• Health insurance and the income effect

    Frequently, when I mention to a trained economist that I do not have a degree in economics I am told that is probably to my advantage. I haven’t been brainwashed. They’re joking, but only partly.

    I got a sense of the benefits of missing the indoctrination when I read about John Nyman’s theory of demand for health insurance. I saw it first in Health Economics: Theories, Insights, and Industries Studies, by Santerre and Neun. It was explained as an alternative to the standard theory. (A key implication of the standard theory is that all care that is valued below marginal cost is “inefficient” moral hazard.)

    That surprised me. To me Nyman’s theory, or at least some elements of it, is intuitive, even obvious. I believed it before I even knew about it. I could not imagine it not being the standard theory. Thing is, I didn’t really know what the standard theory was. Such were the benefits of studying physics and statistical image and signal processing. Not a peep about demand for health insurance in those classes.

    I’m now reading Nyman’s book, The Theory of Demand for Health Insurance. Here’s his first description of one of the problems with the standard theory that his addresses:

    [The conventional theory] ignores the fact that many expensive procedures would be unaffordable without insurance, especially for those households with low incomes and low net worths. For example, a person with a $50,000 net worth would not be able to afford to purchase a $300,000 liver transplant procedure. The patient may be too ill or the illness may be too urgent to save the difference, and banks may be reluctant to lend money for the transplant because of the risk that the borrower will not be able to pay off the debt. For expensive, lumpy procedures like this, a consumer without insurance may receive only minimal palliative care, even though with insurance the consumer would have access to the appropriate expensive procedure.

    I would not make too much of the dollar values in this passage. They’re just illustrative. For less well off individuals and families, a $100 office visit or a $1,000 minor procedure could be as out of reach as a $300,000 transplant is to a $50,000 net worth household. Would we call the provision of Medicaid to such poor families the source of moral hazard in such cases? If so, why do the poor experience moral hazard for the same services that the wealthy may pay for out-of-pocket, which by definition is not moral hazard care? Is moral hazard income dependent? Should it be?

    There is an income effect to insurance that is crucial to recognize. Labeling every bit of use that an individual would not have purchased with his own money as “inefficient” moral hazard really misses a key point. So does failing to recognize that one individual’s catastrophic range is not the same as another’s. To some $1,000 or even $100 really is a lot of money, money that could improve their health, even save their lives. To others $100 or $1,000 is next to nothing.

    Again, what’s the cost of “efficiency”? Are we even sure that’s what we’re buying?

    • I have gone round and round on this issue with the free marketeers. I agree that the most efficient way to run health care is for everyone to have enough money to pay directly for their own care. We would, ideally, never use insurance to pay for health care. But, it just wont work. I like the example of the premie baby better than the liver transplant. No matter what we do, it is going to cost a lot to provide quality care for early premies. The specialized equipment and the dozens of staff needed around the clock will cost a lot in any system, and it comes at a time when few have real savings.

      So, we have to have insurance, but insurance distorts markets, or so I believe. I dont see a way around it. As you point out, lots of medical care is now beyond the means of many people, and the number of those people is growing. If the is economic moral hazard in providing that care, I would also note that not providing care is a social immoral hazard.


    • This is obliquely pertinent. It is an interesting insight stated more clearly than many of us manage to make.



    • “There is an income effect to insurance that is crucial to recognize. Labeling every bit of use that an individual would not have purchased with his own money as “inefficient” moral hazard really misses a key point. ”

      Thank you for pointing this out. This is the point that I was trying to make previously during the discussions on “moral hazard”. I may not have expressed it as well as you did here but I agree fully.

      If you look at a “strict” definition of moral hazard, ANY use of insurance could be considered moral hazard. I would prefer a definition of moral hazard for health insurance which defines it as use of insurance beyond what is expected (i.e. people are induced to seek medical care since it is “free”). In this later definition, I really think that there is very little use of insurance which meets the definition since health care is never really free. There are always costs for co-pays, time, and transport as well as the pain and risk of damage from the health care itself.

    • there is a Health Affairs Commentary by Nyman on inefficiency and moral hazard

      Is ‘Moral Hazard’ Inefficient? The Policy Implications Of A New Theory
      John A. Nyman
      Health Affairs, 23, no. 5 (2004): 194-199

      • @jamzo – The paper is here, and I believe it is ungated: http://content.healthaffairs.org/content/23/5/194.full.pdf+html

        @Mark – I had made this point too, though in some posts I adhered to the standard economic theory to illuminate it.

        @steve – As I’m sure you know, there is nothing anti-free market about insurance. A perfect market would offer every product anybody wanted to buy. It doesn’t make sense to outlaw insurance and claim the market is then “free.”

    • “As I’m sure you know, there is nothing anti-free market about insurance.”

      I am actually torn on this. Remember your piece about prostate care? If we all had enough cash on hand to pay for all of our own care, and if all three methods of treating prostate cancer had equivalent outcomes, wouldnt we expect the cheaper treatment to dominate? I guess this also assumes that the patient and physician both have access to this information. Once you are insured, especially in a large group, isnt that decision less influenced by cost? What am I getting wrong here?


      • @steve – Insurance doesn’t solve all problems and it introduces others. Which and to what extent depends on design. But there is deviation from optimality no matter which way you go, just for different reasons.

        • @steve – One more thing I’d add about insurance: it’s a waste of time to try to argue it into nonexistence. That will never happen. I do think that HDHPs will grow, so that’s not a useless thing to push for if one wants to. But wishing for an insurance-free world is like hoping for unicorns. The latter may even be more likely.

          You know my point of view. I try to work within the realm of the possible. We can make the system marginally better even if we can’t achieve any one person’s ideal.

    • I’ve read some of Nyman’s work but have a little bit of a hard time understanding exactly how his approach differs from standard insurance theory.

      Standard theory does indeed tend to describe health shocks as income or wealth shocks, but I don’t see how the standard approach REQUIRES that these health shocks be thought of as “affordable” through savings or other means.

      If you can afford to treat an illness, you can think of illness as a shock to income. If you cannot afford to treat it, you can think of the illness as a shock to health. Either way though, if there is diminishing marginal utility to income and health, then there is reason to insure against the possibility of illness. And that’s just maximizing expected utility. Seems standard to me. What am I missing?

    • “I’ve read some of Nyman’s work but have a little bit of a hard time understanding exactly how his approach differs from standard insurance theory.”

      I think this depends on what you’re calling ‘standard insurance theory.’ One thread of Nyman’s book responds to the particular (and widely held) view expressed in a brief article by Mark Pauly from 1968. Pauly’s point was the moral hazard argument–that the additional health care consumed by insured patients is 1. moral hazard, and 2. therefore, welfare decreasing. This argument implicitly assumes that the additional care consumed was a movement down the demand curve: that is, it was due to the reduced price of the care. The other way to look at it is to assume that the additional care is due to a shift in the demand curve, rather than a movement along it: a person who is sick values health care differently than he does when he is well. If you can picture the supply/demand graph, you can see that this difference in interpretation reduces the amount of moral hazard/inefficiency.

      As I understand it, Nyman’s main point against the standard economic view of health insurance has to do with the interpretation of moral hazard as necessarily welfare decreasing.