Health Care Spending: International Comparisons
Matthew Yglesias notes the following expenditures for health care for different countries.
- United States: $5,267 on health care/ $2,364 is government spending.
- Canada: $2,931 on health care / $2,048 is government spending.
- France: $2,736 on health care / $2,080 is government spending.
My question is so what? Do the above numbers take into account what basically amounts to a large subsidy that many countries benefit from when it comes to pharmaceuticals? Does it take into account differences in populations? For example certain segments of the population tend to suffer from certain diseases such as sickle cell anemia, Tay-Sachs disease. Not to mention that Americans are some of the fattest people on the planet. What about issues like violence such as gun violence. Do these things increase health care expenditures or are they all trivial (and another possibility is that while each one is individually small in its impact is the combined effect significant)? What about differences in laws and medical policies. For example, how do other countries catagorize and treat premature/low birth weight babies? Are they treated the same as here in the U.S.? In the U.S. these babies are treated as live births and many of these children fight mightily (and consume large amounts of health care resources) and die. A double whammy from the perspective of gross aggregate statistics in that both health care expenditures go up, as does the infant mortality rate. Similarly for the eldelry. Do any of these European countries have assisted suicide laws that could also result in lower health care expenditures?
The point is that such comparisons of gross statistics as this could be quite misleading. Spouting out these numbers as if they show us something is at best misleading and at worst is intentionally dishonest. How come the supporters of socialized medicine never report things like wait times for various procedures? I bet France is pretty good, because from what I’ve read their system is pretty good (but many people in France also pay a fair amount out of their own pocket). Also, what about success rates for various procedures? What is the survival rate for various heart operations in the U.S., Canada, France and yes even Sweden? Maybe the U.S. ranks low in these areas too, but shouldn’t we look at the statistics on these things as well?
As for the French system and paying out of pocket, I’ve read where often the out-of-pocket expense can often be 30% or more (for example, some prescription drug plans cover only 35% of the costs). Gee could this be a factor in keeping health care expenditures low? Here in the U.S. people like to get as much as they can for as little as they can when it comes to health care (and who wouldn’t?). So, people like a health plan that costs their employer quite a bit, but also has “free” eye-care so they don’t have to pay for glasses. Of course, this is “free” to the employee either, but the perception is that it is and with the way employer provided health care works it subsidizes those who needs glasses at the expense of those who don’t. But the bottom line is that there is little out-of-pocket expenditure. Could this be one factor in driving up health care expenditures?
This kind of brain dead comparison across countries isn’t really that helpful. The implication of these simpleton posts is that if we simply switched to the same system as the French we could save $2,500/year and that isn’t chump change for most of us. But does this have to be the case? Does it even have to be the case that we’d save $1/year? This is ultimately the real question (from a utilitarian/money stand point) and these kinds of moronic posts don’t even come close to answering this question.
Comparisons like the ones you have above do give much more heat than light—their main function is to outrage rather than to enlighten. But that doesn’t mean that enlightment isn’t possible or necessary. In order to make prudent choices we need more and better statistics not fewer or poorer.
For example, I’d really like to see some statistics that actually demonstrate that the U. S. health care system is genuinely excellent. Too many (I’m not saying you, Steve) of the opponents of change in the status quo rely solely on bland assertions.
I’d like to see some real statistics that demonstrate the excess demand theory. Rising costs do not ipso facto mean that there’s excess demand. I just don’t believe that health care demand curves have the kinds of shapes that are being described. As I’ve said before if I sprain my ankle I don’t want both ankles X-rayed no matter who bears the freight.
The statistics do, at least, point out that our supposedly “private” healtcare system is actually paid for by quite a bit of public funds.
And as you point out, people expect to go to the doctor and not pay anything at all. That’s not a free market.
I believe in free markets. When there is no free market, then you get an inefficient allocation of resources.
One of the problems with our healthcare system is financial. Too many people go to see a doctor, the doctor orders a couple of tests and because they only have to come up with a $20 co-pay, fail to even ASK what all those tests will cost and if they are truly needed. The patient may find out later from doctor or insurance statements that that one vist acutally cost several hundred dollars.
This was less of a problem when insurance would reimburse the patient for some of the charges because the patient had to pay the doctor up front and was more likely to question the doctor. Now, with benefit assignment, doctors can charge for excessive and unneeded services at will. Each consumer should be actively involved in controlling their own healthcare costs by asking questions and demanding answers before accepting services.
Example problems:
The US has the highest proportion of teen births. Teen mothers are much more likely to have low birth weight babies with problems.
Here’s an article written without much heat about it. Notice that our over 65s spend considerably much more than France’s as well, in the same proportions as our younger ages groups do when compared to similar age groups in France. And our over 65s have primarily government-run health care, like in France. Is there any reason then to expect that expanding Medicare to everyone would lead to a change, then?
