Amy Finkelstein: The Costs and Benefits of Universal Health Insurance

This WSJ commentary from Amy Finkelstein of MIT discusses the costs and benefits of adopting universal health insurance:

The Cost of Coverage, by Amy Finkelstein, Commentary, WSJ: Thanks to widespread concern about the millions of Americans without health insurance, several states have recently mandated universal coverage... Massachusetts enacted legislation...; other proposals are brewing in California, Pennsylvania and elsewhere. Such reforms are likely to affect health-care spending...

For evidence of how such programs can lead to increased spending, just look at the effects of the introduction of Medicare in 1966. Medicare provides health-insurance coverage to virtually all Americans aged 65 and over. Prior to its enactment, only about one-quarter of these individuals had any meaningful health insurance. As a result of Medicare's introduction, about three-quarters of the elderly ... gained health-insurance coverage. (For perspective, this is a similar increase in the share .. as ... will happen in Massachusetts under its new universal coverage program.)

Research I conducted shows that Medicare had a substantial effect on the health-care sector. By 1970, the program caused a 37% increase in hospital spending. This is an enormous number. If I extrapolate from the Medicare ... to ... the ... overall spread of insurance -- both public and private -- between 1950 and 1990, it suggests that it is responsible for about half of the sixfold growth in real per capita health-care spending during this period.

Why does increased health insurance lead to increased health spending? One factor is that when individuals have insurance, they tend to consume more health care. ...

Another reason is that hospitals and doctors respond to the increased demand for health care by changing some of the ways in which they practice medicine. For example, hospitals were more likely to adopt new medical technologies after Medicare... because ..., with greater insurance coverage, there were more people who could afford these new technologies. ... All of this contributes to higher health-care spending.

Of course, the effect of health insurance on health spending tells us only of the costs of expanding health insurance coverage. We can also ask, what are the benefits? And once again, we can learn something from the Medicare experience. Robin McKnight of the University of Oregon and I have examined the data, and as best we can tell, Medicare did not have any effect at reducing elderly mortality in its first 10 years of existence. Of course, mortality is only one measure of health, and it is possible that other aspects of health improved. It is also possible that in the long run, the new technologies adopted ... because of Medicare had important health benefits that our 10-year analysis would not capture.

While the health benefits from Medicare therefore remain uncertain, we found clear evidence of a different type of benefit: It provided substantial financial protection to the elderly. Prior to Medicare, they faced the risk of large out-of-pocket medical expenditures. About one in 10 of elderly individuals spent one-fifth of their annual income on medical expenses... By 1970, we estimate that Medicare had reduced this risk of extremely large out-of-pocket medical expenditures by half.

What all this means is that, as best we can tell, the elderly were not foregoing life-saving treatments prior to Medicare. Rather, they were getting these treatments, albeit at large or even enormous personal financial cost. But overall Medicare did not so much save lives as it did provide financial security. This is the goal of insurance -- not to prevent an awful event from occurring, but to make sure that if it does occur you are not devastated financially.

The Medicare experience offers valuable lessons for today. Recent state efforts to create universal health-insurance coverage would reduce the fraction of the population in a state without insurance by similar amounts as did the introduction of Medicare... And if that program is any guide, we may perhaps see changes in the structure of the health-care system, such as the development and adoption of new medical technologies. We are likely to witness an improvement in the financial security of the currently uninsured. There are also likely to be increases in health-care spending -- quite possibly substantial ones.

The paper she co-authored with Robin is noted and briefly discussed here. The paper itself is here. Here's a non-technical summary.

Here are some statistics related to the paper Robin has used in presentations. This table shows that medical spending is very skewed: If you order people according to their spending on health care, the top 10% of spenders account for 72% of all spending and the top 1% of spenders account for 30% of all spending. 

Share of health care spenders	Cumulative share of US medical spending
Top 1%	30%
Top 5%	58%
Top 10%	72%
Top 30%	90%
Total population	100%

Source: Berk & Monheit (1992)

It's the people in the tails, of course, that use most of the resources and receive most of the benefits from adopting universal insurance coverage.

Comments

[dale]

At first glance this paper just seems to consider the costs of converting uninsured people to insured. Of course there will be additional costs. Although, I wonder if they factor in the current costs of providing health care to the uninsured?

But the big thing to keep in mind is not just the costs of adding the currently uninsured- but the potential vast savings of converting all currently privately covered to a universal, single payer type system. There the savings of eliminating the overhead of the private insurers should be substantial. And if I read Krugman correctly- those savings may be enough to pay for adding the currently uninsured.

[Peter Schaeffer]

This strikes me as very real world. It also shows that any single-payer system will be extremely expensive and/or require deep rationing. As I have oft demonstrated, the Singapore approach (predominatly out-of-pocket payment with low prices) yields astoundingly cheaper care with 85% public satisfaction.

It also shows that a vast proportion of health care dollars yield no actual benefits. For different reasons, this is a subject that neither the left or right seem inclined to discuss.

[ninjaplease]

Outsource the patients to the far east! that'll drop those costs!

[Peter Schaeffer]

Ninjaplease,

Not in our system. Actually there is small (so far) trend towards offshore health care at lower cost.

Cheaper to adopt Singapore's system than export the U.S. to Singapore.

[Movie Guy]

Are we not already "outsourcing" some medical care to Asia? Operations, for example?

So said my Sony television...for about an hour a few weeks ago. Was that an embedded commercial message?

[DRR]

I had some Dental work done in Monterrey last year. Satisfactory results & uber cheap. Of course it was compunded by the joy I experienced at selling hard working American dentists down the river with my benedict arnold consuming but I'm just heartless like that.

[ilsm]

Movie Guy,

Telecom and data transmission capability exists for CT radiography, for one example and there are more, to be read in Mumbai where the pathologist makes less than 10% of a US pathologist.

But the AMA doesn't run docs im Mumbai, they are just the top .05% intellectually of a billion person population and are smart enough to make it there rather than emigrate.

We can outsource a lot of things.

[ilsm]

"just look " ..........

The piece is vitriole about single payer health assurance.

My peave is I can start with 'just look the defense budget was $100B in 2006 dollars in 1947, then they passed the national security act and it ballooned to $580B in constant dollars in 2006.'

The national security act is bad!

Medicare has been a huge success, and may even have made people over 65 more healthy, as well as more at ease about not being on 'welfare' when they need help, after working all their lives.

[anne]

"Why does increased health insurance lead to increased health spending? One factor is that when individuals have insurance, they tend to consume more health care."

Yes; less is more especially when it comes to the health of children as the Administration is now arguing. Whose children, though? What we need is less health care, less, less, less. So lunacy, so immorality ever continues.

[Peter Schaeffer]

ilsm,

Defenese spending in 1947 was 5.5% of GDP, in 2006 3.7%.

However, your response along with others shows that cost control is not likely to be a prominent feature of future U.S. health plans.

Amy Finkelstein is way ahead of you.

Maybe I should point out that Singapore's life expectancy is 4 years higher than the U.S. with medical outlays (percent of GDP) 75% lower. Of course, Singapore compensates by having lower infant mortality, general death rates, etc.

[Isabel]

If I remember well, the last time we discussed Singapore, Peter Schaeffer, we came to the conclusion (or at least I did!) that it's not at all unfrequent that people run out of resources when facing high healthcare bills. Correct me if I'm wrong, but I think that would be the equivalent of letting those top 1%, or 5%, spenders (that account for 30% and 58% of all expenses respectively) go untreated. It would certainly be a lot less people than the 47 million of insured (I'm not saying that ALL 47 million go untreated, mind you, but I would expect a fair amount of them to do so). It looks like a good deal. On the other hand, the odds are that among those are your old parents, or old grand parents, or your old self, according to your age. Tough isn't it?

[elvis]

Defenese spending in 1947 was 5.5% of GDP, in 2006 3.7%.
Yet we manage to outspend the rest of the world. We've come a long way.

Peter, the Singapore system looked interesting the last time you mentioned it. But I seem to remember some apples/oranges there. Also, how do they deal with poor folk? Are payments to clinics based on the patient's income?

[anne]

"However, your response along with others shows that cost control is not likely to be a prominent feature of future U.S. health plans."

Yes; there is no cost that is too costly for war and occupation of Iraq. After all, $2 trillion is cheap at twice the price.

Yes; there is no cost that is too costly for war for the military. After all, $622 billion a year is cheap at twice the price.

What is too expensive? Health care. Health care is always too expensive, especially for, say, the ill. Say what?

[ilsm]

Peter,

"Defenese spending in 1947 was 5.5% of GDP, in 2006 3.7%."

In 2006 the US spent 50% of the world expenditures on arms. We only have a bunch of terrorists hanging in caves and on internet cafes in Pakistan to worry about.

In 1946 the Reds were west of the Elbe, and the Europens were broken down.

There was far less foregone opportunity in the 1947 budget than in the 2006.

The 1947 defense budget privatized national security and created the military industrial complex allowing local opportunity costs to compete with national opportunity costs.

Local jobs programs versus opportunities lost in more productive spending.

Eisenhower recognized this in 1961.

[Isabel]

"Health care is always too expensive, especially for, say, the ill."

