments are based on what can be sold to Congress instead of what is best for people. Hillary Clinton and Howard Dean, for example, both opposed single-payer health care, because they thought it was “unworkable”. They didn’t mean the system, of course, the system works fine, they meant the idea of selling it to Congress with the insurance and pharmaceutical industries lined up against them.

    This is also why I don’t think the analogy works between health insurance and auto care. First off, most everyone can make an informed decision about auto care based on the mechanic’s estimate of costs, almost no one can make a similar decision about health care. But Second, imagine if all the auto companies were already making billions of dollars per year selling “auto repair insurance” which basically just spread out the cost of repairs, and added unregulated profit to the bottom line of the companies that offered it.

    They would have no motive to ever fix your radio, much less your transmission, and they certainly wouldn’t reveal to you how little it actually cost.

    The two easy solutions are diametrically opposed: We could have a single-payer system, or we could just make insurance illegal. Both solutions have some sound economic arguments, particularly if you can force people to shop around enough. But which one you accept depends on whether you are talking about cars or lives.

  39. 39 39 iceman

    33-34 re: Singapore, I’d just offer that there are surely things you can do with a 5.5 million highly homogeneous city-state population. Also probably helps if you’re not so big on due process.

    Carney – the “infectious disease” angle is a fair point for what it is, but is also why we have a CDC, and I imagine is a relatively small component of healthcare. I’d add that while it’s often suggested that single payor etc. is a “public health” issue, this is the only respect in which that’s literally true. (Sometimes people claim workplace productivity is directly impacted in a meaningful way but I find that hard to take seriously.)

    Dave – the only way I can see that single-payor *isn’t* more expensive – more demand but not supply — is if we restrict access to esp. newer procedures, *for everyone*. Surely the longer-term impact on innovation means much more unnecessary death and suffering. And I never understand why people say a solution is to ban private insurance. They usually point to the existence of profit, as if that’s some pure waste that’s blobbed onto a standard cost structure, rather than what’s created by learning how to do things better / cheaper. Of course if that reasoning were applied to just about any other area of the economy our focus would instead rightly be on what was hindering competition.

  40. 40 40 Don Boudreaux

    In one of his comments on this tread Steve L. asked why health insurance isn’t supplied more like car insurance. This question reminded me of this 1992 essay in Regulation by Suzie Feigenbaum, titled “‘Body Shop’ Economics: What’s Good for Our Cars May Be Good for Our Health”:

  41. 41 41 Floccina

    It reminds me of this excerpt from a discussion between Russ Roberts and Robin Hanson on econtalk:

    “Is health care different? Does government have to regulate it because we “need” it? Bigger choices have bigger consequences, but that doesn’t mean by itself that it should be regulated or subsidized. It’s different because people feel it is, but what actually is different is not obvious. Historically, medical spending was 1-2% of the economy at most. Recently up to 16%. Historically had plenty of doctors, but still people didn’t spend much. When you have an accident or condition, it’s argued that that’s when you’d spend the money–even everything you’ve got. But even then people don’t spend everything, don’t get every test or treatment. People do value health insurance but at the moment of choice, like in the emergency room, they often choose less than the largest choice possible. Opposite of usual intuition. Even in a situation of crisis, the fact that you’d be willing to pay a lot for it doesn’t mean that therefore markets don’t work. People really like food and would pay a lot for it, but they don’t have to because competitive supply of food keeps the price down. Same could be true of medical care if competitive supply of medical care were allowed. Health care is also claimed to be different because it is hard to evaluate the quality of health care. Because you don’t know how your own body works as well as someone else (a doctor) might, it’s hard to evaluate if a doctor or treatment is good. But you can get on the Internet; and just because you don’t know as much as an expert doesn’t mean you can’t be trusted to consume it. People vastly overestimate how much they know about most things they buy. We make do. If you are ignorant you do have to delegate to someone, but why government rather than private?”


  42. 42 42 Floccina

    I am going to reiterate a point that I made on the last post:

    “Medical care market might have an advantage over the auto mechanic market (also life and death BTW) in that your GP could act as go between for the consumer and if GP knew their customers where paying they might look for the monetary well being of there customers alone with there health.”

    If at least GP’s knew that most of their patients pay directly they would surely try to save there customers a buck when possible as it is they give to little thought to costs.

    As evidence I have be seen MD’s switch to a generic drug only after being told that insurance was not paying.