Thanks Dave. You’re right, I do think the current system has lots of problems and there is much room for improvement. The problem is I am not convinced that socializing medicine is the way to go. It could, at least in theory, be a rotten way for the U.S. to go…or using another countries system could be really good.
We don’t have a private system. Note that about 45% of the spending is from the government. Could that government spending plus the current legal and regulatory institutions be part of the problem? I’m willing to entertain that notion if those on the Left would be willing to entertain the notion that we need more market incentives might also help.
Excellent post, Steve.
You can’t test untestable hypotheses. “The United States health care system is excellent” is an untestable hypothesis.
I question, therefore, whether trying to quantify such an entity is relevant.
But if you insist on trying, here are some questions:
a) If you want health care, can you get it?
b) Are you pleased with the services when provided? Do you think they were worth the money?
c) Do doctors come to the US to train, or do doctors go abroad to train?
d) Do patients come to the US for services, or do patients go abroad for services?
e) (and here we start to go down the slippery slope of poorly defined tests) … for a given procedure on patients with given risk factors, what is the “success” rate?
f) If you are 65 years old, what proportion of the population has had a successful knee replacement? Hip replacement? Age 70? Age 80?
g) What is the conditional life expectancy for nationals? (i.e., ignoring immigrant populations that were not able to partake of the nation’s healthcare system in the 1950s, say)
h) Given a particular set of risk factors, what is the infant mortality (i.e, for an otherwise healthy baby delivered by a healthy mother, etc.)
None of these are terribly dispositive, however, and I am left again feeling that one’s priors cannot be disproved based upon the answers to any of those questions.
Regardless, the aggregate dollar volume is not even one of the questions in my list. What was the exchange rate at the time? What proportion of income does that represent? Does that represent personal preferences or demographics or fundamental cost differences? Even if you say that our paperwork costs are 25% higher than in other countries, this does not mean we are wasteful. We have different privacy restrictions, different reimbursement systems that encourage extra hospital admin but also leave at least some flexibility for growth, etc. We have to pay doctors more in this country to get them to take the job, we have a litigation system that, relatively speaking, favors the plaintiff. Therefore, all of this “waste” does provide us with things that we value, even if the relationship is not readily apparent …
I guess you can argue that those who defend the status quo aren’t trying to answer those questions, but neither are the critics of the status quo. I don’t know why the burden of proof should therefore rest on the “defense” in this argument.
I also think that the approach taken in Thacker’s post above is helpful. But I don’t think you can in general prove or disprove prior beliefs regarding the quality of US health care because you cannot clearly identify a testable hypothesis that completely describes the question.
Michael, have you ever actually attempted this strategy? I have on several occasions citing chapter, verse, and specific CDC studies. The typical response is that the doctor draws himself up to his full height and tells you to go to hell (in the nicest possible way of course).
Victor, what a great set of questions. You are an actuary aren’t you? I can answer the first couple of questions.
That depends on lot on where you are. In major metropolitan areas anyone can get the health care they need. How it gets paid for is another question entirely. In rural areas health care may be very hard to come by.
The last health care I was pleased with that I received is when I was wheeled into the hospital unconscious and bleeding 25 years ago (it’s pretty hard to be a rational optimizer under those circumstances, BTW). They did a great job. Since then, however, I’ve been physically and psychologically abused nearly every time I’ve sought health care and had actionable malpractice committed on me twice. It’s always been very, very overpriced.
Approximately 20% of all doctors practicing in the United States were trained overseas. In some areas (NYC, for example) it’s over 30%. In some specialties it’s over 40%. I think that probably dwarfs the number of doctors who train here and then return home to practice (or even that stay here to practice). It’s actually extremely difficult for foreigners to get into US medical schools.
It used to be mostly coming over here but the tied is turning. I suspect that the number that go overseas for treatment is at least as high as vice versa. I’ll do some digging.
If you could, try to avoid counting those who go overseas for plastic (okay, silicone) boobies. Seems to me like that could bias the results.
Dave Schuler: In answer to your question, I have done this with success on my few visits to doctors and dentists and refused services based on cost. The key is to remember that despite whatever attitude the doctor has, you are the consumer of thier services and you never surrender the right to say “NO”.
The typical response is that the doctor draws himself up to his full height and tells you to go to hell (in the nicest possible way of course).
Funny — professionals (doctors, accountants, lawyers, etc.) rarely try to pull that kind of thing on me; when they do, something on my face must start flashing like a B.S. detector, ’cause they almost invariably trail off into CYA mode in the other direction.