Indeed. As a matter of fact, one can imagine that the majority of those 47 million uninsured are in good health, like the rest of the population (70% of people spend 10% of money). So, they get to keep all their money to buy plasma TVs or whatever. Much better deal.

[save_the_rustbelt]

Mark:

The problem with the chart is that the people in the brackets change constantly.

My mother-in-law was a light utilizer for 78 3/4 years, and a heavy utilizer for 2 months. Fairly typical.

Some people with chronic conditions are heavy all of the time, most are not.

Trauma patients are extremely heavy utilizers, usually for a brief period of time.

No offense, but when economists look at healthcare they are looking at the entire forest, they usually know nothing about the trees. That worries me.

[Isabel]

With the Singapore system, the problem is solved: just let natural selection weed out the heavy users, whoever they are.

[baileyman]

And where in this work is included the analysis of the fraud worked into the system by industry lobbying?

[ilsm]

baileyman,

Fraud goes into the price of things.

[Peter Schaeffer]

ilsm,

In 1946 defense spending was 19.2% of GDP. Throughout the Cold War it was around 10% of GDP. The 1947 number (5.5%) was part of short-lived (1947 - 1950) military retrenchment after WWII. At 3.7%, the U.S. is well below the levels needed to contain the Soviets.

The United States faces conflicts other than terrorism. Various wars in Asia are quite plausible. Large scale resources will be needed to fight them. I am not advocating any of them. However, we shouldn't pretend that OBJ is our only potential or actual foe.

[Peter Schaeffer]

Isabel,

Care to produce any data to support your theory? I have already shown that death rates are lower in Singapore versus the U.S. (4.28 versus 8.26).

I find your reference to "natural selection" odd. Are you suggesting that the population of Singapore is genetically evolved?

[anne]

"Various wars in Asia are quite plausible."

Australia, I always knew Australians were a fiercely dangerous people. I think the problem is kangaroos. We must be ready to fight Australian kangaroos at a moments notice. Hand me the kangaroo traps, mate.

[anne]

Yes; there is the answer, war in Asia, war with Singapore and then steal those Singapoean health potions. There's no health like Singaporean health, if your healthy that is. There's no health insurance like Singaporean health insurance, if you already have health insurance that is.

[Peter Schaeffer]

Isabel,

There are anecdotal reports of people running out of money from health care costs in Singapore. However, I really rather doubt that this is a significant problem. Two points come to mind. First, the medical outcomes data is all positive. If any significant number of people were doing without health care, it should show up in the data. Second, consumer satisfaction is quite high (85%). Are 85% of Americans happy with our system?

If the top 1-5% of U.S. health care consumer went untreated, wouldn’t that show up in the data?

[Peter Schaeffer]

Anne,

As you well know, I am not part of the "invade the world, invite the world, in debt to the world" mindset (unlike, say ....). However, Iraq runs around 1% of GDP. If 50% of U.S. health care spending is wasted, then that would be 7+% of GDP.

[reason]

Contrary to some who see this as a right wing plot, this paper seems pretty balanced. What it doesn't address of course is the administration costs which is the essense of for example PK's argument (it only looks at medical costs however they are defined).

UHI would imply some real redistribution, but adequate modern medical treatment for the poor always will. (It will also never be possible to avoid some sort of rationing, the law of diminishing returns also applies to medical technology.) I think economists (and this is a theme recently) should reexamine their reluctance to distinguish between needs and wants.

Re Peter Schaefer und Singapore - I don't think you are necessarily wrong for standard medical treatment. But the very poor and catastrophic costs need special consideration. I'm always suspicious of pat simple solutions, there is always a catch.

[evagrius]

"Why does increased health insurance lead to increased health spending? One factor is that when individuals have insurance, they tend to consume more health care."

Logic dictates, therefore, that not having any insurance will result in individuals not "consuming health care".

"Not consuming". Again, a false view of health care as some type of commodity or product. I really wish that health care providers were included in defining health care.

It might be wise also, to realize that mortality is basically determined by how early one received health care and how well one ate . Early child health care and nutrition are the best preventative health measures and programs that a country can do for its population. I really wish people in the U.S. could grasp that but, somehow, it seems that health care is always seen as a response to catastrophe rather than as a steady, if unexciting, process of care.

[Peter Schaeffer]

Anne,

Kangaroos come in various shapes and sizes. Wallabies are smaller and very cute. The petting zoos in Oz are lots of fun for kids.

[anne]

"As I have oft demonstrated, the Singapore approach (predominantly out-of-pocket payment with low prices) yields astoundingly cheaper care with 85% public satisfaction."

Fine, then begin by simply describing the system, or referencing a article describing the system, and showing from where the low prices stem.

[Noni Mausa]

"Why does increased health insurance lead to increased health spending? One factor is that when individuals have insurance, they tend to consume more health care. ...Medicare did not have any effect at reducing elderly mortality in its first 10 years of existence"

Oh gimme a break. Two grand old arguments which have been disproven, at least internationally.

Do people "consume" more health care under universal coverage? I suppose so - but mostly when they need it. We in Canada have to be referred for unusual treatments, and this acts as a pretty good barrier against the hordes of hypochondriacs waiting at the gates. Left out of Amy's equations are the people with conditions which are never treated because they hate doctors, can't be bothered, or can't get through the door to see anyone.

As for mortality in the elderly, it's one-to-a-customer, no matter what you do.

The unspoken subtext to health care arguments, ignored by the studies which prove that properly treated patients are cheaper than those allowed to fall into emergency health situations and then rescued, is that dead people are even cheaper yet. A pine box from The Brick is cheaper than even an MRI, and its effects are permanent.

A strict monetary economic model of health care ignores the inherent value of living human beings, a value which has nothing to do with dollars. The two value systems interact, of course, but when the monetary value system rules, the unspoken effect is to depend on the moneysaving properties of neglect.

A baby is infinitely valuable, must be born, kept safe and warm, and this is correct and proper. But money for schooling? food in the fridge? medical care? safe parks and playgrounds? not so much.

As for Mr Schaeffer -- forget Amy's arguments comparing Medicare and non-Medicare health care in the USA, and look at a big fat international example right before your eyes. Every single Canadian has single-payer health coverage from cradle to grave, and our system is far cheaper than yours and has healthier and longer-lived citizens.

Sheesh.

Noni

[anne]

How do you transfer the savings of a health care system such as in Singapore to America? The savings seems to me to come from lower administrative costs, which is just what Amy Finkelstein is denying a reduced insurance company role can achieve here. I am however always interested in borrowing ideas internationally, and Singapore appears to have proper health care outcomes.

[anne]

Noni Mausa:

"Every single Canadian has single-payer health coverage from cradle to grave, and our system is far cheaper than yours and has healthier and longer-lived citizens."

Agreed, always agreed. I have absolutely no idea why the Canadian system cannot be a fine partial example for us.

"As for mortality in the elderly, it's one-to-a-customer, no matter what you do."

Clever :)

[ken melvin]

The Singapore System was debunked here before, and before. Running it around the bush again doesn't make it better.

Prior Medicare, when elderly got sick they spent what they had then died. This form of health care was known as pinch the purse, i.e., the Drs. pinched and if the patient had no money they were sent home to die.

[Noni Mausa]

Peter Schaeffer:"Cheaper to adopt Singapore's system than export the U.S. to Singapore."

Peter, we had this discussion, on this list, a scant 8 weeks ago, here: http://economistsview.typepad.com/economistsview/2007/01/paul_krugman_a_.html

And you were in on that discussion.

To summarize, Singapore has roughly 5% fewer old people than the US, 5% fewer minors, and such a tight housing market that few can stay there who are not employed. Their system appears to work because it has minimized the number of dependent people and filtered effectively for healthy and functional residents.

I too can have a cheap health care system if I just vote the dependent and sick people off the island.

But that's a feedlot, not a society.

Noni

[Peter Schaeffer]

Elvis,

As I remember it, poor people are handled via Medishield and the extensive public hospital system.

To the best of my knowledge, clinic reimbursement is not tied to patient income.

Reason,

“It will also never be possible to avoid some sort of rationing, the law of diminishing returns also applies to medical technology”

Indeed. Presumably this is why you post under “reason”.

The Singapore system is not that simple. It is quite true, that payment is predominantly out-of-pocket. However, there are many components (for the poor, etc.). Should I post all of the links for details?

What most folks don’t know is that modern Singapore was founded by avowed left-wing socialists. The UK actually considered interning Lee Kuan Yew as a communist when he returned to Singapore after WWII. Providing for all of Singapore’s people was never in doubt.

I recommend reading LKY’s memoirs for insights as to how the Singapore system evolved. True, they are around 2,000 pages. However, the sections addressing medical care aren’t that long.

[evagrius]

Does Singapore have pre-natal, post-natal and early infant health care and good nutrition for infants?

If it does, that would explain why their health outcomes are so good.

says...

"One factor is that when individuals have insurance, they tend to consume more health care."

What a tired argument. Who sits around saying I'm going to the Dr. today because I have insurance. Most people avoid the Dr. whether they have insurance or not.

Also, the population of Singapore is less than Atlanta so I am not sure of the relevance.