    BTW my proposal of what might work is here:

  43. 43 43 Floccina

    RichardR everyone with above median income and close to median healthcare spending will pay every penny for their own healthcare.

  44. 44 44 Mlanier

    Steve I think you state the adverse selection problem correctly, but draw the wrong conclusion. When someone sells a used airconditioner, the customer uses signalling to reduce the information disparity. For instance, they might call a friend who knows more, or they may google the air conditioner. They may even evaluate it via the seller (is the seller totally unknown? Does he live in a part of town that makes me think it could be a scam.) In general advertising exists primary to send a signal to customers that the items they are selling are of reputable condition. In fact, we have a national policy to reduce this information disparity- its called the BBB. Additionally we have the Consumer Protection Act.

    But in the case of medicine, the argument seems to be that expert opinion can never be validated or verified because the cost of validating or verifying can be death.

  45. 45 45 iceman

    Steve L, Don Boudreaux et al — I’ll repeat an earlier question I posted on this topic: given the recent lionization of Ken Arrow on this blog, do we think we overstated the case in his ‘seminal’ paper on healthcare economics and/or others have overstated the implications?

  46. 46 46 iceman

    “do we think *he* overstated…”

  47. 47 47 Nick

    You don’t mention that there’s a large amount of risk involved, and that the adverse selection problem is on the side of many consumers, rather than a few large producers. Rather than pricing each consumer’s insurance policy individually, with more and more intrusive statistical investigations into the precise nature of the risk each individual faces, we might just mandate health insurance and a single actuarial fair price.

    Many different things combine to make the health care market unique; that we can find these characteristics in other less regulated markets is not evidence that an unregulated market for health care will work. I doubt your flippant dismissal of the problems of adverse selection in the health care market is going to convince anyone who has thought carefully about this topic who disagrees. So who is this post written for?

  48. 48 48 Josh

    Is one main difference here between health markets and other markets the fact that none of theee other markets involve matters of life and death, at least not directly, whereas health care often is a matter of life and death? Someone who’s middle age who may still have another good 30 or 40 years of life would perhaps be willing to live out his remaining years in servitude to have life saving heart surgery or a robot heart. But in the past long ago didn’t I read you advocating (correct me if I’m wrong as I’m not certain I’m right) that we should provide insurance benefits to people who would or should have bought it but did not? There’s probably more subtlety to that that in forgetting but I swear I thought I remember you writing something like that.

  49. 49 49 iceman

    Nick – There’s no “single actuarial fair price”, that is a formula for a so-called death spiral (only offset by the coerciveness of your mandate I guess).

    I think you’re overstating your case. Policies aren’t underwritten that granularly, the best insurers can do is create pools of people presenting broadly similar risk factors (which is how “death spirals” are avoided). Of course within a pool people who don’t experience an adverse event end up “subsidizing” those who do, but in a way that is the essence of insurance, based on what is unknown and unpredictable. As opposed to subsidization that is “known” ex ante in an actuarial sense, which was the whole basis for the ACA.

    Arguably, knowing the insurance company knows some things about me should actually be conducive to forming pools, by reducing my concern that I’m getting “picked off” by people entering a pool who know they’re higher risk.

    I’ll throw in that while I get the “intrusive” sentiment, IMO it’s not really on the mark here; the fact is health insurance is a pretty intimate contract if you think about it – you’re asking someone to subsidize every aspect of your lifestyle. Of course easier to express those sentiments if you call it a “right”.

    All this said, the specter of genetic testing would seem to raise existential issues for the concept of insurance

  50. 50 50 Nick

    iceman – “only offset by the coerciveness of your mandate I guess”. Yes, of course. See the full sentence.

    The general point of my comment was not to advocate for a particular system, but rather to argue that the original post needs to do a lot more work to justify the flippant claim that the (massive) information problems endemic to the market for health insurance are similar to those in other markets, such as for air conditioners. It leaves out many of the issues and understates the extent of the problem.

    Smart people have been thinking about and researching this issue for a long time, this post contributes very little to any understanding of it. I’m not advocating for any side.

  51. 51 51 Bob Murphy


    This was a great post. However I am sure you benefit more from my nitpicking than my praise.

    You wrote: “I’m quite unclear on why people are unwilling to let the price system work its magic in the market for health care.”

    Ironically, earlier today I saw Justin Wolfers tell somebody on Twitter: “Take an economics class, and you’ll learn that private sector competition fails in health insurance markets.”

