As a dollar amount, U.S. per capita spending for physician services was the highest in the OECD in 1999: $988, compared with an OECD median of $342. . . .
In 1996, the most recent year for which data are available for multiple countries, the average U.S. physician income was $199,000. The comparable OECD median physician income was $70,324.* The ratio of the average income of U.S. physicians to average employee compensation for the United States as a whole was about 5.5. Germany’s was the next highest, at only 3.4; Canada, 3.2; Australia, 2.2; Switzerland, 2.1; France, 1.9; Sweden, 1.5; and the United Kingdom, 1.4.
Mankiw posts three discussion questions. I’m just going to take a stab at the second one:
On the issue of doctor training: Suppose that in country A physicians get free training through a taxpayer-financed educational system, while in country B physicians finance their own education and then, once trained, are paid higher fees. (a) If country A classifies these training expenses as education rather than healthcare spending, which country would report higher healthcare costs? (b) Is that difference in healthcare costs real or an artifact of labeling? (c) In which country would doctors, once trained, have more incentive to work long hours? (d) In which country would there be more doctors? (e) Which country’s system, in your judgment, is more efficient and equitable?
(a)-(b) I get Mankiw’s rhetorical point. And I guess it makes sense to count educational costs as part of the production costs of healthcare. But $199,000 – $70,000 = $119,000. (Updated – I made a subtraction mistake the first time.) So the higher med-school costs people pay in the U.S. get made up in two years out of a career of 30-40 years.
(c) The short answer is that physicians would have more incentive in country B (the U.S.), because the marginal return on labor is higher. But do we want doctors working longer hours? Medicine is not like, say, baking bread – the more hours you put in, the more good stuff you end up with. We are already the country with the highest hourly wages, and one of our major problems is not lack of doctoring capacity, at least not in aggregate; on the contrary, we have the problem of overutilization of many types of services (the expensive ones). Put another way, because we have higher wages for expensive procedures, we have doctors working longer hours doing those procedures by prescribing more of those procedures than are medically appropriate. Because we overcompensate for some services and undercompensate for others (relative to each other), we have too few of some kinds of doctors (family practice, for example) – but that is a product of the way we pay for healthcare, not the way we produce doctors.
(d) You should get more doctors in whichever country gives you the higher aggregate returns to being a doctor. Right now that’s the U.S. But the key point here is not the financing of medical school; it’s the constraint on the number of doctors enforced by the American Medical Association. That’s why, as Mankiw points out, we actually have fewer doctors per capita than the average OECD country.
(e) I’ll leave that as an exercise for you.
* Yes, it says “average” for the U.S. and “median” for the OECD. I can’t tell from the original paper if that is accurate or not. The rest of the paragraph says it is dealing with averages. In any case, I think it’s fair to assume that the median U.S. doctor made well over $70,324 in 1996.