[anne]

Noni, I agree with you but I am trying to understand how self-insuring for medical care could work in America or Canada, since Singapore essentially calls for self-insuring with a provision for care for the poorer. Simply as an exercise, how might Canadians self-insure for health care? Say public provision of catastrophic care, with all other health care self-insured.... I am thinking this through, as an exercise.

[me]

"One factor is that when individuals have insurance, they tend to consume more health care."

What a tired argument. Who sits around saying I'm going to the Dr. today because I have insurance. Most people avoid the Dr. whether they have insurance or not.

Also, the population of Singapore is less than Atlanta so I am not sure of the relevance.

says...

how might Canadians self-insure for health care?

Right now self insurance is for things not covered by health care - like prescription drugs, eye exams, cosmetic surgery, etc.. The rules change slightly from province to province as there are no national standards.

Fortunately we have a generic drug industry so for most drugs people can request generic drugs. Most provinces also have plans to help out low income people with drugs.

[ilsm]

Peter,

1946 a lot of stuff from 1944 was being paid for.

Wars in Asia.

Malarky.

I am in the defesne business.

[anne]

A fine response as to why and how the Canadian health care system can and does work; thank you.

[Isabel]

"If the top 1-5% of U.S. health care consumer went untreated, wouldn’t that show up in the data?"

I actually don't know. I'm a little mathematically impared, and I'm not an expert in healthcare, but there is plenty of people here that can answer, for sure. It's an honest, albeit theoretical question: if 1% of americans were left untreated (let's assume, for the sake of the argument, that they would actually die, although, of course, some people have virtually endless resources) would that really show in the outcomes?

I'm not saying, of course, that that is what happens in Singapore. I'm just trying to understand if the "elimination" (sorry!) of heavy users could make a difference on the budget but not on the outcomes.

[anne]

Isabel, when you come on helpful articles on the Portuguese or Spanish health care systems, in English, please reference them.

[Bernard Yomtov]

As a result of Medicare's introduction, about three-quarters of the elderly ... gained health-insurance coverage. (For perspective, this is a similar increase in the share .. as ... will happen in Massachusetts under its new universal coverage program.)

I do not understand this comparison. In MA about 12% of the population is uninsured, not 75%. I take it that she means that the three-quarters of the elderly who got Medicare were about 12% of the total population.

But this is a silly comparison. MA's uninsured are precisely not the elderly, who have Medicare. Medicare is a very poor basis of comparison.

Besides, aren't there other benefits than what Finkelstein mentions? Let's think about medical technology for a minute. We are constantly being told that we must pay high prices for prescriptions because this encourages research. Doesn't paying for medical technology have the same effect? If Medicare stimulated research into medical technology didn't that provide some benefits not accounted for on Finkelstein's balance sheet?

[anne]

By the way, Isabel, are the prime European health care systems simply taken so much for granted and so well liked that relatively little is written about them?

[anne]

Bernard, I did not understand the statistical analogy either and still do not.

[Isabel]

Anne, I have the feeling that American radars only detect English-speaking realities. Hence the comparison with Canada and UK most of the times (both with a gate-keeper system, as far as I know), when there are systems, like the French and the Belgian one (to speak of two that I know better) that allow a freedom of choice that americans would greatly appreciate (and rightly so!). I suppose that lots of things are written about various health care systems, but if they are not written in English, they don't exist... It's a pity, because with a limited knowledge of the various solutions tried elsewhere, it's easy to buy into the silly old "socialized medicine" thing.

I'll try to find something useful in English!

[crack]

Singapore is a poor proxy for anyplace. It's form of government is basically benevolent nationalistic fascism. I like the place a lot, but much of its industry is at least part owned by the government. Any political dissent is stifled with libel lawsuits. The main english language paper is part owned by the government. To maintain the ratio of ethnicities, ethnic Chinese are given incentives are given to have children. There are valid reasons for the decisions that Singapore has made, but it really is a place like no other.

[piglet]

ilsm has made a correct point. Finkelstein says that health care spending increased after Medicare was created. But for several reasons, health care spending increased in all developed countries during that period, with or without a change in health insurance policy. Coincidence is not evidence of causation. It is likely that Finkelstein vastly overstates the cost effect of Medicare. She points out herself that the uninsured before Medicare used to have enormous health care costs that they paid out of pocket. These expenses just shifted, with the extremely valuable effect of providing financial security to those affected.

[piglet]

PS: "As I have oft demonstrated, the Singapore approach (predominatly out-of-pocket payment with low prices) yields astoundingly cheaper care with 85% public satisfaction."

This characterization of the Singapore model is not accurate. I have posted this in an
earlier thread but I gladly post it again: -------------------------------------

The Singapore health system

"Medisave is a compulsory medical savings scheme with funds available to meet a portion of future personal or immediate family’s hospitalisation, day surgery and certain outpatient expenses. (...)

Medishield is effectively a national insurance scheme for catastrophic illness that is intended to cover a significant component of medical expenses from major or prolonged illnesses that are not covered by Medisave. Medishield operates under a scheduled reimbursement system based on days of hospitalisation and type of surgical treatment, offset by individuals sharing costs by way of co-payments and deductibles. Premiums for Medishield (or private insurance alternatives) can be paid from an individual’s Medisave account. (...)

In addition to individuals self-financing through Medisave, Medishield and Eldershield, a significant portion of workers (and their dependents) are covered by private health insurance. Private health insurance, which is often funded by employers on behalf of employees, covers a diverse range of medical expenses that are not typically reimbursed under the 3M system.

Invariably, individuals will still need to pay for part of their medical expenses directly, even after receiving reimbursements from Medisave, Medishield or private health insurance. These amounts generally relate to deductibles, co-payments (under Medisave or Medishield) or for over the counter prescription drugs not covered by private health insurance." [This is the case *everywhere*, in varying degrees.]

This article shows that, contrary to PS's post, insurance, rather than out of pocket payment, does play an important role in the Singapore system. They are not paying their chemo therapy out of pocket after all. Unfortunately the article doesn't tell us how many individuals chose to buy insurance and how the premiums are structured (how affordable are they? Are they regressive? Is there discrimination and adverse selection?) What the article doesn't tell us is what role SOLIDARITY plays in the system. PS seems to assume that solidarity, a principal goal of almost any health care system in the world, is not necessary and even detrimental. Let individuals get along with what they have and everything will be fine. But it now appears that solidarity *does* operate in this system. And to the extent that it doesn't, we have heard about "the plight of Singaporeans who cannot afford healthcare".

The article claims that "the key to Singapore’s efficient health care system is the emphasis on the individual to assume responsibility towards their own health and, importantly, their own health expenditure." However it goes on to point out:

"The use of compulsory savings (that is, the Medisave account) has been very successful as the main source of private funding for hospital expenses. Another key focus of the Government has been to ensure that overall health expenditure does not fall victim to the significant inflationary pressures that have been evident throughout the world. This has been achieved by actively regulating the supply and prices of healthcare services in the country."

Compulsory savings and a strong government control over "the supply and prices of healthcare services" isn't exactly what I would call an "emphasis on individual responsibility". In the light of the above, The remarkable low cost has been achieved primarily by government regulation (which is the main difference to the US system!), and secondarily probably by rationing. A number of individuals reach a point where they are simply denied services because they can't afford it, they have spent their savings account and are not covered by any safety net. I don't doubt that health care cost can be somewhat reduced in this way but ethically I find it unacceptable.

[anne]

Isabel, you have already drawn attention several times to Belgium, and we know to an extent that France and Germany have mixed public-private systems. That we do not draw on such systems for ideas suggests a terrible arrogance in us, but mention France and in a flash we will learn of the heat wave several summers ago and the discussion will be stopped. Mention Canada or England, and we learn of waiting. On and on....

[ilsm]

anne,

All populations form a bell curve given enough numbers, and 'randomness'.

In the US we care more for the tails, good tail if we are selling, the bad tail if we are stopping the buy.

The tails may be better in our system but the 95% band in the middle is much better in other places.

[anne]

Notice this summary series:

http://ezraklein.typepad.com/blog/2005/04/health_care_fra.html

April 18, 2005

The Health of Nation: France
By Ezra Klein

Da' basics: France has a basic system of public health insurance that, as of January 2000, covers everybody in the nation. Before then, portions of the population lacked insurance. The reimbursement rates are wholly uniform, despite the fact that there are actually three health care funds, a main one covering most workers, and then one for the self-employed and one for agricultural workers....

http://ezraklein.typepad.com/blog/2005/04/the_health_of_n.html

April 19, 2005

The Health of Nations: England
By Ezra Klein

Da' Basics: Britain's health care system finds its roots in a document called the Beveridge report. The report argued that the health care system Britain had in the 40's -- which covered about half the country and used political patronage as its sorting mechanism -- should be combined with the rest of the country's fragmented social programs and administered in a uniform way. Thus the National Health Service was created....

