Oren Cass, who served as Governor Mitt Romney’s Domestic Policy Director in the 2012 presidential campaign, shares his thoughts below on Steven Brill’s recent TIME cover story on U.S. health care costs.
Steven Brill’s TIME cover story squeezes enough outrageous sounding numbers into its tight 24,000 words to create an extraordinary amount of heat ($7 gauze pads! million dollar salaries!), but it manages to shed no light at all on the actual problem of health care costs.
His central thesis, as far as I could tell, is that the enormous mark-ups in hospital bills prove the existence of excess profits in the health care system, while the low prices paid by Medicare and in other countries prove the viability of reducing prices significantly. All of that is wrong.
Brill spends much of the article identifying individual line items in medical bills that are dramatically higher than the wholesale cost (or Medicare price) for the same item. Thus, for instance, he is outraged that a patient might be charged $7 for an “alcohol prep pad” when the per-unit cost online is about one cent. He is sure there must be excess profit involved when a hospital charges $20.44 to Medicare for a chest x-ray, but $283.00 to a private patient for the same procedure. Overall, markups he flagged on basic supplies ranged from 6x to 700x, markups on drugs and devices ranged from 2.5-5x, and markups over Medicare were typically 10-20x.
Profit Margins
So what are the profit margins at these outrageously profiteering facilities? Generally around 10%. Brill should have given up on the article right then and there, realizing that the supposed markups he had identified were not actually leading to outsized profits. Instead, he tried to make the case that such a return (11.7% on average for all non-profit hospitals, actually lower at for-profit hospitals) is in fact too large – “the envy of shareholders of high-service businesses across other sectors of the economy.”
This is wrong in so many ways. High-service businesses can be very high margin, particularly when the service provider is a highly-trained professional (see: lawyers, accountants, consultants, etc.). Hospitals are actually highly capital-intensive businesses (as he reports in the context of “gleaming new facilities,” “big shiny buildings,” and MRI machines that are apparently more-or-less “pure profit”). Indeed, it was “a group of glass skyscrapers” at the Texas Medical Center that inspired him to write the article in the first place.
But most importantly, what makes a 12% operating margin particularly high? At no point does he make any attempt to put that figure in context, perhaps because a 12% operating margin is not at all out of the ordinary. Running a quick Google stock screen indicates that a 12% operating margin would be below average for most major corporations across industries. (If Brill’s point was that this margin is too high for a non-profit hospital, that would be an interesting argument and an interesting article… but that doesn’t seem to be where he is going.)
This helpful compilation from the Value Line database suggests that the average pre-tax operating margin across more than 6,000 firms is 17%, and that medical services is significantly below average. Of course, there are many apples-to-oranges concerns in comparing margins across firms of different sizes and industries (and this Value Line dataset may be completely wrong – a more thorough analysis would be welcome)… but Brill doesn’t even appear to be in the ballpark of having a point.
The disconnect becomes even worse when one notices that the majority of Brill’s examples are from inpatient care, on which (according to Steven Brill, citing a McKinsey study) hospitals only earn a 2% margin. If out-of-control overcharging by hospitals is producing margins of 2%, what level of charging should they be targeting?
His contextualization of the profitability of drug and device companies is similarly, well, non-existent. We are told that because a product can be manufactured cheaply and sold at a high price, there must be excess profit. But returning to the imprecise Value Line data as at least a starting point, there is no evidence to support the premise. The Drug, Biotech, and Invasive Medical Supply categories are all relatively high-margin, to be sure. But they are grouped closely with other high-tech industries like software, semiconductors, information services, telecom services, computers, not to mention utilities and railroads.
Such analysis actually overstates the profitability of the medical technology industries because of severe selection bias. Many of the key innovations in the field come out of smaller firms, which suck up enormous amounts of investment before going bust or being acquired. So we see the profits that comes with having succeeded, but not the losses associated with much of the failure. Perhaps the best proxy for the attractiveness of the industry would be the lack of enthusiasm in the investment community. Nature reports an exodus from the field by venture capital; PWC reports ~15% declines in both biotech and medical device investment, with total deals at their lowest level since 1995.
Average Cost vs. Marginal Cost
Brill tries to save his premise by pointing out that hospitals happily accept Medicare patients at lower prices. “As with all hospitals in nonemergency situations, ManorCare [a facility that charges Medicare $571 per day for a patient Brill profiles] does not have to accept Medicare patients and their discounted rates,” he writes. “But it does accept them. In fact, it welcomes them and encourages doctors to refer them. Health care providers may grouse about Medicare’s fee schedules, but Medicare’s payments must be producing profits for ManorCare. It is part of a for-profit chain owned by Carlyle Group, a blue-chip private-equity firm.”
