« The economics of "bailouts" | Main | The realignment of the regulatory powers »
Sherry Glied's new health care paper
It is one of the best health care papers in recent times, it is here, I cannot find an ungated version. Glied reminds us that only about 1/3 of American health care spending comes from private insurance. Moving to international comparisons, the more general point is that:
...there is no persistent and regular relationship between the structure of system financing and the rate of growth in per capita health expenditures in a health system...the efficiency of operation of the health care system itself appears to depend much more on how providers are paid and how the delivery of care is organized than on the method used to raise the funds.
In other words, as I've stressed before, the health care cost problem comes from immediate suppliers, namely doctors and hospitals, and not from health insurance companies.
The best parts of the paper concern equity. It is GPs which help the poor, not additional spending on technology or surgery; see p.18 for other comparisons along these lines. Furthermore, and this you should scream from the rooftops, consider this:
...patterns of health service utilization in developed countries suggest that the marginal dollar of health care spending -- money used to purchase high tech equipment or specialist services -- is less progressively spent than the average dollar.
In other words, egalitarians should not allocate marginal government spending to health care. And there is evidence that the more a government spends on health care, the less it spends helping people in money ways. That is, there is crowding out.
Finally, Glied offers a summary comparison:
Putting $1 of tax funds into the public health insurance system effectively channels between $0.23 and $0.26 toward the lowest income quintile people, and about $0.50 to the bottom two income quintiles. Finally, a review of the literature across the OECD suggests that the progressivity of financing of the health insurance system has limited implications for overall income inequality, particularly over time.
Highly recommended.
Posted by Tyler Cowen on March 19, 2008 at 07:45 AM in Medicine | Permalink
Comments
Defenders of America's current health care system often point to the waiting times for hip-replacement surgery as evidence of our system's superiority. "Americans can get new hips in a matter of days," they boast, "while the Canadians and Europeans have to suffer for months until their socialized medicine comes through."
What this argument overlooks is that private insurance pays for only a small percentage of hip replacements in America. Most are paid by Medicare, and the remainder are mainly Medicaid.
Posted by: Peter at Mar 19, 2008 9:24:21 AM
You say "the more a government spends on health care, the less it spends helping people in money ways." Would you please specify how you'd like to see the government help people in other ways? It seems to me that more often than not, every gov't program (or any tax increase necessary to pay for it) is routinely scorned on this and other libertarian blogs.
Also, please don't use this paper to flog the idea of ending employer-purchased insurance, or otherwise fragmenting the risk pool. Here's the likely consequence of that, from a Jon Cohn article:
"Absent a substantial restructuring of the insurance industry, the people with preexisting medical conditions . . . would still struggle to find decent coverage outside of the workplace. So it's quite possible that only relatively healthy people would opt into the individual market.
"Once these more robust specimens fled employer groups, however, the cost of insurance for those remaining behind would go up--since insurance becomes more expensive without the contributions of relatively healthy people to offset the costs of those with high medical bills. And, as the cost of the employer insurance went up, even more people would start dropping it (or their companies might simply stop offering it). Appearing on ABC's "This Week" recently, McCain told George Stephanopoulos that "I want the families to make the choice." But, for Americans who are sick or poor or both, the McCain plan could mean fewer insurance choices than they have now--or no choice at all."
from http://www.tnr.com/politics/story.html?id=7e9b013b-2fb7-45c7-91cb-e05218063a33
Posted by: Frank at Mar 19, 2008 9:51:40 AM
IMHO In the discussion of healthcare we should always make clear the distinction between costs and spending. I have heard some economist say that costs are not rising but spending is rising. Cost may be to too high due to excessive licensing but I understand that we are getting more and better treatments (probably more than it beneficial) and that this is the real problem.
So did Tyler really mean to say “the health care cost problem” or the health care spending problem? He could mean the health insurance cost problem since it is difficult to buy insurance that does not cover too much (although my Blue cross policy seems reasonably affordable at $303/month for my family of 4, one of us a breast cancer survivor and who also had a $30,000 bill for another condition 2 years ago).
Posted by: Floccina at Mar 19, 2008 11:04:07 AM
Frank this:
So it's quite possible that only relatively healthy people would opt into the individual market.
..has been shown to not be true. It is the risk averse, who also tend to be more healthy, who opt to buy health insurance. They could be described as wise (conservative/risk averse) people. It seems that Government action usually hurts the wiser more conservative/risk averse. For example with the current bailout of bad mortgages the wiser more conservative/risk averse people who have been sitting on cash waiting for/hoping for a housing crash to buy are hurt to help the un-wise.
