Dean- I understand how patents make new drugs more expensive than they would be otherwise because the marginal cost of producing an extra pill is tiny. We see proof of this once they come off patent.

You nave never, to my knowledge, presented any evidence that this is true of medical procedures. Your swat at GE makes me guess that you think MRIs are a lot more expensive and if companies could make generic MRIs, these would be much less expensive.

Do you have any basis for this assertion? It seems lots of companies besides GE make MRI machines and, unlike drugs, the marginal cost of producing an MRI (and the servicing I suspect) is pretty high.

plus which, unlike a pill, which patients take pretty independently, most procedures require the time and attention of a highly trained individual for some period of time.

Samuelson also refers to an "undesirable income transfer from the young to the old", which is rather odd since the money in question is not going to the old themselves but to the health-care and insurance industries. Does Samuelson think that old people should receive no more health care than young people?

Samuelson is more open in disapproving of the young-old money transfer in the case of Social Security - he thinks the retirement age should be increased and/or benefits should be cut. Of course McCain is on record as disapproving the transfer, which is the basic idea of Social Security.

The media should take more notice of the implications of this conservative attitude, and probably Democrats could make it an effective point of attack with older voters.

3 companies supply most of the market for MRI magnets, although other companies sometimes resell the magnets under their own brand. GE is the largest, followed by Siemans, and Philips. Hitachi, Toshiba, etc. are bit players. In other words, I disagree, the market is very concentrated.

I think because of the capital cost of MRI equipment, unlike the capital cost of pill-making equipment, MRIs wouldn't drop more than about 2/3 of their price. If the market is currently $1.9bn, it'd save about $650m. In other words, I agree, the savings would not be like the incredible savings of the drugs.

Where there is a difference is in other diagnostic techniques. A lab test might cost $5 to perform, but the patents bring the price to $18, which is resold for $30. If people could get a $10 price, they'd save money. Insurance companies would also be more interested in promoting diagnostics, because they are much cheaper than unnecessary or late treatments.

Mr. Duncan- two points
1. Three major suppliers may actually be sufficient to produce adequate competition. I think that is the approximate number of suppliers of LCDs and those have plummeted in price and increased in size and quantity.
2. The overwhelming majority of lab tests performed are not under patent. I would need to see some numbers on the ones that are to be convinced that they are a significant portion of health care costs.

Dean Baker wrote, The other problem with Samuelson's analysis is that often the "expensive" procedures are not really expensive in the sense of using substantial resources, they are only expensive because the government hands out patent monopolies.

Erik L wrote, You nave never, to my knowledge, presented any evidence that this is true of medical procedures. Your swat at GE makes me guess that you think MRIs are a lot more expensive and if companies could make generic MRIs, these would be much less expensive.

Actually...

There are other problems with the health care industry besides things possibly costing too much.

Namely, there is a huge amount of market failure because (a) there is too much unnecessary treatment and (b) there is too much of a financial and professional incentive to treat.

So, if you have more competition, you won't necessarily get an optimal solution, because you'll still be plagued by too much treatment.

This is really standard knowledge in the world of health economics. It's based on the work of the Dartmouth group, who showed that there are extremely large regional disparities in amount of treatment---even controlling for the severity of the cases---that have zero correlation with outcome.

So even if you could have more, cheaper MRIs, you might end up getting just more unnecessary MRI scans. This certainly seems to be the case when hospitals in the same metro area "compete" in capital investments like scanners.

Mr Duncan wrote, Insurance companies would also be more interested in promoting diagnostics, because they are much cheaper than unnecessary or late treatments.

Again, sometimes yes, sometimes no.

E.g., the highest level committee responsible for reviewing the issue recommended less use of the PSA test for prostate cancer, because often from a cost benefit point of view, it's unwise to test.

So if you have cheaper PSA tests, you'll at best mildly increase the incentive to test, which in most cases is actually going in the wrong direction.

Unfortunately, epidemiology is something most people (especially doctors) seem incapable of understanding, as witnessed by the ridiculous posturing of some of the MDs who are pro-treatment quoted in the Washington Post article on this; and also some absurd letters to the editor.

(Such as one idiot who wrote that one cost of the PSA test, which is possibly unnecessary prostate biopses, isn't a cost because he claimed such biopses are a simple office procedure. Except for the fact that you can e.g. get a nasty cardiac infection from the "simple office procedure".)

Fact is that the "market" for health care (here I mean health care, not health care insurance) is so completely screwed up, any attempt to discuss this market in the context of a semi-rational supply-and-demand system is flat out incorrect.

Right on the mark, Dr. Baker

Dean,

Your forgot the other protection that you often talk about that drives up health care costs. The lack of competition from doctors from outside the US to drive down wages of doctors, just like they do in lower wage jobs.

How to begin ...? The real answer is in removing the assumption that ANY third party (insurance, government or other) should be the primary payor for healthcare, OR that the "employee benefit" model (creating a huge base for the statist quo that got us into this mess?)should be the norm.

If health and wellness were first of all the province of self-responsible adults who treat their bodies with respect and care, and only routine physical and screenings were what the vast majority of us needed for keeping "illness" away ... all of this would be academic.

We could then provide for the rare exceptions -- where someone was actually "sick" or "ailing" -- with community, charity and other VOLUNTARY methods.

Getting from here to that status is the goal, NOT getting everyone under an umbrella of "insulation" for every hangnail or sniffle ... (Please recall that the under "car insurance model" often cited as comparison, we DO NOT insure for oil-changes, brake jobs, etc., but for major accidents, and the liability and hospitalization expenses that ensue from them.)