He makes the same argument about drug companies, noting that they gladly sell their products in price-controlled markets overseas: “those regulated profit margins outside the U.S. remain high enough that Grifols, Baxter and other drug companies still aggressively sell their products there.”
Apparently he missed the day in Economics 101 that reviewed the difference between average and marginal cost. Seriously, there is typically a day dedicated to the topic because it is a hard one; a lot of students intuitively make the same mistake Brill does.
Here is the problem: Suppose PharmaCorp spends $10,000 to develop a drug and can then produce each pill for $1. Suppose there are 1,000 patients in the U.S. who need the drug. If it produces 1,000 pills it will have spent a total of $11,000 so it has an average cost per pill of $11. To earn a profit, it must charge at least $11 per pill. But now suppose France invites PharmaCorp to sell 1,000 pills to French patients, provided it sells the pills for $2 each. This is a great deal for PharmaCorp because the marginal cost of each pill is only $1. So PharmaCorp can spend an extra $1,000 to produce the pills and earn $2,000 selling them.
Brill sees PharmaCorp excitedly selling in France at $2 per pill and concludes that this still allows a sufficient profit margin for PharmaCorp… but obviously PharmaCorp cannot sell all of its pills for $2; its average cost per pill is still ($12,000 / 2,000 = ) $6.
The same goes for ManorCare, or any hospital. Once it has built the hospital and all the rooms and hired and staffed all the nurses and doctors, the marginal cost of filling a bed for a night is low and it welcomes Medicare’s patient for $571. But its average cost per patient is still much, much higher. Not every patient can pay $571. And by establishing the below-average-cost Medicare price, the government actually drives up the price paid by private parties, forcing them to implicitly subsidize the Medicare coverage.
Unfortunately, this error permeates almost everything Brill writes, as he goes line by line through hospital bills identifying prices far in excess of the actual cost of the product or service provided. If the marginal cost is so low, he seems to be saying, why must the price be so high? But as the unremarkable operating profits (remember, 2% for inpatient care!) demonstrate, average prices are not ending up much higher than average costs.
The More Interesting Question
None of which is to say that medical billing is a paragon of transparency and efficiency. To the contrary, its opacity is a problem and serious health care reform would address it. But in focusing on mark-ups, Brill has not identified an actual source of potential savings. Yes, we could bring private pricing closer to average cost if we did not ask those patients to subsidize Medicare’s lower payments, but the result would be higher Medicare costs and an increase in government spending. We could demand that hospitals stop marking up individual items on bills, but the result would be either bills that incorporated all charges into a single and even less transparent line (Treatment for Heart Attack: $100,000) or bills that actually allocated fixed costs to individual patients and looked more confusing than ever (Facility Depreciation: $20,000).
The shame of it all is that an important article could and should be written about, as Brill describes his mission, “Where’s all that money coming from? And where is it going?” Because the really remarkable thing is not that someone is making huge profits with massive markups, but rather that someone is not. All of those enormous costs for treating a patient actually go to pay for things, not to line the pockets of scheming industrialists. But what? How much of a hospital’s expenditures are construction? Capital equipment? Doctors? Supplies? Management? Bureaucracy? And each of those things that it buys – an MRI machine, a pacemaker, a cancer injection – where does that money end up? How much of it goes to researchers? To the acquisition of start-ups that create new intellectual property? To TV advertisements? How much of a doctor’s income goes to the cost of her education? To her malpractice insurance?
It is by understanding the relative sizes of those buckets that policymakers will begin to understand a path forward in controlling health care costs. Maybe the executive salaries that Brill repeatedly harps on are a huge source of cost. More likely (and the reason Brill never moves beyond innuendo to actually quantify the cost) they will turn out to be a drop in the bucket.
In perhaps the most telling line of his article (it’s in there, you just have to read 22,000 words to get to it), Brill writes: “however much hospitals might survive or struggle under [single-payer Medicare for all Americans], no doctor could hope for anything approaching the income he or she deserves (and that will make future doctors want to practice) if 100% of their patients yielded anything close to the low rates Medicare pays.” Why that applies to doctors, but not every other segment of the health care value chain, he never bothers to explain.