[anne]

http://ezraklein.typepad.com/blog/2005/04/the_health_of_n_1.html

April 20, 2005

The Health of Nations: Oh, Canada!
By Ezra Klein

Da Basics: Canada care is unapologetic, no-holds-barred single-payer. The single-payer, by the way, is not Canada as a whole, but each specific province, so it's not quite as monolithic as we think. It's financed by taxes, but the taxes vary from province to province, so there is a certain amount of variation in how the system pays its bills. But I'm going to stay away from that -- keeping you guys still for health policy is dicey enough, if I start throwing in tax policy, my blog will have tumbleweeds blowing through it (and maybe a shoot-out in the saloon, but that's another story)....

http://ezraklein.typepad.com/blog/2005/04/health_of_natio.html

April 21, 2005

The Health of Nations: Germany
By Ezra Klein

Da Basics: Germany was the first nation to enact mandatory health insurance, doing so way back in 1883. The system is funded through employer contributions, with half the money coming from your paycheck and half coming from your employer. Participating Germans -- about 90% of the country -- are enrolled in 'sickness funds', some of which are organized by geographical region, some of which are organized by trade, and some of which are organized by company. The funds are a mix between private and public entities and are all nonprofit. They can't discriminate, and can't charge customers at different rates corresponding to their health/age/lifestyle. That means no cherry-picking....

[anne]

http://ezraklein.typepad.com/blog/2005/04/health_of_natio_1.html

April 22, 2005

The Health of Nations: Japan
By Ezra Klein

Da Basics: Japan's health insurance is another one of these employer-based systems, and has been since 1922. Universal insurance was achieved in 1961, through the National Health Insurance Act. Employers with 700 employees are required to operate insurance plans for workers and their dependents. The plans are called 'society-managed insurance'. About 1800 of these employer-run plans exist, with 85% of them being single company programs and the balance being jointly administered by two or more companies. The boards of these plans are 50% company reps and 50% worker reps, much like in Germany. Dependents are required to enroll in the plans and the whole thing is funded through payroll taxes. These employer-based, 'society-managed insurance' groups cover 26% of Japan's population....

[piglet]

Again to Finkelstein: This summary of her study notes:
"Finkelstein estimates that the introduction of Medicare was associated with a 23 percent increase in total hospital expenditures (for all ages) between 1965 and 1970, with even larger effects if her analysis is extended through 1975. Extrapolating from these estimates, Finkelstein speculates that the overall spread of health insurance between 1950 and 1990 may be able to explain at least 40 percent of that period's dramatic rise in real per capita health spending.

This conclusion differs markedly from the conventional thinking among economists that the spread of health insurance can explain only a small portion of the rise in health spending. This belief is based on the results of the Rand Health Insurance Experiment (HIE), one of the largest randomized, individual-level social experiments ever conducted in the United States. The HIE compared the spending of individuals randomly assigned to different health insurance plans. Based on these comparisons, the estimated impact of health insurance on hospital spending was at least five times smaller than Finkelstein's estimates of the impact of Medicare on hospital spending.

Finkelstein suggests that the reason for the apparent discrepancy is that market-wide changes in health insurance - such as the introduction of Medicare - may alter the nature and practice of medical care in ways that experiments affecting the health insurance of isolated individuals will not."
So the question is how well Finkelstein can establish that it was in fact Medicare that caused the cost increase. On the face of it, it doesn't appear to be plausible that Medicare had the effect of increasing young people's hospital exenditures. And the argument that Medicare changed "the nature and practice of medical care" calls for an international comparison. Can the same be observed in countries with universal health care, and if yes why do those countries manage to keep costs lower? If anybody has time to actually read the paper and tell us how good Finkelstein's evidence is, that would be appreciated.

[Peter Schaeffer]

Noni Mausa,

If the goal is to look at systems more efficient than the U.S., why not look at systems more efficient than Canada? Singapore runs at less than 50% of the cost of the Canadian system.

Anne,

The low cost of the Singapore system can not possibly come solely from lower administrative costs. Yes, they appear to play a part. However, the savings appear to arise from other factors as well.

1. Less futile care. The U.S expends countless dollars on care that yields no benefit at all. Anyone remember Terri Schiavo? To use an anecdotal example, my uncle died a few years ago. Incredible amounts of money were spent in the last weeks of his life. His family insisted on it. However, he bitterly opposed it. They kept him unconscious (with sedatives) to prevent him from ending his misery. What sense did they make?
2. Lower prices. Is it cast in stone that American health care prices are optimal? Drug prices are lower (with and without government regulation) in much of the rest of the world. Compare Prilosec and Nexium prices (they are the same molecule). Only in a country where “price is no object” would Nexium be widely prescribed. This kind of thinking pervades American health care. In a cost conscious system, doctors would prescribe Prilosec, if they prescribed anything at all.
3. Higher efficiency. No one would be surprised to hear that the U.S. can refine a barrel of crude oil with fewer resources than other countries (just an example). No one would be surprised to learn that the Japanese can build cars with fewer inputs. Why then is at a matter of faith that the U.S. is the optimal producer of health care?

Ken Melvin,

Please provide a link to where the Singapore system has been debunked.

Noni Mausa,

Yes, Singapore’s demographics are different than the U.S. However, I adjusted the Singapore data using U.S. demographics (percent over 65) and the U.S. elderly/non-elderly cost ratio. These changes raise Singapore’s costs by 0.7% of GDP (as I remember). Singapore remains astoundingly cheaper than the U.S. or Canada. As for the poor being forced out of Singapore, that doesn’t make any sense. Singapore’s neighbors are much (Indonesia) or substantially (Malaysia)) poorer than Singapore

Evagrius,

I can’t answer your questions. However, Singapore does have the lowest (apparently) infant mortality in the world.

Anne,

The Singapore system uses self-insurance for the middle and upper classes. The poor are handled via Medishield (as I remember) and the public hospital system.

Me,

How well is health care in Atlanta working?

Ilsm,

The 1946 data does (apparently) include post-WWII costs. The 1950 – 1970 data (roughly 10% of GDP) does not.

Isabel,

This isn’t too hard to guess at. Take a look at the death rate data. If curtailing care to the top 5% caused one out of five people to die, the death rate in Singapore would be much greater than the U.S. In fact, it is much lower. However, look at it another way, if 5% of the population was being cut off, would 85% be happy with the system? How many friends and relatives would the 5% have?

Crack,

When the U.S. manages to provide as good a life to its poor people as Singapore, I will be a bit more willing to accept criticism of Singapore. However, your claim that

“To maintain the ratio of ethnicities, ethnic Chinese are given incentives are given to have children”

is not true based on my knowledge. I have close friends from Singapore who assert otherwise. Note that they are not Chinese (ethnic Malays) and were part of the political opposition when they lived there.

All,

Correction. Medifund is the system for the poor, not Medishield. See Medisave, Medishield and Other Subsidy Schemes.

Piglet,

According to the WHO, 62% of all health care expenditures in Singapore are out-of-pocket. That qualifies as “predominantly” in my opinion.

However, I would be the first to agree that the Singapore system works because of the interaction of several factors including required savings and regulation. That is why I always refer to it as a “system”, not just as an example of the virtues of relying upon “out-of-pocket” health payment.

[Peter Schaeffer]

piglet,

Finkelstein wrote

"If I extrapolate from the Medicare ... to ... the ... overall spread of insurance -- both public and private -- between 1950 and 1990, it suggests that it is responsible for about half of the sixfold growth in real per capita health-care spending during this period."

Note that Medicare is not solely responsible for increased health care spending. Third party payment systems in general are according to Finkelstein.

[johnchx]

piglet writes: So the question is how well Finkelstein can establish that it was in fact Medicare that caused the cost increase.

Finklestein's paper can be downloaded without charge from her web page at NBER. Direct link to the paper here. Her 2005 working paper with McKnight (on the mortality and risk-reduction benefits of Medicare) is also available.

The evidence would appear to be pretty good. The key point, which doesn't come through in the summaries or the WSJ article, is that she uses regional variation in the levels of hospital insurance pre-Medicare to estimate the marginal impact of the implementation of Medicare. She also goes to great lengths to incorporate possible confounding factors, including overall trends in hospital spending, income, and fixed regional effects.

For those not entirely averse to appeals to authority, I'll point out that the paper has apparently been accepted for publication in the forthcoming QJE, which means that if there are any major statistical blunders, they've escaped the attention of referees much better trained than I am. FWIW.

I'm afraid the key point of the paper is in danger of getting lost here. The point is this: while there is actual (experimental!) evidence that indicates that individuals with health insurance consume more health care, the size of the increase is small in comparison with the size of the increases in health care spending we've observed over the last few decades (i.e. estimated in the range of 10% - 12% of the toal increase). However, Finklestein's work suggests that, in the aggregate, an increase in health insurance coverage may have an effect on total health care consumption much larger than the direct effect on the consumption of the individuals who get the new health insurance.

How this works -- why I spend more on health care when my neighbor gets health insurance -- is not well understood. And I, for one, would certainly like to understand it better.

[ilsm]

Peter,

You said: "The 1946 data does (apparently) include post-WWII costs. The 1950 – 1970 data (roughly 10% of GDP) does not."

You seem to subscribe to Hoover Institute. They think the US should set aside part of GDP for 'national security'.

That is wasteful unless you make your living from taxpayer largesse.

Certainly is business development without having to work hard and have real threats.