Perhaps this is a good thing because people are by nature risk averse but I am not sure. One could argue that Gov. encouraging risk taking is a good thing. I do not know.
Posted by: Floccina at Mar 19, 2008 11:17:28 AM
Frank,
As always, you're a thoughtful commenter regarding the social welfare implications of particular policy prescriptions. But your perception of libertarian skepticism is quite off-topic, and possibly straw man. Libertarian skepticism of state power as a means of improving social welfare is not categorical, at least not on this blog. This skepticism simply acknowledges that a solution based on state power is bound to have negative public choice implications that are generally ignored by policy prescriptions at hand, and that the liberty interest of individuals place the burden of proof on those advocating for state power (including accounting for public choice considerations) before the experiment actually begins, not on those who would be the unwilling subjects in the experiment.
Posted by: M. Hodak at Mar 19, 2008 11:20:28 AM
A "pre-existing condition" is not a risk. It is a certainty. It is an aberration that it should be paid for by a risk-management product like insurance.
Should there be "egalitarianizing" mechanisms to pay people for pre-existing conditions, forcing the healthy to pay for the sick? Maybe, maybe not. In the case of OBVIOUS behavior-related conditions, as in smoking-related, it seems not.
Posted by: Andrew at Mar 19, 2008 11:33:10 AM
"egalitarians should not allocate marginal government spending to health care"
But should we allocate it to ~primary~ care?
Posted by: Bill Gardner at Mar 19, 2008 11:39:13 AM
Of course, you like the paper that supports exactly what you think already.
Posted by: Andrew at Mar 19, 2008 1:42:39 PM
Of course, you like the paper that supports exactly what you think already.
Posted by: Andrew at Mar 19, 2008 1:42:41 PM
"...the efficiency of operation of the health care system itself appears to depend much more on how providers are paid and how the delivery of care is organized than on the method used to raise the funds."
An important complaint about the current system of private health insurance in the US is the fact that their pay/delivery system is so inefficient (e.g., compared to Medicare). It seems as if the incentives for the health care industry (profit) are poorly aligned with the societal incentive to manage the cost of health care delivery.
I suppose I should go read the original paper, but the quotes would lead me to guess that the paper is going to argue against government spending on universal healthcare. I don't think I'm likely to be convinced by data that current government spending hasn't been done correctly yet, though. No existing central planning nor market approaches seem to do the job quite right yet. It's a Hard problem.
I understand the libertarian impulse to improve the market functioning here in order to improve efficiency and manage costs. However, morally, it's tough to stomach the dramatic inequality that market forces generally provide.
My own intuition is that it's not that helpful to focus exclusively on the overall cost of healthcare. If we get progressively better healthcare at increasing costs, I think that's acceptable. The health care provided to the well-off in the US is also doing pretty well overall. The problem is that the not-well-off do very poorly in the US system. But rather than frame this as just a problem of reducing inequality, I think it would be better to focus on the direct question of how to make healthcare better than it is for the poor.
If your goal is to improve the healthcare available to the poor, what is the best market or regulatory mechanism for this? If you can design a market mechanism that improves healthcare for the poor, sign up me. I suspect you're going to need central planning, though. Probaly central planning with a fairly deft touch to avoid constaining innovation and to allocate somewhat efficienty (e.g., more GPs and fewer MRIs, as the paper suggests). Yes, we should be skeptical of the possibility of high-quality central planning. But I have yet to see a better choice outlined.
Posted by: Paul J. Reber at Mar 19, 2008 1:49:45 PM
Paul J. Reber wrote:
My own intuition is that it's not that helpful to focus exclusively on the overall cost of healthcare. If we get progressively better healthcare at increasing costs, I think that's acceptable.
Do not forget that the returns to healthcare beyond the really cheap basics like vaccinations, antibiotics, emergency care and food fortification are very small. They may even be smaller that the benefits to health from being a little richer in non healthcare ways (like being able to afford a newer safer car or to live in a more rural area or join a workout club, or to work a safer job). Also the poorer one is the more one would presumably be willing to trade health and longevity for more money (fishermen and coal miners come to mind).
Paul J. Reber wrote:
The health care provided to the well-off in the US is also doing pretty well overall. The problem is that the not-well-off do very poorly in the US system. But rather than frame this as just a problem of reducing inequality, I think it would be better to focus on the direct question of how to make healthcare better than it is for the poor.
Some studies have shown that the difference in health between the rich and the poor is even greater in Canada, Australia, New Zealand and the UK than in the USA. Some people have proposed a reason for this is that if you are of the low classes in society it is good for your doctor to see you as walking dollar sign.
