Saturday, November 23, 2013

Why Are American Physicians Paid So Much?

This was the topic of Matt Yglesias last week, with rather strong words about the idea that the US physicians have a pretty strong cartel on the market (cartels of providers keep incomes higher than they would be in less concentrated markets).  Then Kevin Drum joined in with a point about international comparisons:

What that bar graph shows is the number of practicing physicians per 1000 people in each country.  The blue bar is the OECD average, the red bar is Murka.  So the United States has fewer physicians per capita than the average of the richer countries in this world.  But the United States also has much higher physician incomes than most OECD countries!  A puzzle, if you happen to believe in the free-market-fairies, because high incomes in that reality mean that the markets are trying to lure in more physicians, that there is a need for more physicians.

But note this:  The above situation is not a recent one.  US physicians have been earning a lot more (in relative terms, such as when compared with other professions requiring roughly similar basic talents and years of training) than physicians in other countries for decades.  Yet that "signal" (we desperately need more doctors, screams the market) is not translated into more doctors being trained, or at least not being trained in the numbers one would predict if one was wearing the free-markets fairy-hat.

The answer to that paradox is, sadly, the one Yglesias proposes:  Certain institutions have a lot of market power in the US medical markets, and that includes the medical schools and various physician organizations.

On the other hand, the physician labor markets are not free markets in any other country, either, partly, because quite a few countries determine the number of physicians through government controlled funding processes but also because one of the basic requirement for a competitive market:  free entry and exit of providers, cannot hold (and must not hold) when it comes to physicians,

What I mean by the latter is that physicians are carefully screened and selected through often-vigorous competition for available medical school places and then further screened by licensing requirements.

The reason for the importance of this screening and training process is that there is no way consumers of health care can easily, cheaply and quickly determine whether someone is well-qualified as a medical provider without such signals as framed degree certificates or letters after the person's name.  If we really allowed an unregulated and chaotic so-called "free market" to operate in health care markets, I could set up a cosmetic surgery business in my basement.  You don't want that, because of the consumers' inability to judge provider quality.

Why that is especially crucial in health care (rather than in, say, the florist industry) is that the consequences of using a low-quality provider can easily include death or permanent disability.

Got it?  The physician markets are not truly competitive but regulated by either the physicians themselves or by governments or both, because unregulated entry to those markets is a serious health hazard.  In the United States the gatekeepers are the medical schools and various state-level physician associations.  A sizable chunk of all this is self-regulation.

Self-regulation makes some sense, because the lack of quality information means that bad physicians in the barrel infect all the physicians.  Thus, the physician organizations have an interest in controlling the "bad apples."  But the physician organizations also have the human interest of not ratting on their own colleagues and the more general interest of all trade unions:  To keep the average earnings of the existing membership high.

One way to guarantee the latter is by making entry harder and harder, both because that lets existing physicians have less competition and because the new entrants then look better-and-better, what with passing tougher screening.  That improvement, given the consumers' difficulty of discerning individual physicians' quality, serves to improve the markets for all practicing physicians.

Weird, isn't it?

My point is not to argue that the high US physician earnings are somehow undeserved.  The basic economic theory on how earnings are determined isn't about that, in any case, but about trying to understand how (and if) the demand for a particular kind of labor and the supply of a particular kind of labor come together and determine the average earnings of a profession.  When the markets cannot be competitive, the analysis must look at what is actually happening.  Calling for the free-market fairies to fix the problem is just plain silly, because the physician markets cannot be unregulated.

But once something IS regulated, and especially when it is self-regulated, the interests of the existing workers tend not to support increasing the competition.

So if I had to make one guess about the reasons why the United States in this source (covering data between 2005 and 2012)  had only 24.2 physicians per 10,000 people, while Australia had 38.5, France 33.8, Germany 36.9 and Switzerland 40.8,  I would point out the greater market power of US medical schools and physician organizations*.

What about the other arguments usually advanced to explain the higher earnings of US physicians?  Most of the arguments which are not comparing the US to other nations point out the high costs of getting a medical degree.  And those, indeed, are high.  Especially high when we also think about the fact that medical schools are highly subsidized by the government (via Medicare funding a lot of the teaching hospitals, for example), and those subsidies are not paid back by the student loans.

