• Simply put: Moral hazard

    Simply put” is an ongoing series. See the introduction for an explanation of the series and the full list of topics that have been or may be covered. Feel free to suggest other topics in that post.

    We buy all things that are worth more to us than we spend (of our own money) to obtain them.

    When someone else is paying the bill (or some of it), we’ll buy even more stuff because the amount of our own money we must spend to obtain it is even further below the value of the stuff. This is intuitive. The less something costs us, the more we’ll buy.

    That’s why we use more health care when we’re insured. Some people call this “moral hazard,” but that’s not quite right. I’ll explain.

    In the background, no matter who is paying for health care, we each have a valuation of it, a measure of its worth in dollars. That is, for any health care service, there is an amount of our own money we’d spend for it and no more. At some price we’d rather have the cash.

    If it weren’t so impractical health insurance (public or private) could function this way. For every health event the insurance company (or government program) could give us a wad of cash. Then we could decide if the health care we could buy for that amount was worth it. We could decide to keep the cash. There is some amount of money for which you are indifferent to having the cash or having your sprained ankle examined. That’s your valuation.

    The way insurance actually works is that we either see the doctor or get nothing (no cash). If we see the doctor our insurer pays something to the doctor for it, and we pay a little too (a copayment–I’m ignoring deductibles for the moment). Is the insurer spending more than the amount we’d accept to not see the doctor? That is, is the insurer paying a price above our valuation of the health care we receive?

    If so, that’s a “moral hazard” effect. We’re using care that costs more than we think it is worth. That’s what moral hazard means, when applied to health care. In some sense the insurer is getting ripped off. It could have made us happy by spending less, giving us an amount of cash that is one penny above what we think the health care they paid much more for is worth. This reflects an information asymmetry–something we know that the insurer doesn’t. It doesn’t know our private valuation.

    Very often people say that all care that is used by the insured that wouldn’t otherwise be used in the absence of insurance is “moral hazard.” But that’s not right. That would imply that care you couldn’t afford without insurance but think is a great value is moral hazard care. It means that nearly all the care that people with low incomes on Medicaid receive is moral hazard care, yet people who are wealthy who receive the same procedures through private insurance (that they could afford anyway) is not moral hazard care.

    In other words, moral hazard is sometimes confused with an income effect and income transfers (or the effect of in-kind transfers). It really has nothing to do with it. Moral hazard is really about the insurer (public or private) overpaying for care because it lacks the information to do otherwise.

    Another way “moral hazard” is abused is by conflating this “overpayment effect” with “care that doesn’t improve health.” The two concepts are totally unrelated. There may be, and likely are, health improvements derived from care that costs more than you or I would pay out-of-pocket. Just because I’d forgo the care for a wad of cash lower in value than the cost of that care doesn’t mean the care isn’t good for my health. Not all of it may be, but some of it likely is.

    The real problem as it pertains to health is not “moral hazard” per se, but distinguishing between care that is health improving and care that is not. Even if moral hazard did not exist–because the insurer paid exactly what we would out-of-pocket for the care–that problem would remain.

    Further Reading

    Newhouse J. Pricing the Priceless: A Health Care Conundrum. Chapter 1.

    Nyman J. The Theory of Demand for Health Insurance.

    Santerre R and Neun S. Health Economics: Theories, Insights, and Industries Studies. Chapter 6.

    • I know that people often assert that if you give everyone insurance then they will consume a lot more health services and the system will go broke. While this may be true for candy and electronic gadgets if they were provided ‘free’, I don’t think it applies to health care.
      Consider automobile insurance (which is required in most states). Are you more likely to invent a reason to use your auto insurance just because you have paid for it? Are you more likely to get into an accident just so you can ‘use’ your auto insurance? Do you dent your car on purpose just so you can make a claim?
      Consuming health services has costs such as time, inconvenience, and pain in addition to the monetary costs of co-pays, etc. I don’t know anyone who makes up a disease and goes to the doctor for no reason at all. (Munchausen syndrome is a disease.) I don’t see any actual moral hazard in health insurance.

      The opposite is true. People do avoid going to the doctor when they are sick if they don’t have insurance… but this is not a good thing and it is something that we all should be working to avoid.

      • @Mark Spohr – But didn’t the RAND HIE show that people use more health services if it costs less? A lot of other studies show it too. I think this is a pretty believable and robust finding. Even if I didn’t want to believe it, how would I explain the research that shows it?

    • People probably do use more health services if they cost less. The real question is: “Are they making unnecessary use of services?”
      I believe people should have good access to needed care. I don’t believe that people abuse this by consuming unnecessary care (the moral hazard bit).
      Do you have research which shows that people are consuming unnecessary care (other than doctors ordering unnecessary tests and procedures which is a different form of moral hazard)?

    • Mark
      People dont buy health insurance to get diabetes, just like they dont buy auto insurance to get into accidents.