Posted by: Floccina at Mar 19, 2008 2:16:41 PM
There's also the broader question of whether the government should be spending any tax money at all on health care for citizens for any form of "universal coverage", or whether individuals should be allowed to spend their own health care dollars according to their own best judgment for their own benefit, free of government interference (apart from laws protecting against theft and fraud).
For those who are interested in a defense of a fully free market in health care, Lin Zinser and I have written an article on health care history and policy entitled "Moral Health Care vs. 'Universal Health Care'" which recently appeared in the Winter 2007-2008 issue of the journal, "The Objective Standard".
We argue that the current crisis in American health care is the result of decades of government interference and violations of individual rights in health insurance and medicine. Hence the only moral and practical solution to the problem is not more government controls but instead to gradually and systematically transition to a rights-respecting, fully free market in those industries.
The full text of the article is available online for free at:
http://www.theobjectivestandard.com/issues/2007-winter/moral-vs-universal-health-care.asp
or http://tinyurl.com/25zffu
Paul Hsieh, MD
Freedom and Individual Rights in Medicine: http://www.WeStandFIRM.org
Posted by: Paul Hsieh, MD at Mar 19, 2008 2:17:51 PM
A visit to a GP costs 3 to 5 times more in the US than in France. Malpractice insurance costs are often used as an explanation, but the median income of medical doctors after deducting said insurance is $250K/year.
The root cause of the health care crisis is pretty clear: providers have organized themselves into a cartel. The solution is to break up the AMA, transfer licensing to a truly independent authority rather than a doctor's trade union. The transition would be rough, but could be handled using military doctors to vet med school programs much as Reagan used military air traffic controllers to break the ATC strik in the eighties.
Posted by: Fazal Majid at Mar 19, 2008 3:37:17 PM
Frank,
Andrew (@ 11:33) is right. The concepts of insurance and health care subsidization need to be separated. If you think healthy people should subsidize the ill, that is a completely different issue from the proper mechanism for an efficient insurance system. Insurance exists so that people can pay a premium to get rid of large risks.
Someone who already has a condition does not have a risk to get rid of. It would be similar to a person with a house that is on fire and gasoline in the basement seeking fire insurance. No insurer would issue a policy to such a person, or if they did the cost would be the value of the house. The cost at that point is no longer uncertain. The point of fire insurance is of course not to make people with safe or fire-proof houses subsidize those who have flimsy and fire-prone houses.
Unfortunately, group-plans have the side effect of subsidization, which seems to muddle the issue. If you think ill people should be compensated by healthy people for their illness, that would be much better done by a direct transfer.
Posted by: Cliff at Mar 19, 2008 4:00:12 PM
This sounds like a very valuable paper. However, though I have only your quotes to go on, it seems to me you are drawing lessons from them that the quotes do not support.
the marginal dollar of health care spending -- money used to purchase high tech equipment or specialist services -- is less progressively spent than the average dollar.
"In other words, egalitarians should not allocate marginal government spending to health care."
It does not seem to me that that is what the quote is saying. If I understand it correctly, it is merely saying that funding for basic services and general practice is more likely to benefit the poor or medically-underserved (however you interpret "progressive" vs. "regressive" healthcare) than is money for high-end resources. This could be true for at least a couple of reasons: the less-privileged segments of society lack basic services as well as access to advanced care, and must get basic services before advanced care is even meaningful for them, so of course marginal investments do not benefit them immediately until the basic-care gap has been closed; or, given the gap between the privileged and under-privileged in access to basic care, which serves as the gateway to advanced care, investment in high-end services will only benefit the more privileged, who are the ones who get the tests and referrals that make it available - and so, again, such investments will be regressive until the basic-care gap is closed. But neither of those explanations, or the phenomenon of regressivity in general, warrant the conclusion that we should never invest in high-end care. The obvious solution would seem to be to invest as much as is necessary to ensure universal basic care, and then add in marginal dollars to the extent we think we can afford them. In other words, the solution to the problem of unequal benefit from marginal investment is not to abandon advancements in high-end infrastructure entirely, but rather to eliminate the causes of the unequal benefit.
"there is evidence that the more a government spends on health care, the less it spends helping people in money ways. That is, there is crowding out."
Well, that would have to be true. It is true for any two categories of expenditure. (Recall "guns vs. butter"?) That is not an argument against healthcare spending, or against government as a source of healthcare spending. It is simply an observation that all spending decisions take place in the context of a competition among global budgeting priorities.
