« Markets in everything, legal edition | Main | Harry Potter and the Mystery of Inequality »
Measuring health care costs and benefits
The U.S. spends more than any other country on health care, over 14 percent of gdp. A crude economist would take the "gdp" of the health sector as a measure of its economic value. Of course this isn't quite right, since spending money does not automatically increase health or customer satisfaction. Advocates of European health care systems are keen to point this out.
In other words, "p x q" doesn't measure the benefits of the U.S. system. But critics are quick to insist that "p x q" measures the costs of the U.S. system. That is a disconnect. The real costs are what could have been produced with those same resources, or opportunity costs.
Let's say a patient pays $1000 to a doctor, but half of that sum is fraudulent pricing brought on by patient irrationality, non-transparency, fear of death, and fraud. Sound familiar? The real social cost is what the doctor could have produced, had he not been looking after that patient. The real social cost is $500, not $1000.
How does this sound as a debating point: "America doesn't have better health care outcomes than Europe, and in real terms we are spending about seven percent of gdp -- adjusting for relevant corrections-- on health care"? Not so fierce.
Of course one might dispute the 1:2 ratio mentioned above.
But in general, the more American medical care is overpriced in current markets, the more current expenditures overmeasure social costs.
The more you think that government can bargain down prices without wrecking supply, the more you should think that current expenditures overmeasure social costs.
Recall that economists do not find efficiency loss in higher prices per se. Higher prices are bad as a sign of restricted output. But the American medical system has high prices and relatively high outputs, for reasons partially explained by Maggie Mahar. So don't measure cost by the high prices, look for output restrictions, either for health care or elsewhere.
Two caveats: We still must consider secondary economic costs; for instance it is harder to switch jobs for health insurance reasons. Second, medical care is a sector where distribution and efficiency are not totally separate.
But the general point remains: many of the arguments for greater government involvement imply that the true cost of the American medical system is less than the current numbers imply.
Posted by Tyler Cowen on April 23, 2007 at 07:07 AM in Medicine | Permalink
Comments
Most people regard the true cost of something as what they spend to buy that something. You seem to be confusing the common usage of a term and the technical usage.
If there are two possibilities with essentially equivalent outcomes, and one costs X while the other costs 2X, then the first system is the one a consumer will prefer. That we spend 2X is evidence of market inefficiency, market power, etc.
Posted by: richard at Apr 23, 2007 8:05:33 AM
Why doesn't the true social cost depend on the shape of the supply and demand curves? (I guess this is also to ask why (p x q) should ever be a useful measure of costs and benefits, even without irrationality. Two otherwise identical sets of marginal cost and benefit curves could give radically different answers based on (p x q) solely on the basis of ability to price discriminate.)
Posted by: conchis at Apr 23, 2007 9:38:47 AM
Here's an argument you could make in favor of the U.S. system.
The U.S.'s results are inferior to France, but the nation is much more diverse.
The U.S.'s results are superior to Brazil, but the nation is more advanced.
So we land somewhere in between France and Brazil. That makes the spending just about right.
Obviously this an extremely simplified argument, but I believe it has merits.
Posted by: Ted Craig at Apr 23, 2007 9:42:27 AM
Wouldn't one have to do the same calculation w.r.t. European health care costs to get a true comparison?
Posted by: Andy at Apr 23, 2007 9:49:11 AM
What about 'medical tourism'? The same medical procedures (broadly speaking) can be had at a fraction of the cost in (eg) India or Thailand. At least one very large, very new hospital is being built in Thailand expressly for 'medical tourists' from DCs.
What does the difference between the two costs (same procedure in DC/LDC) 'measure'? 'Overcharge' in the DC? Opportunity cost?
Posted by: Sudha Shenoy at Apr 23, 2007 10:23:36 AM
See also Arnold Kling, http://econlog.econlib.org/archives/2007/04/tyler_cowen_cha.html. I don't think the entire difference is one of prices, though.
Posted by: Tyler Cowen at Apr 23, 2007 10:34:27 AM
This is all a lot of mental masturbation. All this terms you use are subjective, and that you represent "French health care" and "us health care" as some sort of monolithic experience or reality is bogus.
How many people consider the cost of anything? They consider the price and their ability to pay it - in the present or over time.
The better question to ask is how to ensure all people have access to good health care and how you accomplish that without sucking up all their future income.
Posted by: ajkrik at Apr 23, 2007 10:47:31 AM
I think perhaps the medical outcome measurements comparing the US to places like France and Canada may be to coarse. As has already pointed out, there is a strong genetic component to health, which I almost never see corrected for. Additionally there are strong behavioral aspects etc. All of these are being attributed to the health care system with the coarse measurements.
But when I look at studies of fine grained measurements, like what percentage of people with hypertention are
a) Being medicated
b) Having their hypertention controlled
the US comes out a fair bit ahead of Europe. I am curious to see other fined grained studies, because I strongly suspect that, for example, the US outcomes for those with most classes of cancer is better. I would also be very unsurprised to find that our outcomes for other very serious conditions is better when you look at the outcomes for those presenting with those illnesses.
Posted by: quadrupole at Apr 23, 2007 11:11:55 AM
wait a minute there. I see some sleight of hand going on. If redistribution of money has zero social cost, why do so many libertarian economists write screeds villifying the welfare state?
If it's actually true that some force is causing patients to pay twice what they "should" pay to doctors, then effectively we're getting a redistribution of money from patients to doctors (who appear to split these rents with health insurers, drug manufacturers and malpractice attorneys). Doesn't this cause perverse economic incentives and don't those changed decisions have a social cost?
