Doctor’s Salaries Aren’t the Problem

Recently, Politico published an article entitled “The Problem of Doctor’s Salaries.” The article is written by economist Dean Baker and the gist is that doctors are overpaid due to an artificial suppression of supply. Baker claims that addressing this artificial suppression will help control healthcare costs. Superficially, Baker’s point seems valid. Unfortunately, he makes multiple incorrect statements about medicine (both in this country and in others), and more importantly, the solutions that Baker offers will not help with problem that he purports to address. Namely, eliminating this “artificial suppression” of doctors will not have a significant effect on healthcare spending.

First, a word on the basic economics. Baker claims that doctors in the US make about twice as much as doctors in other countries, representing a spending excess of $100 billion dollars. He then proposes some changes that would increase the supply and help eliminate this “doctor tax.” But an important point that he seems to miss is that the salary of the individual doctors is not what is important to the system; it’s the aggregate salary of ALL the doctors. And it turns out — having more doctors making less money costs about the same amount as fewer doctors making more. Let’s take a closer look.

How many doctors would we end up with if we eliminated barriers to entry? Well that’s exceedingly difficult to predict, but here are some relevant numbers. According to the World Health Organization, the US had about 2.45 doctors per 1000 people in 2014. If the supply of doctors is indeed artificially suppressed, then the removal of that suppression should cause that number to increase. But how high? The highest per capita number on the WHO’s list is Qatar at 7.7, but rather than taking the most extreme example, let’s use Germany, Italy, Switzerland, and Bulgaria (#’s 7–10 on their top ten list). Those four average out to about 3.9 doctors per 1,000 people. If the United States went from 2.45 doctors (per 1000 people)to 3.9, the number of practicing doctors would go from 900,000 to 1.43 million.

How much of an effect would that have on salaries? It’s hard to find data on how much doctors earn in these countries so for the purpose of illustration, we’ll use Baker’s average salary for doctors in “other wealthy countries”. He never actually provides that number but I did some fancy arithmetic from the other numbers he cites and came up with $138,000. 1.43 million doctors making $138,000 each comes to a total of $197 billion. That’s down from Baker’s estimated current expenditure of $225 billion (again, fancy arithmetic) — a savings of 27 billion dollars. That’s 0.8% of current healthcare spending. Not exactly a meaningful improvement, and that’s using Baker’s unreferenced estimates. In the real world, the effect would be even smaller because doctor’s salaries are determined by insurance company pay scales, and thus are largely unaffected by traditional free market forces.

But Baker’s paper is more than an exercise in faulty economics. He also goes out of his way to paint the US medical system as fraught with abuse intended to stack the deck in the favor of the “overpaid” doctors. Unfortunately, his claims aren’t particularly accurate.

He starts by saying that the supply of doctors is controlled by the number of slots in medical schools and residencies. Realistically though, with only 18,500 US graduates applying for 36,000 residency slots, the number of medical school graduates is irrelevant to the eventual supply of practicing doctors (the rest of the positions being filled by foreign graduates). That said, he’s certainly correct regarding the bottleneck at the residency level. Fewer residency positions means fewer doctors. So why don’t we just train more residents? Baker would have you believe that doctors lobbied to keep that number low. Here’s what actually happened.

Historically the supply of doctors has cycled between shortage and surplus, which is pretty much what you’d expect in a real-world free market. That all changed in 1997. Baker would have us believe that this was the year that “Congress lowered [the number of residency slots]… at the request of the American Medical Association and other doctors’ organizations.” But what really happened was the Balanced Budget Act of 1997. With government spending spiraling out of control, Congress decided to find places to save money. As the major source of funding for residency programs, Medicare was spending about $10 billion a year training doctors. And that number kept going up. With the Balanced Budget Act of 1997, this spending level was frozen, effectively locking in the number of residency positions in the country. This was done as part of an effort to balance the federal budget, NOT to decrease the supply of doctors.

To be fair to Baker, back in 1994 the American Medical Association (AMA) warned us about a looming surplus of about 165,000 doctors by the year 2000. And who knows — maybe the AMA was doing some hardcore lobbying to get this restriction included in the Budget Act. But less than a decade later, the doctor shortage became obvious and the American Association of Medical Colleges started lobbying for an increase in the number of medical graduates. Heck, Baker himself casually mentions that even the AMA currently wants more residency slots. So when Baker refers to a “very powerful political lobby” that wants to keep this shortage going, I don’t know who he’s talking about; and of course he doesn’t say.

Now that we’ve established that (a) increasing supply won’t help with costs, and (b)doctor’s aren’t going out of their way to keep up the current shortage, let’s look at a few more specifics from Baker’s article.