    And now I’m echoing iceman above: If you wonder what in the world could’ve made a professional economist say something like that, the answer is the guy you said was the greatest living economist.

    Obviously Arrow was amazing in certain respects, and I come up with excuses to do (Sen’s) proof of his Impossibility Theorem in as many settings as possible. But I think guys like that have huge blind spots when it comes to the actual operation of market economies. I would feel way better with you in charge of economic policy than Ken Arrow (even back when he was alive).

  52. 52 52 iceman

    Bob Murphy – well said. I’ve read Arrow’s paper on healthcare and wondered why that particular work was supposed to be so pathbreaking and brilliant, maybe because I can’t put myself back in the era it was written.

    Nick – OK it seemed like an endorsement, of a system you also seem to concede would be “actuarially unfair” (and coercive). Obviously this topic is so complex no blog post can assess it comprehensively; however I think Steve’s insight about the information problems working both ways — and which direction is probably the most tractable to work from — IS an important and, at least for me (and I suspect most people), original angle to contribute to the discussion.

  53. 53 53 Nick

    iceman – We tend to think that the market for health care insurance is unique in the magnitude and severity of the information problems. Death spirals are not uncommon if the industry fails to price risk appropriately, leading to significant losses in welfare.

    We’ve known since at least Akerlof in 1971 that information asymmetries exist, to some extent, in ALL markets. Observing this fact once again is fine; pretending it has some relevance to the current debate strikes me as singularly obtuse. The argument that the market for health insurance suffers from severe information asymmetries is not refuted by the observation that all markets suffer to some extent from information asymmetries.

  54. 54 54 iceman

    1) I think “death spiral” is overrated (see 49), requires a *systemic* reason insurers are unable to price (-adjust) accurately. Seems surest way is to forcibly over-broaden pools, which is the essence of the ACA.
    2) I agree merely pointing out info asymmetries in other markets is not groundbreaking stuff (but does shift the burden back to exactly why they’d be so much more severe and intractable in healthcare). The main insight of the post is that most people probably don’t think as much about of the reverse information asymmetry that Steve describes; and the fact that prices can be such effective assimilators of information should inform our conclusions as to the most effective way to resolve the decision problem.

  55. 55 55 Richard D.

    SL: “… your friends … might rethink their positions in a hurry
    if they faced different prices for different procedures. The fact
    that people don’t bother to weigh these decisions is a terrible
    side effect of the fact that we’ve made costs invisible to the customer.”

    Market failure, and reason for such?

  56. 56 56 Richard D.

    Lee Hibbs: “Health care is 1/6 of our economy, but nobody wants to spend 1/6 of their income on it.”

    However, if it’s 1/6 of the economy, they DO spend 1/6 of their income
    on it. But they’re blisslessly unaware of this fact – appropriate,
    considering the topic title –

    PS “Everybody wants to go to heaven, but …”

  57. 57 57 chad

    38. “Americans spend about $9400 per year on health care, and we get terrible results. Norway is a distant 4th at $6,600 per year.”

    This is an overly simplistic and naive point to make. Americans have single-payer K-12 education, and we STILL pay more per year on education than anyone else in the world, and get pretty terrible results. The idea that going to single payer automatically makes the situation in the US comparable to other countries is thus disproven by a single example.

  58. 58 58 iceman

    chad – excellent point which I wonder why isn’t made every time this subject is raised – yep it seems in primary education we also spend much more for results that do not hold up particularly well in intl comparisons. Yet the response from advocates of that ‘single-payor’ system is invariably to spend even *more*, while fighting tooth and nail *against* any attempt to evaluate and incentivize efficiency of outcomes.

    Of course as you suggest in healthcare it’s really kind of nonsense to pronounce overall results for the US — as if that can even be measured in some summary metric — as “terrible”, or even unambiguously worse. The reality is there are these pesky things called trade-offs, and we do better in some areas, especially those that require more high-tech spending (and less waiting for such services). As I’ve said repeatedly on these threads, that seems like precisely what one would expect if life-extending medical care is the ultimate luxury good. (Is there any reason to think it’s not?)

  1. 1 Some Links - Cafe Hayek
  2. 2 Some Links - Cafe Hayek
  3. 3 Centralized Cost-Benefit Analysis Fails the Cost-Benefit Test - Cafe Hayek
Comments are currently closed.