Working with percent of GDP denies the question: is the stuff needed? If needed is it effectively done?

The percent of GDP for national security was 10% or above in the years 1952 through 1959. It was 8.1 of GDP in 1970 the Vietnam debacle was going on then.

It blipped during the Reagan splurge to 6.1% a couple of times.

What was spent on national security from 1950 to 1970 was not really that effective nor efficient, we lost two wars during that period.

And there is no way the world today 'requires' the kind of spending for war materials that were spent then.

I also will note the CIA estimates were over blown throughout the cold war and the Soviets not that strong.

After 1970, the beginning of the demise of the cold war we built a lot of expensive things that are not well justified by need nor by the performance we get.

I think I can make the case for gross inflation rather than hedonic improvements. But that is a separate hypothesis.

Very expensive and the most obvious result is the checks the taxpayer underwrites.

In terms of real dollars we spend about what was spent in 1970 at less than 4% of current GDP.

GDP is not a good target for defense spending.

Need, solution to need and opportunity against other uses of the scarce resources are the way to decide on defense.

And if the money is just going down a hole we cut our loses.

[anne]

Yes; I also finished reading the paper, but am left with the same impression I began with. How lucky we are to have had Medicare all these years, and Medicaid, and let's have more because health insurance has given us a means of providing for our health and stimulated the health care industry to better and better cater to our health needs. Let's have more.

[anne]

There is so much of interest here, even to ideas that I have given little consideration such as a measure of self-insurance. But, the question that is really being raised is what we value. Ben Bernanke is telling us again of the fearful problem of "entitlements." Amy Finkelstein fear Medicare especially, though Medicaid must be included. Why though is there no fear of alternate spending? Why is there no fear of spending on defense, no matter how soaring? Until the question is answered properly, I have not the least concern about "entitlements" and want more.

[piglet]

PS, I restate that in my reading, it is strong government regulation, together with other factors such as favorable demographics, that make the difference in Singapore. Your emphasis on out-of-pocket payments in my view is misguided. Singapore starts with forcing people to put money aside (which Germany for example does also but in a different fashion) and then they use this money to pay for the doctor. That is not so different from having insurance pay: it is actually a variant of UHC. On top of that, there are substantive insurance components within the Singapore system. So to describe this system as "predominatly out-of-pocket payment" is misleading at best. Especially if you suggest implementing a similar system in the US, it is quite dishonest to describe it in a way that ignores the importance of government regulation. When you say to an American audience, "let's dismantle health insurance and let people pay for their health care needs out of pocket", that is quite different from saying "let the government tightly control the health care sector and let's introduce a compulsory health care saving scheme".

[Noni Mausa]

Peter: "If the goal is to look at systems more efficient than the U.S., why not look at systems more efficient than Canada? Singapore runs at less than 50% of the cost of the Canadian system."

Noni: "I too can have a cheap health care system if I just vote the dependent and sick people off the island...But that's a feedlot, not a society."

Sorry it wasn't obvious to you that my primary goal is NOT cheapness. (The Pine Box model comes out way ahead there.)

Noni

[piglet]

johnchx, I'll have to read the paper when I have time but for now the methods you describe (estimate the marginal impact of Medicare and incorporate confounding factors, oh my goodness - that's what makes quantitative social science so hard) don't appear to be that robust.

"How this works -- why I spend more on health care when my neighbor gets health insurance -- is not well understood." And actually it is doubtful whether that is the case. Again, I say: international comparison. I would also like to compare the insurance coverage among working age people in 1966 to today. Remember, it's not that there was no health insurance in the US prior to 1966. Most working people were presumably covered by insurance, as they are today. So why did those 12% of the population have such an impact? If the mechanism is not well understood then purely statistical evidence that leads to rather implausible conclusions should be treated with caution.

[me]

big Ben should be lecturing Congress about the defense budget, not the social security surplus.

Healthcare in Atlanta? Peter, when I get a bill for $600 for bloodwork and Quest labs settles for $26 from my insurance company I would say something is rotten.

As for those suggesting that expenditures would double with a single payer system ignore the rest of the civilized world that cover everyone at half the cost.

[anne]

Remember, the paper is essentially a polemic against health care as a right and thoroughly unconvincing. The comments here are the more interesting. Drugs are becoming an increasing factor by cost in health care. Partly, this is because of development of increasingly effective drugs but importantly for another reason entirely.

In 1980, drug companies were allowed to patent products that were developed at public expense and would previously have been sold in generic form. From 1980, international drug companies have become the most profitable of companies in any American stock market sector. Where every other developed country negotiates drug prices, we do not and in not doing so the cost of drugs is far higher for Americans.

Should we then remove drug coverage from any insurance or negotiate drug prices and add all sorts of drug coverage as is done elsewhere?

[dale]

"It also shows that a vast proportion of health care dollars yield no actual benefits. For different reasons, this is a subject that neither the left or right seem inclined to discuss."

If we include the works of Michael Marmot and Richard Wilkinson into this discussion of health care costs we will realize that the very act of providing universal coverage- of decreasing inequality- and strengthening social solidarity- will make for healthier people- in and of itself.

[donna]

Ridiculous. Everyone who IS insured is paying for everyone who is uninsured.

[JRossi]

Very interesting study, and not one whose conclusions I could have predicted ahead of time.

[Isabel]

Still on Singapore - from a comment on a post in Belle Waring's blog where she marvels how cheap healthcare is in Singapore but doesn't really know how it works:

http://examinedlife.typepad.com/johnbelle/2004/04/im_not_going_to.html

I really can't imagine this working in the States or in "the real world", as the poster says:

"Since I'm actually a citizen of the nanny state, let me fill in some details for what nah alluded to.

Each month, 20% of my paycheck is automatically docked, and together with an amount equal to 13% of my pay, added to an account in the Central Provident Fund. Think of it as a (near) nation wide pension fund for individuals. The CPF contribution is allocated to three accounts--the ordinary, special and medisave accounts (22, 5 and 6% respectively) and given a small interest rate (the last I checked, 2.5% pa for the ordinary account and 4% for the special and medisave accounts; all tax free).

I can use the ordinary account for the purchase of (public) housing, college education of my children and certain approved investments. The special account is the pension fund proper and cannot be touched until I reach the retirement age. The medisave account can only be used for medical expenses, both for myself and my dependents, and to purchase medical insurance.

The 13% employer's contribution figure is variable and can be used as a powerful macroeconomic tool. For instance, in times of recession, it is reduced so as to decrease the cost of labor. In better times, it is returned to a higher figure.

The accumulated CPF funds are invested, largely overseas, by a statuory board in charge of such matters. In other words, it is put to work generating foreign currency savings, etc. Another macroeconomic feature.

Since the money I have in my CPF account is in my own name, it is a bit misleading to think of it as taxation rather than enforced savings. Obviously, I would understand if "God-fearing" Americans are uncomfortable with it--but to someone who grew up in the system, it is pretty much invisible to me, just part of the furniture of the social world I live in, so to speak (much as social security tax might be to Americans?).

Caveat: nothing I say above amounts to a defense of the system, or imply any favorable comparison with an other real world system. To every social and economic benefit provided by the CPF scheme, there are also other social and economic costs. Nothing comes for free."

[James Killus]

"Ridiculous. Everyone who IS insured is paying for everyone who is uninsured."

No Donna, that is simply untrue. Someone who is uninsured in the U.S. must first exhaust all their financial assets before public assistance becomes available. Where I live (Contra Costa County, California), for example, MediCal is means tested, requiring an income of less than $2000 per month and zero (yes, zero) financial assests, including IRA and 401(k) accounts, before someone is eligible.

I'm aware that people like to believe that the uninsured receive medical care on an emergency basis in the U.S., but in reality, they are mostly just allowed to die.

On the topic of Singapore, a little checking also informs me that the "Medisave" accounts in Singapore are involuntary, that about 80% of hospital beds are in public facilities, and the degree of regulation of physicians would be called "socialized medicine" in the U.S. Add the fact that Singapore has a population of less than 5 million, and one suspects that the adoption of the "Singapore model" for the U.S. would actually result in some voluntary HSAs (since involuntary contributions are actually taxes) and little else.

[Isabel]

"Shared responsability" vs. single-payer:
http://www.newamerica.net/publications/articles/2006/universal_health_coverage_as_used_in_france_belgium_can_work_here_4597

[Arne (not anne)]

"If the top 1-5% of U.S. health care consumer went untreated, wouldn’t that show up in the data?" [PS]

People who spend more than the average at the end of life are clearly in the top 1% of health care consumers. Cutting that spending would have a far larger impact on spending distribution than on mortality distribution, so I suspect that PS is wrong.

[anne]

http://www.newamerica.net/publications/articles/2006/universal_health_coverage_as_used_in_france_belgium_can_work_here_4597

June 30, 2006

Universal Health Coverage as Used in France, Belgium Can Work Here
By Steven Hill

A new report from the Citizens' Health Care Working Group established by Congress has concluded that the federal government should guarantee basic health care to all Americans. Many proponents, as well as opponents, of health care reform equate universal coverage with a Canadian-style, government-run, single-payer system. But a survey of successful health care systems around the world shows this is an incorrect assumption.