      However, given the choice between the trade vs generic–if you are not paying versus, say, replacing the door after a ding or using paintless dent repair if it is on Geico’s nickel–well, you know what path I am going down. Moral hazard is moral hazard.


    • Brad,
      You seem to be saying that a patient choosing a more expensive trade name drug over a generic is acting on moral hazard. I think in this case the patient is choosing higher quality when the option is available to them. This is not irrational or immoral. It is also not unnecessary care.

      I am still looking for evidence that patients seek out and consume unnecessary care to illustrate moral hazard. I just don’t think this exists.

    • Mark
      I disagree.

      the literature is replete with similar studies.

      If you are not convinced on the Rx front, lets offer the same auto owner the opportunity to get a new door after a bang up–except the choice is after-market vs original manufacturer replacement. If no added expense, the consumer takes the original–however, if they pay OOP, they ask for the after market. Thats moral hazard.

      If we disagree on that, no problem, I can live with it.


    • If you are looking for moral hazard in the health care system, I think the best candidate would be doctors performing procedures and tests which are not beneficial to patients but which increase the doctors income at the expense of the insurance companies. Patients are innocent victims in this scam (and they often are real victims of procedures which cause harm and no benefit).

      I came across this passage from George Bernard Shaw (The Doctors Dilemma – 1909) which I think still applies to this situation:

      “It is not the fault of our doctors that the medical service of the community, as at present provided for, is a murderous absurdity. That any sane nation, having observed that you could provide for the supply of bread by giving bakers a pecuniary interest in baking for you, should go on to give a surgeon a pecuniary interest in cutting off your leg, is enough to make one despair of political humanity. But that is precisely what we have done. And the more appalling the mutilation, the more the mutilator is paid. He who corrects the ingrowing toe-nail receives a few shillings: he who cuts your inside out receives hundreds of guineas, except when he does it to a poor person for practice.

      Scandalized voices murmur that these operations are necessary. They may be. It may also be necessary to hang a man or pull down a house. But we take good care not to make the hangman and the housebreaker the judges of that. If we did, no man’s neck would be safe and no man’s house stable. But we do make the doctor the judge… I cannot knock my shins severely without forcing on some surgeon the difficult question, “Could I not make a better use of a pocketful of guineas than this man is making of his leg? Could he not write as well—or even better—on one leg than on two?”

      – Thanks to Paul Levy:

      • @Mark Spohr – Physician-induced demand is not moral hazard necessarily. It’s physician-patient and/or physician-insurer information asymmetry. I’ll discuss those in subsequent posts. There are many reasons for increased utilization and cost. Moral hazard has a precise definition and doesn’t encompass as much of the problem as most people seem to think.

    • Brad,
      I agree that many studies have shown the equivalence of generic drugs and I myself choose generics. However, patients are told in many ways that trade drugs are better and are seeking quality so will choose them if given the choice. This is not acting on moral hazard. It is seeking quality in their care.

      The analogy of automobile insurance applies here also. When repairing cars, there are in some cases cheaper “aftermarket parts” which can be used to cut costs. However, consumers demand original parts for their perceived quality and insurance companies have been sued and received sanctions for using “inferior quality” parts in repairs.

    • Austin,
      I am not an economist so don’t have the same frame of reference. The website of The Economist defines:

      ” Moral hazard means that people with insurance may take greater risks than they would do without it because they know they are protected, so the insurer may get more claims than it bargained for. ”

      In the context of health insurance I would assume that the insurance companies would expect that people would not take greater risks to their health and that they would expect that people would not generate a need for health care beyond care for normal disease. This is where I came up with the concept of “unnecessary care”. I don’t think people generate care beyond treatment for routine illness and that they don’t generate extra illness just because they have insurance.
      I don’t know how this fits into your frame of reference of valuation and payment.

      • @Mark Spohr – Moral hazard is not taking risks with one’s health. I’ve given the definition in the post, but it sounds like it wasn’t clear. Here it is again: If you’d pay $10 for a service but not a penny more and your insurer pays $20 for you to get it, that’s moral hazard for health insurance. What it means is that you’re using $20 of resources when you’d be just as happy receiving $10.01 in cash. The extra $9.99 is, quite literally but only in an economic sense, “wasted.”

        That has nothing to do with how valuable the services were to your health. Nothing. That’s a completely different question.

        Getting back to “risk” and connecting it to the definition you found at The Economist, for health insurance, it’s not about taking greater health risks. It’s about putting the insurer at greater cost risk than one would accept oneself. You wouldn’t spend $20 for that hypothetical service, but the insurer must pay it. By the same token, one who is insured for flooding will build in a flood zone, but wouldn’t without the insurance. Once the risk is transfered the insured really does behave differently than (s)he would otherwise.

        However, moral hazard isn’t the only reason we use more health services when insured. I explained that in the post. There’s an income effect too. Even if you would pay $20 for that hypothetical service, you can’t if you don’t have $20. The insurance effectively transfers income at just the right time. That’s not a moral hazard effect.

        Is any of this making sense?