Wednesday, December 17, 2014

Why Price Competition Doesn't Work in Most Health Care Markets

The New York Times has published a good article on the wildly varying prices for diagnostic procedures and the apparent stickiness of those prices at the upper tail of the distribution.  The quickest possible look tells us that the prices are not set based on some kind of marginal cost thinking, as simple market models assume.  For example:

With pricing uncoupled from the actual cost of business, large disparities have evolved. The five hospitals within a 15-mile radius of Mr. Charlap’s home here charge an average of about $5,200 for an echocardiogram, according to an analysis of Medicare’s database. The seven teaching hospitals in Boston, affiliated with Harvard, Tufts and Boston University, charge an average of about $1,300 for the same test. There are even wide variations within cities: In Philadelphia, prices range from $700 to $12,000.
You don't need to know anything more than that to know that the markets are not truly competitive, that consumers are uninformed about the prices (except after the fact when it's too late to shop around) and that the supply side has price-setting power.

In other words, competition doesn't lower prices*.  Rather the reverse, in fact:




Although medical groups cite malpractice lawsuits for the high prices in the United States, some studies suggest that is not a major factor. California, which caps malpractice liability for noneconomic damages, has the highest average charges in the country for outpatient echocardiograms, according to a recent study in the Journal of the American College of Radiology.

What did predict price in a region, according to the analysis? The more machines, the higher the bills. “You would think increased competition would cause price to go down, but it’s the opposite,” said Dr. Stein, the study’s author, who was then a student at Temple University School of Medicine.

Bolds are mine.  The bolded sentences match a lot of older research in this field.  The higher the supply of a diagnostic service, the higher its prices.  If you've taken Econ101 you know that the reverse should be the case if we wish to use competition to lower the costs of health care.

Pricing of medical goods and services is much more complicated in practice than the scissor diagrams of supply and demand from economics textbooks would suggest.  Consumers are pretty uninformed, the intermediaries set price ceilings with varying success, but many of those apply to only certain groups of consumers, such as Medicare patients, and many prices are set to cross-subsidize other services and goods (or patient groups) which are producing losses. 

Add to that the non-profit status of many health care firms.  Basic economic theory cannot say that much about the way non-profits would set their prices**, even under conditions of complete certainty.

Boring stuff, eh?  Except that it could truly hurt your wallet, after your illness has already hurt you.

Here's a bit more crunchy-dry stuff:  The health care system works on the basis of an agency between the patient and the provider.  The latter is expected to help the former to get the best possible outcome, as the patient's more informed agent.  But those agents also have their own incentives:  If a test is painless, has no side effects and the patient is insured, why not order the test?  It could be of some value and it could also provide revenue for the provider.

But if any tests are used above their medically indicated levels, the costs of diagnostic tests will rise and ultimately raise the costs of health insurance for all insured.

The major point of the NYT article, in my view, is how it subtly suggests that markets cannot solve this particular problem.  Countries such as Japan have solved the high costs by setting rules for the prices of various diagnostic services and by requiring those prices to fall over time.  But that's not free-market conservative stuff:  It's direct intervention with a market which really cannot comply with those simple scissors models, however hard it tries.  Because it's too different.

There's too much not-knowing, too much asymmetric information, too many incentives which don't results in clear-cut outcomes.  And the conservatives would like to leave the markets to compete in price!
----
*Exceptions exist:  Basic preventive services, routine checkups, eyeglasses and dental services can all benefit from price competition.  But note that those all share frequent use rates (which help to teach consumers about the product bundles), relatively healthy consumers with energy to learn about prices and much less uncertainty about the efficacy of the offered services.

**Because if firms don't try to maximize profits, what do they try to maximize?  Output?  Then they would price so as to set the prices equal to average costs (so as not to make losses).  But clearly the average costs cannot fluctuate as much as the prices in the above examples.