The problem with that argument is not that medical student loans aren't high or that those wouldn't need to be paid back from incomes which are sufficient for that plus a proper standard of living.  The problem is that those loans cannot explain the high incomes** all through the physicians' working lives:

 Orthopedic surgeons again topped the list as the highest earners, with a mean income of $405,000, followed by cardiologists ($357,000) and radiologists ($349,000).
The lowest-earning specialists were similar to those in Medscape's survey report from last year: internists ($185,000), diabetes physicians/endocrinologists ($178,000), family physicians ($175,000), and pediatricians ($173,000). HIV/infectious disease physicians ($170,000) dropped to the lowest-earning position, which was previously occupied by pediatricians.***

The second argument justifying physicians' high incomes has to do with the long hours physicians work in the US.  But that argument is tied to the small numbers of physicians per population.  If there were more physicians the average working day would be shorter.   It's a chicken-and-egg problem, but essentially something which could be changed without endangering patient welfare, say.  Indeed, one could argue that extremely long work-shifts endanger patient welfare more.

Then there's the argument that has to do with what one deserves.  I have already talked about how the market-determined earnings may not have much to do with what we truly deserve or not, but the more detailed argument has to do with the difficult and exhausting type of work physicians do and the  life-or-death power the work involves.  Those are also arguments which in themselves are not incorrect.  But as the same arguments would apply to physicians working in any other country, they cannot explain the international differences.

Finally, the idea that the quality of physicians might be better in the US because of tougher control of the various intake numbers is worth thinking about.  This may well be true, though it's hard to construct studies of that kind (because of the nebulous aspect of the whole concept).  But note that if the number of physicians is to be kept small because of that quality argument, we are then right back in the problem of long working hours for physicians, and, more importantly, the lack of access to physician care by many who have a medical need for it.  Alternative solutions exist for those who worry about lowering quality with increased intakes of new students in medical schools, say.  Those include using more other medical workers to carry out various routine tasks, allowing task-licenses for paramedics who get additional training in a narrowly defined medical field, and in general re-defining tasks in ways which make the few physicians go further.  Doing this would not, however, change the fact that physicians would face more competition and this would probably lower their average incomes.

It's worth pointing out that physicians are paid well (in relative terms) in all those OECD countries, and nobody is trying to change that.  The crucial question is how to make the American system more responsive to the health care needs of its people, and in that context looking at the various cartels in health care is necessary.  The physicians are not the only medical profession or industry with market power.  Just think of the pharmaceutical industry or the hospital industry.

Added later:
I want to stress, once again, the fact that salaries based on markets or other practices of similar sort are NOT the measure of the value of someone's labor in some ethical or moral sense.  Otherwise we would argue that hospice workers taking care of the dying are only worth the minimum wages which they often receive.  Neither does any of the above mean that individual physicians participate in the cartel-keeping or that they choose the profession for the high earnings or that they wouldn't deserve their earnings in the moral sense.  The question is simply to answer the question why American physicians are better paid than physicians in other comparable countries.

*Sure, the data may not be strictly comparable, and, sure, Canada has even fewer doctors per 10,000 inhabitants (20.7).  But some of the countries with almost as low physician/population ratios as the US either regulate physician numbers centrally for financial reasons or allow other medical professionals do more of the tasks which are here defined to be done by the physicians only.

Note that all the Nordic countries have more physicians per population than the US.  Even Armenia and Egypt do.  Even the UK with its "socialized" health care system has more physicians per capita.

**Female physicians earn less than male physicians, on average, though the earnings gap is getting slightly smaller (30% overall now, 17% in primary care).  The reasons for the wage gap are probably partly the same as in any occupation (women's greater responsibilities for children and household chores which translates to lower hours at paid labor, societal steering or personal preferences which lead women to "choose" somewhat lower-paying specialties etc.). but an additional reason in medicine is the fact that women are fairly recent entrants (in large numbers) in the field, so we are comparing an average man and an average woman in an occupation where the average man is quite a bit older and thus has the income advantages experience provides.

***Whether the earnings differences in that quote are truly reflecting relative scarcity etc. is another question on which a long post is desired.  For instance, the relative ratio of specialists to general practitioners seems not be declining (but rather growing) in most countries.  Yet health care planners have for years complained about too many traditional-type specialists and not enough general practitioners and gerontologists.  Indeed, Herodotus mentioned that about medicine in ancient Egypt:

The practice of medicine is very specialized among them. Each physician treats just one disease. The country is full of physicians, some treat the eye, some the teeth, some of what belongs to the abdomen, and others internal diseases.