Plus, from a utilitarian perspective, as long as eta is positive, the redistribution itself is a social cost even ignoring incentive effects, since almost all of the beneficiaries have more wealth and income than the average.
So no, I don't think you can quite handwave all that away.
Posted by: Michael Sullivan at Apr 23, 2007 11:33:08 AM
If there are a large number of patients paying $1,000 for a procedure that costs $500 to produce, isn't it logical to think that the doctor has spent a substantial part of that $500 surplus attracting the patient? There are a few barriers to entry, but it's hard to believe medicine is so uncompetitive market that producers are earning many economic profits.
Posted by: Nathan at Apr 23, 2007 1:23:27 PM
Michael,
Because you can choose not to have medical care for yourself (or your family). There is no choice in supporting the welfare state.
Posted by: Xmas at Apr 23, 2007 2:08:45 PM
Does this presage your Sraffian turn, Tyler? True costs rather than marginal costs?
Posted by: Dan Karreman at Apr 23, 2007 2:11:58 PM
ajkrik said:
"The better question to ask is how to ensure all people have access to good health care and how you accomplish that without sucking up all their future income."
While we are at it, why don't we ask how we can make humans fly? It is precisely because many market oriented people understand the limits of scarcity and how supply and demand allocates scarce resources through the pricing system, that we dont sit around arguing for pie in the sky hypotheticals such as "HEALTH CARE FOR ALL WITH NO TRADEOFFS."
Posted by: John Pertz at Apr 23, 2007 3:22:34 PM
National health schemes tend to ration care, keeping 'costs' lower. But to the extent that people are sicker longer (waiting for treatment), they work and produce less. Where does this opportunity cost of production forgone show up in these cost comparisons?
An example - Australia looks likely to finally add remicaid (a successful but expensive drug for treating inflammatory disease of the bowel) to the Pharmaceutical Benefits Scheme, years after it was accepted by private health insurance in the US. Private insurance had an incentive to pay for this treatment since it can prevent the need for expensive multiple surgeries. In the meantime 28,000 patients have experienced severe pain, removal of large parts of their bowel in many cases, inability to work, and so on. Where does this cost manifest in GDP?
Posted by: PJ at Apr 23, 2007 3:27:35 PM
If I understand the argument, the doctor provides $500 worth of services but ("fraudulently") charges $1,000. This means there's $500 of legitimate services and $500 just transferred to the doctor for nothing. So, the cost in resources is the $500 of effort the doctor put in. The rest has no net cost to the economy--your loss at overspending by $500 is offset by the doctor's gain at getting an extra $500 for nothing.
This argument works fine if it's fraudulent overcharging without resources put out. But, if the doctor ordered an unnecessary test, or our system required a lot of unnecessary paperwork, then there a total of $1,000 of social cost. Society used $500 of doctor resources and wasted $500 on paperwork or useless tests that produce no value.
Posted by: John Pinkerton at Apr 23, 2007 4:21:20 PM
John Pertz: First of all, humans can fly. If you mean "flap flap flap I'm a bird", you're right it might be silly to debate if I take your post's meaning.
There is absolutely nothing "pie in the sky" (whatever the heck that is anyway) about ensuring that everyone has decent health available to them. If I understand your post it implies that all things are restrained absolutely by "market forces" and that providing health care freely can be wrought from "will" alone.
Of course it is a scarce resource be cause it is by definition limited. But I never said there would be no "tradeoffs."
It is precisely because people sit around mentally pleasuring themselves with intellectual debate about how "supply and demand allocates scarce resources through the pricing system" that nobody bothers to work out the logistics of actually getting it done.
I know it can be done - i.e. providing quality health care to everyone in an economy. There are far too many faulty assumptions in the argument that "scarce" resources preclude providing everyone a piece of this pie.
People can fly, it just depends on what the meaning if "is" is.
Posted by: ajkrik at Apr 23, 2007 7:51:34 PM
From the CIA world fact book – US GDP per capita is $43,600 in 2006. For France the number is $30,100 using PPP.
If the French had the additional $13,500 per capita that would bring them up to parity with the US, how much of it would they choose to spend on health care versus something else?
Turn the question around. If we started with the same per capita GDP as France and then allocated the rest to get up to current US GDP per capita, how much would we allocate to healthcare versus something else.
I suspect that much of our healthcare spending is simply because we can afford it.
Posted by: Scott at Apr 23, 2007 8:14:56 PM
The waste in medicine isn't a transfer.
Massive rent-seeking dissipates the potential transfer. I don't see many people getting rich from health care.
Posted by: Douglas Knight at Apr 23, 2007 9:28:58 PM
I think TC is quibbling about terminology here. From the point of view of a typical American or a Medicare actuary, a transfer to a special group constitutes a cost, even though, technically speaking, it may not be economically inefficient. And as others have pointed out, there are economic inefficiencies involved in actuating those transfers. Attempts to capture those rents will tend to attract resources away from the productive sector until the rents are fully dissipated.
Posted by: knzn at Apr 24, 2007 3:01:37 PM
Don't forget the unequal distribution of that cost, and the economic distortions introduced by tying healthcare to employment while at the same time providing little to no employment security.
Posted by: Peter Clay at Apr 25, 2007 7:29:56 AM
大家好，我是臺灣人，從臺灣一個人搬家來到美國，環境很陌生，感覺很孤單。以前在臺灣幾家知名的徵信社工作過，我是一個優秀的徵信工作者，希望早點找到適合自己的工作。希望通過貴站，認識更多的朋友。
Posted by: 謝文豪 at Apr 2, 2008 2:36:28 AM
