Arnold Kling  

Are health economists naughty or NICE?

PRINT
Simonizing Economics... The Music Copyright Tax, Revis...

The lead story in today's Wall Street Journal reports


If a treatment helps people, should governments and private insurers pay for it without question? Or should they first measure the benefit against the cost, and only pay if the cost-benefit ratio exceeds some preset standard?

The U.S. generally follows the first course. Even the most cost-conscious insurers say they'll pay the price if a drug works and there aren't other options. Britain openly and unapologetically adopts the second course. If a drug or type of surgery costs a lot and helps only a little, it says no.


A British health commission, with the acronym NICE, evaluates the cost-effectiveness of treatments. They determined that certain medications for Alzheimer's only offer limited in benefit in a minority of cases, so they disapproved the medication.

I like the idea of using cost-benefit analysis to evaluate health treatments. If individuals were making the decision about whether or not to use the medication, then such information would be helpful. Of course, under the British system, most decisions are effectively made by government.

My instinct is that this particular outcome is inhumane, because I suspect that the cost-benefit calculation is distorted by the high price of the medication. If the medication were sold at marginal cost (i.e., the cost of manufacturing and administering the drugs), my guess is that this cost would be even less than the low benefits.

What the cost-benefit calculations tell me is not that people should be denied the drug, but that the drug company is pricing the drug too high relative to its benefits. Maybe at a lower price the company cannot recoup its investment in research and development. But that is how the market ought to work.

If American insurance companies and Medicare are willing to pay high prices for the drug because it provides a benefit that is greater than zero, then to me that illustrates how American health care has gotten into its fix. That is what I call in "gray area medicine" in my forthcoming book.


Comments and Sharing





COMMENTS (12 to date)
Tom L writes:
Maybe at a lower price the company cannot recoup its investment in research and development. But that is how the market ought to work.

What are you talking about? Granting a drug manufacturer a monopoly for a period of time incentivizes innovation, but also ensures that some people don't get the drug even when their benefit is above the marginal cost of producing it. I agree that perhaps we over incentivize development, but how can you say that we shouldn't do so at all?

Dan Landau writes:

Since you have written a book, you have undoubtedly thought the issues through and it would be fool hardy to argue. However, why miss a good argument?

The drug company is trying to maximize profits, as we teach it does, and Milton Friedman said it should. To charge a lower price, the drug company requires some sort of government subsidy. That sounds like the government picking winners. Why should government be better at picking winners in drug research than anywhere else?

D. F. Linton writes:

So is your judgment that the NICE approach is "inhumane" and your suggestion of what sounds for all the world like "production for need and not for profit" the outcome of type C or type M reasoning?

Hans P. Deuel writes:

I thought you might be interested in some concrete numbers that I posted a while back, that shed a somewhat different light on this matter.

spencer writes:

Dan -- I suspect what you are missing is the change in volume.

If the drug company can sell x at $1 it may increase profits by sell 110% of x at $0.91 if the greater volume and greater income cover marginal costs and do nothing to increase fixed costs. This is why we use marginal analysis. My numbers may not be the best -- off the top of my head --but I think you can get the idea.

This is especially true if the drug company can price discriminate -- still sell x at $1 and another 10% of x at $0.91 without losing the original x customers at $1. This is what they do now by selling at $1 in the US and $0.75 in Canada or Europe.

Tracy W writes:
Maybe at a lower price the company cannot recoup its investment in research and development. But that is how the market ought to work.

I presume therefore that you sell your own labour at marginal cost rather than trying to recoup your investment in education and human capital? So you think it is only right to charge what you'd make as unskilled labour - maybe $8 an hour? After all, you think that's how the market ought to work.

Bob Dobalina writes:

No, Tracy, I bet he tries to maximize profits.

If someone wanted to pay Arnold $50 million per year to be a janitor, I'm sure he'd do it without fretting about the sunk costs of his education.

Bob writes:

The discussion is overly focused on the cost of producing treatment, ignoring the value consumers put on it. What we really need is price discrimination - charge Bill Gates $1m, Arnold $500, and grad students $10. Innovators make their money and you maximize the surplus.

Arnold Kling writes:

Hans,
There is a paper by Uwe Reinhardt and others called, "It's the Prices, Stupid." He argues that health care suppliers overcharge, and what we need are buying collectives to negotiate them down. The Clinton health care plan attempted to create such collectives.

A competing narrative is "It's the technology stupid," as a subsequent paper argued. I tend to believe the competing narrative, because the data on trends toward increased specialization and use of advanced technology is more striking than the trends in physician incomes.

Robert writes:

A few years back, Raymond Orbach, a senior poobah at the Department of Energy, hit the lecture and colloqium circuit with a talk that presented the history of federal research expenditures from ~1950 to the present. Finding 1: Even adjusted for inflation, the federal government has spent increasingly large sums on scientific & technological research. Finding 2: Since its inception, the NIH's share of the federal science budget has grown monotonically, until it now spends more on research than all other federal agencies combined (DOD, DOE, NSF, NASA, ...)

The point of the talk was of course that the government ought to be giving more of that money to the DOE and less to the NIH, but we can leave the normative aspects of this budgetary breakdown behind and simply say that federally-funded medical research is a very big business.

Now, advancing medical technology is probably a Good Thing, or at least is insofar as physicians know how to use it properly. But one side effect of rapid improvement in the potential quality of medical care is that it gives hospitals and other medical care interests an increased ability to compete on factors other than price. Judging from hospital advertisements, at least their marketing departments feel that being able to offer the Latest Medical Technology is a major selling point, perhaps even enough of a selling point to justify raising the cost of other treatments to susidize the Latest Medical Technology.

Now, I can't make an ironclad case that in the absence of federally-funded medical research, health care providers would compete on price, but it appears the day is approaching when one can fly to India, have bypass surgery done, recuperate, fly back, and pay for the whole thing out-of-pocket, for approximately the cost of a 10% deductible in having the procedure done stateside.

Phil writes:

Isn't it possible that the cost exceeds the benefit to the *average* user (which is what NICE is considering), but that the reverse is true for the users who are willing to pay the most?

A Rolls-Royce is not a cost-effective form of transportation for the masses either, but it doesn't follow that they are mispricing their product.

Chris Bolts writes:

In regards to your last point Arnold, why should drug companies price their medications at marginal cost if Medicare and Americans are willing to pay a higher price? I'm willing to bet that like the textile industry R&D costs for a pharmaceutical is high in the short-term, but once a drug is successfully completed and patented, it's marginal costs will decline over time due to economies-of-scale, but people's demand for the drugs will be relatively inelastic, especially if it is one of mandatory use. Sure, the drug companies can be beneficent and price the drugs at or slightly above marginal cost, but that wouldn't still help fix their public image of being heartless bastards trying to profit off of someone's misery (of course, I don't believe the last point).

[quote]The discussion is overly focused on the cost of producing treatment, ignoring the value consumers put on it. What we really need is price discrimination - charge Bill Gates $1m, Arnold $500, and grad students $10. Innovators make their money and you maximize the surplus.[/quote]

But what if Bill Gates values the drug less than the grad student? Should the grad student be given a subsidy to offset the high price of the drug at the expense of Bill Gates simply because Bill Gates can afford to give the student a subsidy? Greater transparency in the healthcare industry without more government intervention is what is needed to help bring the true cost of healthcare down and I don't think price discrimination alone will help do that.

Comments for this entry have been closed
Return to top