But I’ve been told I have a “don’t @#$! with me” kind of face…
Amusing that everyone is using that same page on French health care costs. Crooked Timber, me, you….
Note that it is not that certain plans only pay for 35% of the prescription. That is all the State system pays for. People then buy further insurance to cover the extra costs…and even then they face a heavy co-pay at certain hospitals.
And it is most certainly not a single provider system ,like the VA systems that some are talking up (or our own dearly beloved NHS, one of the few rich world health services that is truly awful).
Must thank Krugman (where Matt got those numbers) though, every time he starts to opine on how the Europeans do things I get paid for a few more articles as I explain how he’s not quite got his facts right. He’s my employment plan.
Well Tim I’m glad Krugman is doing some good then with his Op-Ed articles. 😉
This is only the pretense of an argument from Steve as to why citing these numbers as a prima facie case showing some of the problems with our system vs. others isn’t reliable.
For he hasn’t shown even ball park guesstimates of what the correction factors from these considerations might be, let alone that they fully account for the differences shown in costs.
Might they? Sure, I guess. Might they not? Possibly. And Steve himself agrees that it may or may not make enough difference to close this apparent cost factor gap.
So, Steve hasn’t actually argued that the raw number comparison is misleading, and only wrongly, for various reasons, seems to show at least a cost disadvantage to our health care delivery system. He suggests that may be the case, but then fails to offer the slightest data to show what, if any, adjustments to these numbers would be more accurate.
Not that I blame him. Digging through the weeds to try to get quantitative numbers on such factors would surely be a lot of work.
But since he declines to do this, no less than those he criticizes for ignoring these adjustment factors, he hasn’t shown they’ve erred in relying on this comparison.
An example from the past: in 1980, the American industrial worker was the most highly paid in the world. By about ’89, that figure had slipped to where we were no longer even in the top ten, but at the bottom of the second ten, about 18th, 19th highest paid.
Critics made the point that since necessities like food and clothing were so much cheaper here, American workers were doing better than a gross comparison would indicate. So the comparison of workers was done on what was called a purchase parity basis, not only what one earned, but how far that went, after necessities were paid for at prevailing (and differing) national prices.
By that measure, the American worker had not done as badly, but still, we had gone from best well paid, to about 9th. Considerably better, yes, but still indicating a large relative decline over a short period of time.
So, Steve’s total of alleged possible confounding effects could cut the apparent cost difference in half, and yet not change the basic conclusion already reached on a naive basis from the data as shown.
No…I haven’t. I have instead argued that they might be misleading, although I am pretty sure they are misleading.
Why should I do Matthew’s homework for him? I don’t get paid to dredge up such data and do the analysis. Matthew on the other hand does get paid to write articles. So does Krugman, Matthew’s source. Krugman gets paid to do nothing other than sit in his office and do research similar to this. Why isn’t he doing it? Could it be because it could undermine his point?
This would be a good thing for either Matt or Krugman to look into, but I doubt they’ll do it because it is so much easier to write a superficial pieces of crap and get paid the same amount of money.
In other words, my position stands. We probably wont see a savings of $2,500/year. Maybe there will be a savings, but we don’t no for sure until a more serious analysis is done.
Really? I thought my final concluding comments were,
This kind of brain dead comparison across countries isn’t really that helpful. The implication of these simpleton posts is that if we simply switched to the same system as the French we could save $2,500/year and that isn’t chump change for most of us. But does this have to be the case? Does it even have to be the case that we’d save $1/year? This is ultimately the real question (from a utilitarian/money stand point) and these kinds of moronic posts don’t even come close to answering this question.
Gee, guess I really didn’t write that. Tell me O’ Sage, what exactly did I write?
According to OECD data (table 10) from 2001, health care accounts for 13.9% of GDP (14.6% in 2002), and 44.6% of health care expenditures in the US were paid by public systems such as Medicare, Medicaid, the VA and other military care, public health clinics, and other programs. However, when one includes tax subsidies and public employee benefits, the current tax-financed share of health spending is nearly 60%.[Woolhandler] Government mandates and regulations add another layer of public expense to health care in the US. From 1970 to 1996, state and federal mandates increased 25-fold, an annual growth rate of 15%.
It is estimated that 15 percent of the total increase in health care costs (representing $10 billion in 2001) is attributed to government mandates and regulations. Indeed, the “Health Insurance Portability and Accountability Act (HIPAA) alone will add billions of dollars in new compliance costs to the healthcare system.â€Â[PriceWaterhouse]
Woolhandler S, Himmelstein D; Paying For National Health insurance – And Not Getting It; Health Affairs (21) July/Aug 2002, pp.89-95.
PriceWaterhouseCoopers, “The Factors Fueling Rising Healthcare Costs,” April 2002; pp.1-13