For example, the World Health Organization rates France as having the No. 1 health care system in the world. France's system covers everyone. It also is noted for its short waiting periods, affordability, freedom of physician choice, doctors who still make house calls, exemplary gynecological care, quality health care for immigrants and the poor, all while spending about half what Americans pay to fund a health care system ranked 37th in the world.

Yet France does not have a single-payer system, nor does it have what is commonly referred to as government-run, socialized medicine. Neither does Germany, Japan, Austria, Belgium and the Netherlands, yet they also provide universal coverage and quality health care at a fraction of what we pay in the United States.

How do these non-single-payer nations provide such excellent care? These nations employ a system that blends a flexible mixture of public and private, with most doctors, nurses and other professionals working for private medical groups, not for the government as in the Canadian or British-style single payer systems. Many hospitals also are in private hands, while others are in public hands.

The funding for health care in these nations is best described as a "shared responsibility" -- employees, employers and the government all contribute a pre-determined amount. Both workers and their employers are subject to mandatory payroll deductions, and government chips in any shortfalls for poorer individuals, depending on income level or employment status.

The contributions from individuals, employers and government are deposited into private insurance funds that are non-profit and government-regulated (sometimes known as Sickness Insurance Funds, or SIFs). Additional private insurance can be purchased for premium services, such as a private room in the hospital.

But here's the key part: the SIFs sit down at the bargaining table with the government and representatives from professional associations of doctors and health care professionals to set exact fee structures. They negotiate strict cost controls that have prevented expenditures paid by consumers from approaching anywhere near exorbitant U.S. levels. Cost controls are essential to the success of these "shared responsibility" systems....

[anne]

Thank you, Isabel.

[JRossi]

I'm not surprised that medicare coming in did not decrease mortality rates in the first 10 years. A lot of the mortality was already in the pipeline (heart disease, cancer). It would be interesting to know if mortality rates declined later.
It is of course impossible to extrapolate morbidity and mortality and cost statistics from the elderly to the non-elderly who would be covered by universal insurance. A properly designed controlled trial would be needed to address the issue, and one is probably not forthcoming. There are reasons to think that costs would not rise as fast for younger patients given that they are relatively health c/w the elderly, but there are no definitive answers.
But reading between the lines on Finkelstein, I do think universal care would need to be accompanied by some sort of rationing mechanism. And she is right that universal care is to a great extent (not exclusively) a financial protection issue. It's also an equality issue and a humanitarian issue.

[Isabel]

Anne, this might be a bit of an overkill, but if you have the time and the inclination...

http://ec.europa.eu/health/ph_information/dissemination/hsis/hsis_13_nhs_en.htm

[Peter Schaeffer]

ilsm,

You asked for a comparison of 1947 defense spending versus current. I provided it. You compared the Cold War / Soviet threat period to the present. I came up with the numbers (average 1950 - 1970 9.5% of GDP). You mentioned 1946. Found that number as well.

You can argue that we spent too much back then and/or that we spend too much now. Since I am an anti-neocon, I am inclined to agree about the present. However, 3.7% is near post-WWII lows for defense spending and health care spending is at an all-time high.

[Peter Schaeffer]

Piglet,

You wrote

“I restate that in my reading, it is strong government regulation, together with other factors such as favorable demographics, that make the difference in Singapore.”

Demographics are the easy part here. Some time ago I wrote

“Pioneer10 gave statistics for the over 65 population in Singapore, the U.S., and Europe. I will use his data. Another source Seniors Spending Four Times More for Healthcare gives spending data by age in the U.S. Applying the U.S. cost ratio to a modified Singapore population does raise outlays modestly.

Assuming Singapore starts with health care at 3.7% of GDP and the over 65 group as 8.5% of the population, raising the elderly to 12.5% (as in the U.S.) increases outlays to 4.06% of GDP. Raising the elderly to 16% of the population (as Pioneer10 states for France) increases outlays to 4.4% of GDP.”

Alas, the 0.7% I mentioned above was for Singapore with France’s demographics. Singapore with U.S. demographics would have increased costs of 0.36% of GDP. Do I need to reproduce every number used in my calculations?

I have already acknowledged the role of government in creating Singapore’s system. Note that I wrote

“However, I would be the first to agree that the Singapore system works because of the interaction of several factors including required savings and regulation. That is why I always refer to it as a “system”, not just as an example of the virtues of relying upon “out-of-pocket” health payment.”

My use of the “predominantly out-of-pocket” phrase reflects the facts. 62% is well more than 50%.

I have always advocated the totality of Singapore’s health care system (and other aspects of Singapore as well). I would use somewhat different words… Nonetheless

"let the government tightly control the health care sector and let's introduce a compulsory health care saving scheme"

Is a sentiment that I would readily agree with.

Noni,

You haven’t supported your claims about “if I just vote the dependent and sick people off the island”

However, if you like expensive health care… Welcome to the United States.

Me,

“Healthcare in Atlanta? Peter, when I get a bill for $600 for bloodwork and Quest labs settles for $26 from my insurance company I would say something is rotten.”

Agreed and in Singapore the fee might be $10. Folks with actual experience in Singapore have posted actual costs and they were quite low.

By the way, the standard single-payer system run somewhere around 2/3rd of the US cost, not half. Singapore is 75% cheaper.

Dale,

Single payer systems may improve solidarity and health. Some supportive data would be helpful. However, they also ration care and yes that is a topic most folks are loathe to discuss.

Isabel,

Thank you for the details. Sounds like a success story to me.

Arne (not Anne)

"Cutting that spending would have a far larger impact on spending distribution than on mortality distribution, so I suspect that PS is wrong."

No, I am actually inclined to agree. Note that I mentioned futile care above. My specific example was how a vast sum was spent keeping a relative alive who opposed artificial life support.

Does anyone think this makes sense?

[johnchx]

anne writes: I'll have to read the paper when I have time but for now the methods you describe (estimate the marginal impact of Medicare and incorporate confounding factors, oh my goodness - that's what makes quantitative social science so hard) don't appear to be that robust.

You might want to read section II.C of the paper. It's titled, Robustness. Jeez.

anne also writes: Remember, the paper is essentially a polemic against health care as a right...

Um...you really might want to read the paper before you call it a polemic. It's a moderately complex exercise in econometrics, working with a long-neglected data set. It draws no policy conclusions at all, and has nothing to do with the desirability of a national health care system.

Really...the tendency to instantly classify every piece of research as either "ammo" or "enemy fire" does become tiresome.

[Isabel]

PS, of course it makes sense. But what if your relative had said "do what you can"? When you can a lot, where do you stop?
Let me give you an example, of a Swedish friend of mine that had premature twins. After some trouble they (or one of them) had to be intubated and something went wrong. The child went on to have recurrent respiratory problems, so about age 3 or 4 it was decided to do a tracheotomy. At that age, the doctors explained, is like having a small child permanently with a pebble in its mouth, it can choke anytime without warning. So, this being Sweden, a person was allocated to watch over him night and day (except sundays) until he grows up and can look after himself. The child is doing great. According to the doctors, there are only two countries with decent survival outcomes in these circumstances, Sweden and Netherlands. Incidently, my friend doesn't complain about taxes any more.
Waste of money?

[ilsm]

Peter,

You've got to be Hoovering me!

"However, 3.7% is near post-WWII lows for defense spending and health care spending is at an all-time high."

Today's GDP is much larger than GDP in 1970, when we were losing 6 or 8 first line jets a week in Vietnam, and 20 or so helicopters, saying nothing about supporting 500,000 service people in the RVN.

This year's military spending in constant dollars is about the same as 1970.

With 40% of the miliatary manpower and similar reduced, but hugely expensive, weapons, we are running police stations in Baghdad and getting our clocks cleaned.

And what sense does it make to compare the percent of GDP spent on 'national security' welfare to the money spent on health care?

Quit Hoovering me.

[anne]

The paper by Amy Finkelstein, which I have read, is precisely a polemic against Medicare and extension of the Medicare model for further health care insurance coverage. I could not be less impressed by the selective playing with data analysis. Phooey.

[anne]

This comment is not from me, however.

"I'll have to read the paper when I have time but for now the methods you describe (estimate the marginal impact of Medicare and incorporate confounding factors, oh my goodness - that's what makes quantitative social science so hard) don't appear to be that robust."

[ilsm]

johnchx:

"Um...you really might want to read the paper before you call it a polemic. It's a moderately complex exercise in econometrics, working with a long-neglected data set. It draws no policy conclusions at all, and has nothing to do with the desirability of a national health care system."

Well!!! she said:

"just look at the effects of the introduction of Medicare in 1966. Medicare provides..."

You've got to be Hoovering me!

just look indeed!!

"a moderately complex exercise in econometrics, working with a long-neglected data set."

Indeed!!

[anne]

The paper and the Wall Street Journal commentary are written to attack Medicare and any extension, and the attacks are transparent enough to suggest that we simply set the older ill afloat on ice drifts should there be any left rather than bother to care for them for the fierce cost involved. Not gonna happen.

[Movie Guy]

donna - "Ridiculous. Everyone who IS insured is paying for everyone who is uninsured."