“…like anyone in a position to exploit a cartel, [doctors] also get patients to buy services (i.e, from specialists) that they don’t really need.” Actually, doctors don’t get paid more by sending patients to specialists. There are anti-kickback laws in place to prevent this from happening. In fact, with the recent push towards “Value-Based Care,” some doctors are even penalized for making referrals. Doctors refer to specialists for one reason — because the patient has a problem that is outside the expertise of that particular doctor. A good doctor knows what he knows, and more importantly, what he doesn’t know. A general surgeon who hasn’t operated on a liver since residency shouldn’t be taking out your uncle’s complicated liver tumor.

“roughly two-thirds of [US] doctors are specialists…[but in other countries] roughly two-thirds of doctors are general practitioners.” The two-thirds number is pretty accurate for America, but it turns out, America isn’t too far off from the rest of the world in this respect. According to the European Commission, of 36 countries surveyed, all but Ireland had more specialists than generalists; and even Ireland was barely higher than 1 to 1. Baker provides no reference for his claim so I’m not sure where it came from, but I’d be surprised to find any modern medical system with two-thirds generalists.

“teaching hospitals… have an incentive to offer residencies in specialties from which they can get the most revenue per resident.” I don’t want to get into how hospitals are compensated, but it’s based on the estimated costs to the hospital providing the training, not on the specialty mix. (Here’s a nice article that talks about the process if you’re interested.) “This means they are more likely to train someone in neurology or cardiology than as a family practitioner.” According to the National Resident Matching Program (NRMP), 3,215 doctors matched into Family Medicine positions in 2017, compared to 479 in Neurology. Cardiology (a fellowship) matched 855. And it’s probably worth mentioning that Internal Medicine (the main category of medical “generalists”) matched 7,101 positions. So again, I’m not sure where Baker is getting his information.

Now that we have debunked some of Baker’s myths, let’s take a look at his “solutions.”

Baker makes three main suggestions. Increase the number of residency spots, let midlevel providers (nurse practitioners and physician assistants) do more, and allow foreign doctors to practice here without completing a US training program. The first suggestion is a no brainer and is supported by just about everyone. As I mentioned earlier, it wouldn’t do much for healthcare’s financial concerns, but it certainly would increase the number of practicing doctors, which would be a net benefit for everyone. And believe it or not, the next two suggestions are also fairly reasonable. Just not for the reasons that Baker suggests. There is an ongoing debate about midlevel providers working without supervision as well as about allowing foreign doctors to practice without completing a US residency. Some say quality will suffer. Some say they are just as good. I don’t want to go off on that tangent so for the sake of discussion, we’ll just say what everyone agrees on — these providers are certainly better than no providers at all. So what happens to costs when they join the workforce? Turns out, we have some data on this.

In 2016 UC San Francisco published a study that looked at medical costs in states where Nurse Practitioners (NP’s) are allowed to practice independently. Their findings were interesting, but not entirely surprising. They found that average (per patient) primary care costs went down — because NP’s cost less than doctors. But TOTAL costs increased. Why? Because increasing the number of providers made medical care available to people who previously had no access. And it turns out that inexpensive care still costs more than no care. To be clear, the increased access that would come with such a change is a GOOD thing. But a cost savings, it is not.

So this brings us back to the original question. Why do doctors in this country get paid more than doctors in other countries? Well, with factors like expensive schooling, long training, malpractice worries, low job satisfaction, ever-increasing regulations, long work weeks, and fixed pay scales set by insurance companies, that question isn’t going to be answered without writing an entire book — one with the complexity of an Agatha Christie novel. And it certainly won’t be answered by something as simple as “artificial suppression of supply.”

If we want to take on the challenge of fixing healthcare in this country, we have to be honest about what the problems are. Yes, costs are out of control. And with healthcare spending topping $3 trillion a year, Mr. Baker and many like him are anxious to find a quick fix. Doctors are an easy target because they are the face of the system — and their salaries put them near the top of all wage earners. But at only 8% of total healthcare expenditures, doctors could work for free and we wouldn’t be putting a dent in our spending. So no, we don’t have a problem with doctor’s salaries. We do, however, have a problem of too few doctors. Ironically, despite erroneous “facts” and severe flaws in his reasoning, Mr. Baker makes a series of reasonable suggestions. He has recognized the doctor shortage and thought of ways to help overcome it. But his paper is flawed because he has identified the shortage as the root of another problem, rather than a problem in itself. Addressing the doctor shortage will not save the system any money. It will, however, save lives. And isn’t that why we’re here in the first place?