Putting $1 of tax funds into the public health insurance system effectively channels between $0.23 and $0.26 toward the lowest income quintile people, and about $0.50 to the bottom two income quintiles. Finally, a review of the literature across the OECD suggests that the progressivity of financing of the health insurance system has limited implications for overall income inequality, particularly over time.
As she notes, the former sentence implies that government-funded healthcare provision is "modestly redistributive", in the progressive direction - which is a good thing. The second sentence merely implies that healthcare expenditures, while progressive in terms of the distribution of healthcare resources per se, are not a mechanism for overall income redistribution - which is hardly surprising, because they are in general not meant to be. For those who seek redistributive economic policies, this means only that healthcare stands aloof from any such goals; for those who oppose them, this means that they have no grounds for opposing publicly-funded healthcare out of fear of its possible contribution to the collapse of class barriers. These also are good things.
Posted by: Kevin T. Keith at Mar 19, 2008 4:13:39 PM
So, immediate providers are the cost problem. Does this mean that if doctors went back
to visiting people in their homes like they used to do in this country and still do in
countries such as France that are widely viewed as having better (and less expensive)
health care systems than the US, this would help contain our cost problem and improve
our health outcomes?
Posted by: Barkley Rosser at Mar 19, 2008 4:49:33 PM
Is the individual market even insurance? Insurance is the pooling of risk. (Or is my understanding and definition too simplistic?) Can an individual form a pool? Some may say compare individual insurance to auto insurance. But state mandate that every car owner have insurance. Therefore costs and risks should fall on a per capita basis because they are spread over de facto pools of drivers and further segmented by neighborhoods, age groups and vehicle types. Do so-called individual health insurance polices work in a similar fashion?
Posted by: tec619 at Mar 19, 2008 5:49:57 PM
The cost problem of health care comes from providers and not insurance? Isn't that common knowledge among people who follow economics and/or medicine?
Anyway, I'm with Fazal and Milton Friedman--end the licensing, end the AMA's control of the business, open up the division of labor, open up the markets. Until this is done, the labor will get more expensive, visits to a doctor or hospital will skyrocket in price, and the problem will continue to get worse and worse. There also needs to be heavy deregulation. Not wholesale deregulation, but a lot less direct controls, and a lot less of the kind of regulation that serves largely to inhibit competition at the expense of consumers.
Posted by: Jacob Oost at Mar 19, 2008 7:35:40 PM
"the efficiency of operation of the health care system itself appears to depend much more on how providers are paid and how the delivery of care is organized than on the method used to raise the funds"
and
"In other words, as I've stressed before, the health care cost problem comes from immediate suppliers, namely doctors and hospitals, and not from health insurance companies."
"how providers are paid" and "how delivery of care is organized" are bother based upon what insurance will cover which is based upon what medical coders code for what care providers provide based on what insurance will cover. insurance runs the business.
Posted by: c at Mar 20, 2008 10:25:47 AM
tec619,
Yes, the individual market is insurance. It is the group market that is not insurance (or rather, is insurance plus wealth redistribution). In every type of insurance, premiums are based on individualized actuarial calculations and not the characteristics of an arbitrary group.
It is not that risk is "pooled" as in, net risk shared equally among unequal people. What happens is that a company with a large amount of capital can place bets against the insured event (house burning down, getting sick) that are profitable in the long run because the insured is willing to pay a premium to reduce the variance in costs (in other words smooth out the costs over the long run, paying a small amount repeatedly rather than a large amount unpredictably every once in a while).
As long as the bets are not correlated, for example because if one house burns down it is essentially no more likely that another house will also burn down, the risk for the company is minimal. Because they have a large portfolio of bets, mathematics can predict pretty accurately the probability of any given magnitude of gain or loss, which is essentially a function of randomness and on average is profitable for the company.
So in some sense you might say the risks are "pooled." This is the same sense in which individual health care insurance would pool health care risks.
However, the premium paid by each person depends solely on the probability that the insurance company will have to pay out to that person. So people who are healthy pay less than people who are ill, etc. Each person pays slightly more than the expected costs to the person (of health care expenditures, fire, whatever) to provide a profit expectation to the company.
Thus you can see that a) insurance does not save any money over the long run but rather increases costs (but prevents sudden catastrophic expenses that would bankrupt a person). And b) any insurance scheme that involves an arbitrary group of people all paying the same premium is a distribution of wealth from those less likely to file a claim to those more likely to file a claim.
Posted by: Cliff at Mar 20, 2008 11:43:09 AM