James Killus - "No Donna, that is simply untrue. Someone who is uninsured in the U.S. must first exhaust all their financial assets before public assistance becomes available. Where I live (Contra Costa County, California), for example, MediCal is means tested, requiring an income of less than $2000 per month and zero (yes, zero) financial assests, including IRA and 401(k) accounts, before someone is eligible. I'm aware that people like to believe that the uninsured receive medical care on an emergency basis in the U.S., but in reality, they are mostly just allowed to die."

donna,

Your point is well taken in the final analysis as viewed from taxpayer and insured health care perspectives. Further, the U.S. Government is repeatedly asked for large sums of addtional federal funding to bail out the States emergency rooms financial shortfalls and California sits at the top of that list. Now, James Killus likes to scoff at posters who reference materials that support their posts (as he did last week on another Economist's View thread), but let's take a look at some hard facts from Contra Costa County and other counties in the State of California.

Here's are a few examples of what James Killus didn't bother to explain about the financial conditions of the Contra Costa County and other State of California hospital emergency rooms resulting from the treatment of individuals:

CONTRA COSTA COUNTY
Emergency Rooms (ER) Annual Losses, 1999-2000

Total Annual Emergency Rooms Medi-Cal Visits: 26,116
Total Annual Rooms Uninsured Visits: 17,604

CONTRA COSTA REGIONAL MEDICAL CENTER: -$5,009,642
DOCTORS MEDICAL CENTER, PINOLE CAMPUS: -$483,347
DOCTORS MEDICAL CENTER, SAN PABLO CAMPUS: -$1,415,899
JOHN MUIR MEDICAL CENTER: -$273,554
KAISER FOUNDATION HOSPITAL, MARTINEZ/WALNUT CREEK:-$1,517,436
MT. DIABLO MEDICAL CENTER: -$3,451,366
SAN RAMON REGIONAL MEDICAL CENTER: -$285,320
SUTTER DELTA MEDICAL CENTER: -$1,493,721

Contra Costa County Total ER Losses: -$13,930,284

-----

State of California
Emergency Rooms (ER) Annual Losses, 1999-2000

ALAMEDA COUNTY: -$24,469,659
AMADOR COUNTY: -$678,929
BUTTE COUNTY: -$4,528,927
COLUSA COUNTY: -$520,595
CONTRA COSTA COUNTY: -$13,930,284
DEL NORTE COUNTY: -$546,652
FRESNO COUNTY: -$5,283,328
GLENN COUNTY: -$348,896
HUMBOLDT COUNTY: -$875,292
IMPERIAL COUNTY: -$707,104
INYO COUNTY: -$1,123,116
KERN COUNTY: -$5,726,917
KINGS COUNTY: -$2,401,523
LAKE COUNTY: -$1,748,641
LASSEN COUNTY: -$260,171
LOS ANGELES COUNTY: -$98,076,425
MADERA COUNTY: -$4,061,821
MARIN COUNTY: -$1,666,265
MARIPOSA COUNTY: -$456,120
MENDOCINO COUNTY: -$1,181,569
MERCED COUNTY: -$1,557,141
MODOC COUNTY: -$123,153
MONO COUNTY: -$494,655
MONTEREY COUNTY: -$2,203,841
NAPA COUNTY: -$1,938,858
NEVADA COUNTY: -$2,508,576
ORANGE COUNTY: -$16,659,082
PLACER COUNTY: -$701,280
PLUMAS COUNTY: -$537,727
RIVERSIDE COUNTY: -$12,533,966
SACRAMENTO COUNTY: -$6,458,212
SAN BENITO COUNTY: -$98,634
SAN BERNARDINO COUNTY: -$20,232,226
SAN DIEGO COUNTY: -$21,619,259
SAN FRANCISCO COUNTY: -$7,249,355
SAN JOAQUIN COUNTY: -$6,549,915
SAN LUIS OBISPO COUNTY: -$3,741,733
SAN MATEO COUNTY: -$2,702,358
SANTA BARBARA COUNTY: -$2,511,633
SANTA CLARA COUNTY: -$8,699,359
SANTA CRUZ COUNTY: -$2,969,394
SHASTA COUNTY: -$6,451,044
SIERRA COUNTY: -$27,841
SISKIYOU COUNTY: -$1,011,370
SOLANO COUNTY: -$3,502,956
SONOMA COUNTY: -$8,239,599
STANISLAUS COUNTY: -$7,752,453
TEHAMA COUNTY: -$305,004
TRINITY COUNTY: -$417,405
TULARE COUNTY: -$2,207,336
TUOLUMNE COUNTY: -$77,667
VENTURA COUNTY: -$2,087,656
YUBA COUNTY: -$1,936,195

State of California ER Losses: -$324,699,115

State of California Emergency Rooms available:

1990: 407 Emergency Rooms
2001: 347 Emergency Rooms

"Emergency rooms reporting losses (1999 to 2000) increased from 285 to 289, not including those that shut completely. Since 1999 ten more emergency rooms have closed bringing the total number of closures to 19 in the last three years and 60 in the last decade."

"The rapid decline in the number of hospitals has increased the strain for those that remain, led to longer waits, and left entire communities without a nearby emergency room. Many hospitals are faced with escalating losses as more patients seek care in their emergency rooms. The losses create a financial strain on the hospital often forcing the closure of the entire facility."

Emergency Room Closures Since 1999: (at the time of this report)

Pacifica Hospital – Orange County
Santa Ana Hospital Medical Center – Orange County
Valley Community Hospital – Santa Barbara County
Lindsay Hospital – Tulare County
Mercy American River Hospital – Sacramento County
Sharp Cabrillo Hospital – San Diego County
Mission Bay Hospital – San Diego County
Scripps East County Hospital – San Diego County
Mt. Zion Hospital – San Francisco
Doctors Hospital - Pinole

Source: OSHPD Hospital Financial Data Disclosure Report (99-00)

The 4th Annual ER Losses Report "found that uncompensated emergency care costs have continued to escalate dramatically, reaching $635 million in fiscal year 2002 (the most recent year for which data is available), up from $560 million in 2001—an 18 percent increase."..."more than 60 hospital emergency rooms and trauma centers have closed in the last decade, 28 of them in the last four years."

“This is an astonishing level of red ink and it shows no sign of abating,” said CMA President Robert Hertzka, M.D. “We all assume emergency care will be there when we need it, but these losses point to a crisis of extreme proportions and underlines the threat to emergency care around the state. We are all just a heart attack or a car accident away from needing that care.”
Source: CMA’s 4th Annual ER Losses Report: Uncompensated Emergency Care Surpassed $600 Million in FY 2002
September 23, 2004, and here

-----

The State of California Emergency Rooms losses plus related ER uncompensated physicans care:
(numbers rounded down)

1997-98: -$391 million (-$291 mil + -$100 mil)
1998-99: -$428 million (-$316 mil + -$112 mil)
1999-00: -$442 million (-$324 mil + -$118 mil)
2000-01: -$540 million (-$389 mil + -$150 mil)
2001-02: -$560 million
2002-03: -$635 million
2003-04: -$???
2004-05: -$???
2005-06: -$???
2006-07: -$???

[dale]

Peter,
Rationing exists now. It's just that middle class folks with decent insurance don't see much of it. Unless they get turned down for a procedure by HMO, etc.

Take a good look at Marmot and Wilkinson's work. They are both English social epidemiologists with extensive publications. Inequality kills and injures. Over and above the actual effects of not being able to afford decent health care.

[evagrius]

Movie Guy;

People who end up in E.R. are asked to sign an agreement for payment. It's usually less than the actual cost but is still substantial. Those who can't pay are referred to a collection agency that often duns their paycheck if they're employed or find other means to extract payment.
Quite often, the situation is such that there really is no money, period.
The patients are treated only for emergency situations. Long term treatment is often restricted and unavailable.
Single individuals are not eligible for Medical unless they've been declared disabled by Social Security, ( an application process that can take a year or longer).
Some counties in California, such as S.F. and San Mateo, have a program for those ineligible for Medical but it's still somewhat restricted.

[evagrius]

Kames Killus;

"Where I live (Contra Costa County, California), for example, MediCal is means tested, requiring an income of less than $2000 per month and zero (yes, zero) financial assests, including IRA and 401(k) accounts, before someone is eligible."

Actually it's as follows;

To be eligible for MediCal one must be over 65 or under 21, or an individual between 21-65 who is a parent of a child deprived of the other parent, ( single parent or remarried parent), pregnant, ( the foetus is considered eligible- under 21), or declared disabled by Social Security.
The income levels vary, from $600 for a single individual to $1100 for a family of four, ( with different income level eligibility for children depending on age based on the FPL, ( Federal Poverty Level)).
Asset levels also vary. Single individuals may not have assets over $2000. Families may not have assets over $3000 or so going up by increments of about $150 per child.
Children and pregnancy may be asset waivered.
Assets are more important for eligibility than income. Any income over the levels is considered to be available for medical expenses. It's called "share-of-cost".
Asset waiver programs may also be income waivered depending on income for children and those over 65.

This is an extremely abbreviated version of the eligibility rules.

I hope that it wasn't to complicated.

