The Domain of Moral Hazard

By Mike Dorf


My latest column on Justia's Verdict uses the JPMorgan Chase trading loss as a point of departure to discuss a common conservative argument against regulation: That providing people and firms with express or implied insurance will lead those people and firms to take risks that are not cost-justified.  In the column, I briefly discuss bailouts of banks and the auto industry, deposit insurance, and at the end, health insurance.  Although the column does not use the term "moral hazard," it does invoke that concept, which is slightly broader than the notion that insurance dampens incentives.  Moral hazard includes the idea that people will be more willing to spend others' money than their own.  In this post, I want to say a few more words about the idea of moral hazard in the health field.

The ideologically conservative argument against third-party health insurance goes much like the argument against insurance in other contexts: In a well-functioning market, the price of a good or service will be set by the intersection of the supply and demand curves; health insurance enables patients to get health services for substantially less than their true cost (because the insurer pays most of the cost); that leads to too much health care and artificially inflates the price of health care.

The problem with the foregoing logic is not so much that it is wrong, but that it is incomplete.  I can't speak for everyone, but personally, I don't like going to see a doctor.  It's potentially unpleasant.  (E.g., the doctor may need to draw blood.)  It takes me away from other things I need to do (work) or want to do (play, spend time with family).  And it could lead to bad news ("I'm sorry but you have scrofula" or "you need to lose 25 pounds"), which, in principle would be welcome because for most conditions early diagnosis increases the likelihood of a cure, but in fact, most people tend to avoid learning bad news even though such information could help them.

Accordingly, it strikes me that there are strong non-economic disincentives to going to the doctor, so the notion that we need policies that further discourage people from using medical services seems wrong.  Co-pays do in fact suppress demand for medical services but I'm not persuaded that they are mostly suppressing unnecessary, as opposed to necessary, trips to the doctor.

Perhaps the real moral hazard problem in medicine is the moral hazard of the providers (chiefly doctors and hospitals).  Suppose that Dr. X is trying to decide whether to recommend that patient Y undergo some procedure (let's say prostate surgery).  In a perfect world, we would want the doctor to make a recommendation based on the expected costs and benefits of the procedure, without regard to cost.  But in our actual world, of course, that's impossible.  Money spent paying for prostate surgery is money not available for other (medical and non-medical) goods and services (for this patient or for others).  And so a rational system would include some consideration of cost.

Our fee-for-service system does consider cost but in a peculiar way: The more procedures that Doctor X performs, the more money that he makes, and so it is in Dr. X's economic interest to over-treat.

That doesn't mean that any particular doctor will in fact over-treat.  Professionalism is supposed to lead doctors to make recommendations strictly on the basis of the patient's best interest.  But basic human psychology will tend to induce doctors to think that they are making decisions and recommendations based on patient well-being, even when their own economic interest colors their perception of such well-being.  (Most of what I have just said about doctors also applies to lawyers, accountants, plumbers, and others who bill by the hour, but I'm discussing health care here, so I'll just note that fact and move on.)

Do the incentives of fee-per-service doctors cancel out the incentives of patients--deterred from visiting doctors by co-pays and the unpleasantness of the experience?  That would be a happy coincidence, but I doubt it is true.  The factors that keep patients away from the doctor will tend to keep them away from visiting regularly or early, thus meaning that when serious conditions are diagnosed, they're more costly to treat.  Meanwhile, the incentives of fee-per-service doctors are to maximize expensive treatments, not office visits.  Accordingly, it looks like we have a set of economic and other incentives that combine to undermine the cost-effectiveness of medical treatment.

One of the lesser-known features of Obamacare is its support for experimentation around alternatives to fee-for-service payment systems.  (This goes beyond the experiments in Medicare and Medicaid.)  In principle, there ought to be cross-ideological support for these experiments, because they build on the insight that moral hazard is real.  But at least so long as health care reform remains highly politicized, we can expect these alternatives to be attacked.