MediCal has over 150 different eligibility programs for applicants. It all depends on income and assets, age, etc;

[Movie Guy]

evagruis,

I know all about ERs as I've done hospital consulting.

I also know that unisured individuals are treated by ERs. And that the volume of ER patient load is increasing, not decreasing.

The large financial losses of hospital emergency rooms in California speak volumes about what is really happening. Only fools and apologizers would try to dismiss annual state level emergency rooms losses on the order of $630 million for FY2002.

[johnchx]

anne wrote: The paper and the Wall Street Journal commentary are written to attack Medicare and any extension....

Now you're making a claim about motive. Is this simply the baseless allegation it appears to be, or do you have some evidence to back it up?

I'm really puzzled because I don't know by what bizarre logic the findings of this paper could possibly constitute an "attack" on Medicare. And Finklestein & McKnight's 2005 paper takes the position that traditional analyses have undervalued the benefits of Medicare.

Maybe I'm missing something, but it seems to me that you're wildly off base here.

[anne]

http://www.washingtonpost.com/wp-dyn/content/article/2007/02/27/AR2007022702116_pf.html

February 28, 2007

For Want of a Dentist: Prince George's Boy Dies After Bacteria From Tooth Spread to Brain
By Mary Otto - Washington Post

Twelve-year-old Deamonte Driver died of a toothache Sunday.

A routine, $80 tooth extraction might have saved him.

If his mother had been insured.

If his family had not lost its Medicaid.

If Medicaid dentists weren't so hard to find.

If his mother hadn't been focused on getting a dentist for his brother, who had six rotted teeth.

By the time Deamonte's own aching tooth got any attention, the bacteria from the abscess had spread to his brain, doctors said. After two operations and more than six weeks of hospital care, the Prince George's County boy died.

Deamonte's death and the ultimate cost of his care, which could total more than $250,000, underscore an often-overlooked concern in the debate over universal health coverage: dental care.

Some poor children have no dental coverage at all. Others travel three hours to find a dentist willing to take Medicaid patients and accept the incumbent paperwork. And some, including Deamonte's brother, get in for a tooth cleaning but have trouble securing an oral surgeon to fix deeper problems.

In spite of efforts to change the system, fewer than one in three children in Maryland's Medicaid program received any dental service at all in 2005, the latest year for which figures are available from the federal Centers for Medicare and Medicaid Services.

The figures were worse elsewhere in the region....

[anne]

Well, as far as I can tell having health care insurance, the finest of insurance, is an unalloyed good individually and socially, and Medicare has been an instance of a particularly helpful form of health care insurance that could well be used as model for extending public-private coverage. To the extent that having health care insurace increases demand for services, than I have not the slightest doubt the increased demand has been and will be generally warranted.

[anne]

John Chx:

"The point is this: while there is actual (experimental!) evidence that indicates that individuals with health insurance consume more health care, the size of the increase is small in comparison with the size of the increases in health care spending we've observed over the last few decades (i.e. estimated in the range of 10% - 12% of the toal increase). However, Finklestein's work suggests that, in the aggregate, an increase in health insurance coverage may have an effect on total health care consumption much larger than the direct effect on the consumption of the individuals who get the new health insurance.

"How this works -- why I spend more on health care when my neighbor gets health insurance -- is not well understood. And I, for one, would certainly like to understand it better."

Please further explain this interesting comment when there is time.

[piglet]

"The large financial losses of hospital emergency rooms in California speak volumes about what is really happening. Only fools and apologizers would try to dismiss annual state level emergency rooms losses on the order of $630 million for FY2002."

Movie, what you are saying is not inconsistent with what James and evagrius say. The losses you mention that tax payers have to compensate are one aspect of this country's screwed-up health care system. Another aspect is that poor or even middle-class people, when they are uninsured, may get stripped of everything they have in case of sickness, plus may be denied necessary treatment, plus they may be induced to wait too long before visiting the doctor, thus aggravating the condition and increasing the costs. All those pathologies could be fixed by introducing some kind of universal coverage. That's what people are arguing here.

Just recently I read a commentary by a propertarian extremist saying that universal insurance was a "ridiculous" idea because it forces people to buy something they don't want. The system is broken, so don't fix it, that's the message.

[piglet]

johnchx: One of the quotes you ascribed to anne is from me. I still haven't read the paper but I agree that the WSJ commentary is "written to attack Medicare and any extension". Finkelstein at least acknowledges that Medicare has the benefit of providing "financial security". But look at her second paragraph:

"For evidence of how such programs can lead to increased spending, just look at the effects of the introduction of Medicare in 1966." She says that Massachusetts' new universal coverage program is comparable in scope to Medicare, and she claims that Medicare was "responsible for about half of the sixfold growth in real per capita health-care spending during this period". In other words, providing health coverage to 12% of the population has, in this story, increased per capita spending by a factor three.

I repeat that I find that claim implausible and highly suspicious, especially since she can't explain the causal mechanism; it is inconsistent both with the Rand Health Insurance Experiment and with what we know from international comparison. Furthermore I doubt that the kind of statistical analysis she performs can reliably identify the effect of Medicare given that health care spending was increasing anyway.

Taken in sum, her statements can only mean one thing: don't try universal coverage. We can't afford our per capita spending to triple again, do we? She doesn't have to spell this out explicitly, but yes, this commentary is an attack on universal coverage.

[Movie Guy]

piglet - "Movie, what you are saying is not inconsistent with what James and evagrius say. The losses you mention that tax payers have to compensate are one aspect of this country's screwed-up health care system."

(I'm going to try a different response format to save space)

piglet,

Let's return to what donna said and review again the reply that James Killus provided:

donna - "Ridiculous. Everyone who IS insured is paying for everyone who is uninsured."

James Killus - "No Donna, that is simply untrue. Someone who is uninsured in the U.S. must first exhaust all their financial assets before public assistance becomes available. Where I live (Contra Costa County, California), for example, MediCal is means tested, requiring an income of less than $2000 per month and zero (yes, zero) financial assests, including IRA and 401(k) accounts, before someone is eligible. I'm aware that people like to believe that the uninsured receive medical care on an emergency basis in the U.S., but in reality, they are mostly just allowed to die."

James Killus gives the impression that the uninsured are denied emergency room treatment and are "just allowed to die" which is not what happens at most emergency rooms in the State of California, or throughout the United States as evidenced by the large financial operating losses.

Contra Costa County had emergency rooms losses of $13,930,284 in 1999-2000, excluding additional losses of related physician treatments and fees. Such emergency room treatment in Contra Costa County involved, among care for others, support of 26,116 Medi-Cal patient visits and 17,604 uninsured visits. There is no evidence that I am aware of that the emergency rooms in Contra Costa County turned away the uninsured individuals and that they were told to just go die. The point is that uninsured individuals appear to have received emergency room treatment as evidenced in the emergency room financial losses to the tune of almost $14 million in 1999-00.

The State of California sustained emergency room losses of $324,699,115, excluding further losses of related physician treatments and fees. Rolled together, such emergency room expense losses in the State of California involved losses of $391 million (1997-98), $428 million (1998-99), $442 million (1999-00), $540 million (2000-01), $560 million (2001-02), and $635 million (2002-03) let alone whatever the running losses have been since then.

Emergency rooms have increased their fee charges substantially in recent years. Some of the fees are rather staggering depending on the hospital and geographic location. It appears that the hospitals are attempting to recoup some of their operating losses due to treatment of uninsured and illegal alien medical treatment. The increased costs of receiving emergency room treatment appear to be passed along in the form of higher insurance health care premiums based on a formula depending on the provider and percentage of claims resulting from emergency room treatment. As such, donna's point appears to be well founded. Similarly, the additional funds sought from State governments and the Federal government to keep emergency room open impacts taxpayers. Donna and others, once again, help defray such funding allocations.

I am not questioning what egrius stated. I am, though, questioning what James Killus stated with respect to the ethical operation of and treatment standards of emergency rooms, notably in Contra Costa County as well as other counties in the State of California. His other point, similar to that of egrius, is well founded.

Of course, the health care system is broken in many respects. The U.S. government has attempted to take some pressure off of the increasing volume of traffic flowing to emergency rooms with the expansion of minimum cost or free health care clinics across the nation. The ultimate responsibility for the operation of the clinics rests with the counties, not the federal government. Yet, the problem of illegal aliens (who also use emergency rooms) rests with the federal government, though some States, counties, and municipalities are reluctant to assist in identifying illegal aliens. The matter of private health care insurance coverage, as we know, is built around employer-provided health care plans and those plans are declining in number and scope of coverage. Yet, the attempts to improve the possibility of reducing or pooling costs of individual health care insurance provision (say, an individual plan for a self-employed or retired individual/family) is apparently being rebuffed by some.

We are a long way from sorting out the future provision of health care in the United States. But closing our emergencies rooms isn't the answer, nor is the increasing volume of emergency room traffic driven, in part, by larger numbers of uninsured individuals. Just look at the losses of emergency rooms in Los Angeles County, California in 1999-00: -$98,076,425. That's seven years ago. $98 million in losses. Staggering. And this doesn't begin to address emergency room cost overruns in other parts of the nation.